Made available in DSpace on 2016-08-10T10:44:41Z (GMT). No. of bitstreams: 1 SERGIO MAURO MEDEIROS DE AZEVEDO.pdf: 470800 bytes, checksum: f30d7d2534b39350a2f1c74490c5de5e (MD5) Previous issue date: 2007-01-31 / The present study aimed to evaluate the behavior of arsenic (As) in typical Cerrado soil (Oxisol), such as adsorption and transport in different horizons. In experiments samples were used in two different depths, from 0 to 0.20 m, 0.80 m to 1.00 m. Samples were prepared, characterized and tested with the application of an arsenic solution. For the experiment were used percolation columns made of PVC tubes with samples to simulate a natural soil profile section made in the laboratory and application of a known concentration of arsenic solution. The arsenic content in each sample was quantitated by spectrophotometric analysis of plasma atomic emission induced argon (ICP-AES). Graphs of arsenic adsorbed depending on the equilibrium concentrations were prepared and the results showed that the oxisol analyzed in its natural pH conditions and so was used in this study has a high adsorption capacity for arsenic to possess desirable characteristics such as high levels of aluminum oxides, iron oxides, ghoetita and hematite fine particle size and large specific surface area. In environmental terms, this is a positive feature, therefore, to keep the layers of earth, is unlikely to be transported to the deep part to become a source of contamination of groundwater by arsenic. / O presente estudo teve como objetivo avaliar o comportamento do arsênio (As) em solo típico de Cerrado (Latossolo) quanto à adsorção e transporte em diferentes horizontes. Nos experimentos foram utilizadas amostras coletadas em duas diferentes profundidades, de 0 a 0,20 m e de 0,80m a 1,00 m. As amostras foram preparadas, caracterizadas e submetidas à aplicação de solução de arsênio. Para o experimento foram preparadas colunas de percolação confeccionadas com tubos de PVC com amostras a fim de simular um trecho do perfil do solo natural em laboratório e realizada a aplicação de uma solução de arsênio com concentração conhecida. O teor de arsênio em cada amostra foi quantificado através de análise por espectrofotometria de emissão atômica por plasma induzido de argônio (ICP-AES). Foram construídos gráficos das concentrações adsorvidas em função das concentrações em equilíbrio e os resultados demonstraram que o solo analisado, em suas condições naturais de pH e da forma que foi utilizado neste trabalho apresenta alta capacidade de adsorção de arsênio por apresentar características favoráveis como altos teores de óxidos de alumínio, óxidos de ferro, ghoetita e hematita, granulometria fina e grande área superficial específica. Em termos ambientais, esta é uma característica positiva já que, ficando retido nas camadas do solo, o arsênio dificilmente será transportado até o lençol profundo e não se tornará fonte de contaminação de águas subterrâneas.

Estudamos o modelo epidemiológico denominado susceptível-exposto-infectado (SEI) na rede quadrada por meio de simulações numéricas. Nesse modelo, cada sítio da rede pode estar susceptível, exposto ou infectado. Um sítio susceptível nas vizinhanças de um infectado se torna infectado com uma certa probabilidade e exposto com probabilidade complementar. Sítios infectados ou expostos permanecem para sempre nessa condição. Mostramos que os aglomerados gerados a partir de um único infectado numa rede repleta de suscetíveis são os mesmos aglomerados presentes na percolação isotrópica. Calculamos os expoentes críticos dinâmicos com bastante precisão permitindo colocar o modelo SEI na classe de universalidade da percolação dinâmica. / We have studied the epidemiologic model called susceptible-exposed-infected (SEI) on a square lattice by numerical simulations. In this model, each site of the lattice may be susceptible, exposed or infected. A susceptible site in the neighborhood of an infected site becomes infected with a given probability, or exposed with a complementary probability. Infected and exposed sites remain forever in these states. We have shown that clusters generated by a single infected site in a lattice full of susceptible are the same clusters as in the isotropic percolation. The critical dynamic exponents were calculated with large precision allowing to put the SEI model into the dynamical percolation universality class.

Neste trabalho foi avaliado o desempenho de liners compostos frente a soluções de \'CU\'\'CL IND.2\'.2\'H IND.2\'\'O\'. Foram utilizados um solo argiloso compactado (CCL) da Formação Corumbataí (interior do estado de São Paulo) e um geocomposto bentonítico (GCL) de fabricação nacional. Quatro corpos de prova foram preparados para a realização do ensaio de percolação em coluna, sendo dois montados com CCL sobre GCL e outros dois com GCL sobre CCL. Os corpos de prova com configurações distintas foram submetidos à percolação de duas soluções de \'CU POT.2+\': uma com concentração de 0,024M e outra com 0,012M. Após o início da percolação da solução contaminante houve aumento da condutividade hidráulica, porém os valores mantiveram-se na ordem de \'10 POT.-11\' m/s, compatíveis com as exigências requeridas em diversas legislações ambientais. O liner composto mostrou-se mais eficiente como barreira de controle de fluxo comparado com Musso (2008), que estudou ambos materiais separadamente. A configuração dos liners não influenciou os valores de Rd obtidos, ao contrário da concentração das soluções. Observou-se que os resultados de Rd são inversamente proporcionais às concentrações utilizadas. Neste trabalho, os valores de Rd obtidos foram maiores que em Musso (2008), evidenciando a vantagem de se combinar diferentes materiais para a contenção de íons. / This research evaluated composite liners performance when permeated by \'CU\'\'CL IND.2\'.2\'H IND.2\'\'O\' solution. It were used a compacted clayey soil (CCL) samples from Formação Corumbataí (state of São Paulo) and a geosynthetic clay liner (GCL) manufactured in Brazil. Four specimens were prepared to column percolation tests. A pair of them was made with CCL on top of GCL and the other two with GCL on top of CCL. Specimens with different configurations were subjected to two \'CU POT.2+\' solutions: one with a concentration of 0,024M and other with 0,012M. Just after solution percolation began, there was an increase of hydraulic conductivity, but remaining around \'10 POT.-11\' m/s, in accordance with environmental laws. The composite liners demonstrated to be more efficient as barrier when compared to Musso (2008), which has analyzed both materials separately. The configuration of the liners did not influenced Rd values, unlike solutions concentration. It was observed that Rd results are inversely proportional to used concentrations. Rd values were larger than those obtained in Musso (2008), showing the advantage of matching different materials aiming ions retention.

by Hon-chor Lee. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1995. / Includes bibliographical references (leaf 56). / by Hon-chor Lee. / Introduction --- p.1 / Series expansion for the conductivity of a linear random resistor network / Chapter 1. --- Introduction --- p.4 / Chapter 2. --- Comparison of the EMA with symbolic simulations in 2D --- p.5 / Chapter 3. --- Comparison of the EMA with Bergman and Kantor's findings --- p.6 / Chapter 4. --- Conclusion --- p.7 / Current moments of linear random resistor network / Chapter 1. --- Introduction --- p.10 / Chapter 2. --- Review of the definition of current moment --- p.10 / Chapter 3. --- Tremblay et. al.'s findings and symbolic simulation of current moments --- p.11 / Chapter 4. --- Conclusion --- p.13 / Effective medium theory for strongly nonlinear composites: comparison with numerical simulations / Chapter 1. --- Introduction --- p.15 / Chapter 2. --- Variational principles --- p.16 / Chapter 3. --- Formalism of EMA --- p.17 / Chapter 4. --- Comparison with numerical simulations --- p.19 / Chapter 5. --- Discussion --- p.21 / Percolative conduction in two-component strongly nonlinear composites / Chapter 1. --- Introduction --- p.25 / Chapter 2. --- Spherical inclusions --- p.25 / Chapter 3. --- Effective medium approximation in the vicinity of the percolation threshold --- p.27 / Chapter 4. --- Acknowledgment --- p.29 / Percolation Effects in Two Component Strongly Nonlinear Composites: Universal Scaling Behaviours / Chapter 1. --- Introduction --- p.31 / Chapter 2. --- General Scaling Relations for Two-Component Composites --- p.33 / Chapter 3. --- Estimate of Critical Exponents --- p.35 / Chapter 4. --- Numerical Simulations --- p.38 / Chapter 5. --- Discussions and Conclusions --- p.40 / Chapter 6. --- Acknowledgment --- p.40 / Improved effective medium theory for strongly nonlinear composites / Chapter 1. --- Introduction --- p.46 / Chapter 2. --- Formalism of Improved EMA --- p.47 / Chapter 3. --- Comparisons with numerical simulations and HS bound --- p.50 / Chapter 4. --- Discussions --- p.51 / Conclusion --- p.54

Wong Yuk Chun. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1999. / Includes bibliographical references (leaves 71-74). / Abstracts in English and Chinese. / Abstract --- p.ii / Acknowledgments --- p.iii / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Motivation --- p.2 / Chapter 1.2 --- The Scope of the Project --- p.2 / Chapter 1.3 --- An Outline of the Thesis --- p.3 / Chapter 2 --- Related Work --- p.5 / Chapter 2.1 --- Percolation Effect --- p.5 / Chapter 2.2 --- Percolation Models --- p.6 / Chapter 2.2.1 --- Site Percolation --- p.6 / Chapter 2.2.2 --- Bond Percolation --- p.8 / Chapter 2.3 --- Simulated Annealing --- p.8 / Chapter 3 --- Electrical Property --- p.11 / Chapter 3.1 --- Electrical Conductivity --- p.11 / Chapter 3.2 --- Physical Model --- p.13 / Chapter 3.3 --- Algorithm --- p.16 / Chapter 3.3.1 --- Simulated Annealing --- p.18 / Chapter 3.3.2 --- Neighborhood Relation and Objective Function --- p.19 / Chapter 3.3.3 --- Configuration Space --- p.21 / Chapter 3.3.4 --- Annealing Schedule --- p.22 / Chapter 3.3.5 --- Expected Time Bound --- p.23 / Chapter 3.4 --- Results --- p.26 / Chapter 3.5 --- Discussion --- p.27 / Chapter 4 --- Thermal Properties --- p.30 / Chapter 4.1 --- Thermal Expansivity --- p.31 / Chapter 4.2 --- Physical Model --- p.32 / Chapter 4.2.1 --- The Physical Properties --- p.32 / Chapter 4.2.2 --- Objective Function and Neighborhood Relation --- p.37 / Chapter 4.3 --- Algorithm --- p.38 / Chapter 4.3.1 --- Parallel Simulated Annealing --- p.39 / Chapter 4.3.2 --- The Physical Annealing Schedule --- p.42 / Chapter 4.4 --- Results --- p.43 / Chapter 4.5 --- Discussion --- p.47 / Chapter 5 --- Scaling Properties --- p.48 / Chapter 5.1 --- Problem Define --- p.49 / Chapter 5.2 --- Physical Model --- p.50 / Chapter 5.2.1 --- The Physical Properties --- p.50 / Chapter 5.2.2 --- Bond Stretching Force --- p.50 / Chapter 5.2.3 --- Objective Function and Configuration Space --- p.51 / Chapter 5.3 --- Algorithm --- p.52 / Chapter 5.3.1 --- Simulated Annealing --- p.52 / Chapter 5.3.2 --- The Conjectural Method --- p.54 / Chapter 5.3.3 --- The Physical Annealing Schedule --- p.56 / Chapter 5.4 --- Results --- p.57 / Chapter 5.4.1 --- Case I --- p.59 / Chapter 5.4.2 --- Case II --- p.60 / Chapter 5.4.3 --- Case III --- p.60 / Chapter 5.5 --- Discussion --- p.61 / Chapter 6 --- Conclusion --- p.62 / Chapter A --- An Example on Studying Electrical Resistivity --- p.64 / Chapter B --- Theory of Elasticity --- p.67 / Chapter C --- Random Number Generator --- p.69 / Bibliography

In this dissertation we solve several combinatorial problems in different areas of mathematics: automata theory, combinatorics of partially ordered sets and extremal combinatorics. Firstly, we focus on some new automata that do not seem to have occurred much in the literature, that of solvability of mazes. For our model, a maze is a countable strongly connected digraph together with a proper colouring of its edges (without two edges leaving a vertex getting the same colour) and two special vertices: the origin and the destination. A pointer or robot starts in the origin of a maze and moves naturally between its vertices, according to a sequence of specific instructions from the set of all colours; if the robot is at a vertex for which there is no out-edge of the colour indicated by the instruction, it remains at that vertex and proceeds to execute the next instruction in the sequence. We call such a finite or infinite sequence of instructions an algorithm. In particular, one of the most interesting and very natural sets of mazes occurs when our maze is the square lattice Z2 as a graph with some of its edges removed. Obviously, we need to require that the origin and the destination vertices are in the same connected component and it is very natural to take the four instructions to be the cardinal directions. In this set-up, we make progress towards a beautiful problem posed by Leader and Spink in 2011 which asks whether there is an algorithm which solves the set of all such mazes. Next, we address a problem regarding symmetric chain decompositions of posets. We ask if there exists a symmetric chain decomposition of a 2 × 2 × ... × 2 × n cuboid (k 2’s) such that no chain has a subchain of the form (a1,...,ak,0) ≺ ... ≺ (a1,...,ak,n−1)? We show this is true precisely when k≥5 and n≥3. Thisquestion arises naturally when considering products of symmetric chain decompositions which induce orthogonal chain decompositions — the existence of the decompositions provided in this chapter unexpectedly resolves the most difficult case of previous work by Spink on almost orthogonal symmetric chain decompositions (2017) which makes progress on a conjecture of Shearer and Kleitman. Moreover, we generalize our methods to other finite graded posets. Finally, we address two different problems in extremal combinatorics related to mathematical physics. Firstly, we study metastable states in the Ising model. We propose a general model for 1-flip spin systems, and initiate the study of extremal properties of their stable states. By translating local stability conditions into Sperner- type conditions, we provide non-trivial upper bounds which are often tight for large classes of such systems. The last topic we consider is a deterministic bootstrap percolation type problem. More specifically, we prove several extremal results about fast 2-neighbour percolation on the two dimensional grid.

Este trabalho relata resultados de estudos sobre drenagem de camadas próximas à superfície que visam contribuir para aperfeiçoar o entendimento de comportamento da fase não saturada de solos, e disseminar o uso de elementos finitos para estudo de movimento de água em componentes que constituem a plataforma de vias. Descreveram-se e comentaram-se tópicos sobre as relações entre o formato de vias e a drenagem na estrutura de vias, processos para estimativas de vazões percoladas e infiltradas em solos por métodos simplificados, tópicos sobre a evolução do estudo de movimento de água e estimativa de medidas sobre propriedades de solos não saturados. Para estudo de caso se adotou a simulação com elementos de volume de sólido e método dos elementos finitos para com dados publicados sobre propriedades de solo arenoso ou argiloso obter dados para testes estatísticos que ressaltassem as diferenças entre as vazões estimadas com uso da condutividade hidráulica constante e de solo saturado, e a condutividade que incluísse o solo não saturado. Os dados usados foram obtidos por simulação, que estimou vazões percoladas e infiltradas, que foram submetidos a teste de hipóteses sobre diferenças de médias. O objetivo proposto foi atingido discutindo resultados de análise sobre a significância estatística de diferenças entre vazões percoladas. / This work reports the conclusions of studies on draining water from surface layers of a way structure that aim to contribute to the knowledge on the behavior of non saturated ground, and to spread the use of the finite element technique for studies on the water movement inside components that constitute these platforms. It was discussed: an introduction on highways in Brazil, topics that had motivated the studies, methods and processes for treatment of the non saturated phase of the water movement. It was described and commented topics on the relations between the format of ways and the draining in their structure, processes to estimate the percolation and infiltration outflows using simplified methods, topics on the evolution of the study movement of water and estimating measures on non saturated ground properties. For case study it was adopted the simulation using elements of solid volume and the finite element technique obtained from bibliographical data on saturated and non saturated sand and clay properties, including the hydraulic conductivity. The tested data had been gotten by simulation, and had been submitted to the test of hypotheses on differences of percolated water volumes. The main objective was reached discussing statistical results on the significance of the differences among percolated volume of water.

Associamos independentemente a cada vértice v de un grafo infinito G um raio de infecção aleatório R_v e definimos um modelo de percolação sujeito às seguintes regras: (1) no tempo zero só a raiz é declarada infectada, (2) um vértice é declarado infectado em um instante t, t>0, se está a uma distância no maximo R_v de algum vértice v previamente infectado, e (3) vértices infectados permanecem infectados para sempre. Dizemos que há sobrevivência em uma realização particular do modelo se o número final de vértices infectados é infinito. Neste trabalho damos condições suficientes sobre o grafo G para a transição de fase deste modelo, estabelecendo limitantes não triviais para o parâmetro crítico quando os raios R_v têm distribuição geometrica de parâmetro 1-p. Além disto, restringindo nosso estudo para o caso das árvores esfericamente simétricas, obtemos um melhor limitante superior para este parâmetro. Finalmente, concluímos que o parâmetro crítico para o modelo nas árvores homogêneas de grau d+1 se comporta assintoticamente como 1/(2d). / We assign independently to each vertex v of an infinite graph G, a random radius of infection R_v and define a percolation model subject to the following rules: (1) at time zero, only the root is declared infected, (2) a vertex is declared infected at time t, t>0, if it is at distance at most R_v of some vertex v previously infected, and (3) infected vertices stay infected forever. We say that there is survival in a particular realization of the model if the final number of infected vertices is infinite. In this work, we give sufficient conditions on the graph G for the phase transition of this model, by stating non-trivial bounds for the critical parameter when the radii have geometrical distribution with parameter 1-p. In addition, restricting our study to the case of the spherically symmetric trees, we obtain an improved upper bound for this critical parameter. Finally, we conclude that the critical parameter for the model on homogeneous trees of degree (d+1) behaves asymptotically as 1/(2d).

Conselho Nacional de Desenvolvimento CientÃfico e TecnolÃgico / Um nÃmero significativo de redes reais possui caracterÃsticas locais ou nÃo-locais bem definidas. NÃs estudamos como estas caracterÃsticas podem influenciar processos de navegaÃÃo e processos percolativos que venham a ocorrer nas mesmas. Primeiramente, estudamos o problema de navegaÃÃo em redes regulares com ligaÃÃes de longo alcance e sujeitas a um vÃnculo de custo. Neste caso, a rede à construÃda a partir de uma rede regular de dimensÃo d a ser melhorada por meio da adiÃÃo de ligaÃÃes de longo alcance (atalhos) com uma probabilidade $P_{ij} sim r_{ji}^{-alpha}$ , onde $r_{ij}$ à a distÃncia de Manhattan entre os sÃtios $i$ e $j$. Mostramos que a condiÃÃo de navegaÃÃo Ãtima, $alpha = d+1$, permanece Ãtima, independente da estratÃgia de navegaÃÃo utilizada, seja ela baseada em um conhecimento local ou global da estrutura da rede. Em seguida, apresentamos um processo de crescimento de agregados que fornece uma clara conexÃo entre a MecÃnica EstatÃstica no equilÃbrio e o processo percolativo nÃo-local conhecido como PercolaÃÃo Explosiva. Mostramos que dois ingredientes sÃo suficientes para obter uma transiÃÃo abrupta na fraÃÃo do sistema ocupada pelo maior agregado: (i) os tamanhos de todos os agregados devem ser mantidos aproximadamente iguais durante o processo percolativo e (ii) a inclusÃo de ligaÃÃes de fusÃo (i.e., ligaÃÃes que conectam agregados diferentes) deve dominar o processo em detrimento de ligaÃÃes redundantes (i.e., ligaÃÃes que conectam sÃtios em um mesmo agregado). Por Ãltimo, introduzimos um modelo que generaliza a regra do produto para PercolaÃÃo Explosiva que revela os efeitos da nÃo-localidade no comportamento crÃtico do processo de percolaÃÃo. Mais precisamente, pares de ligaÃÃes nÃo ocupadas sÃo escolhidos de acordo com uma probabilidade que decai em lei de potÃncia com sua distÃncia de Manhattan, e apenas a ligaÃÃo que conecta agregados para os quais o produto de seus tamanho à o menor, à ocupada. Nossos resultados para redes regulares finitas em diversas dimensÃes sugerem que, na criticalidade, o expoente da lei de potÃncia tem uma influÃncia significativa nos expoentes de escala, onde observa-se uma transiÃÃo nos expoentes da percolaÃÃo tradicional para os expoentes da percolaÃÃo explosiva (nÃo-local) em determinados casos. / A significant number of real networks have well-defined local and nonlocal features. We investigate the influence of these features in the navigation through small-world networks and in explosive percolation. First, we investigate the navigation problem in lattices with long-range connections and subject to a cost constraint. Our network is built from a regular d-dimensional lattice to be improved by adding long-range connections (shortcuts) with probability $P_{ij} sim r_{ij}^{-alpha}, where $r_{ij}$ is the Manhattan distance between nodes $i$ and $j$, and a is $alpha$ variable exponent. We find optimal transport in the system for $alpha = d+1$. Remarkably, this condition remains optimal, regardless of the strategy used for navigation being based on local or global knowledge of the network structure. Second, we present a cluster growth process that provides a clear connection between equilibrium statistical mechanics and the nonlocal explosive percolation process. We show that the following two ingredients are sufficient for obtaining an abrupt transition in the fraction of the system occupied by the largest cluster: (i) the size of all growing clusters should be kept approximately the same, and (ii) the inclusion of merging bonds (i.e., bonds connecting nodes in different clusters) should dominate with respect to the redundant bonds (i.e., bonds connecting nodes in the same cluster). Finally, we introduce a generalization of the product rule for explosive percolation that reveals the effect of nonlocality on the critical behavior of the percolation process. Precisely, pairs of unoccupied bonds are chosen according to a probability that decays as a power law of their Manhattan distance, and only that bond connecting clusters whose product of their sizes is the smallest becomes occupied. Our results for d-dimensional lattices at criticality shows that the power law exponent of the product rule has a significant influence on the finite-size scaling exponents for the spanning cluster, the conducting backbone, and the cutting bonds of the system. For all these types of clusters, we observe a clear transition from ordinary to (nonlocal) explosive percolation.

Tendo em vista o fato de que o sistema radicular das culturas modifica as propriedades do solo ao seu entorno e da praticamente ausência de trabalhos científicos para avaliação das diferenças estruturais entre solo e rizosfera, o objetivo deste estudo foi aprofundar o conhecimento destas diferenças, avaliando a agregação do solo junto às raízes e fora da sua influência, em diferentes rotações culturais. A amostragem foi realizada na camada superficial do solo (0,00 - 0,05 m) em cada sistema de rotação, considerando como rizosfera o solo aderido às raízes e como solo sem influência das raízes, aquele não aderido a elas. Para avaliar a estabilidade estrutural do solo, foram feitas as análises da argila prontamente dispersa em água, da argila mecanicamente dispersa em água e da estabilidade de percolação. As análises granulométrica e química também foram realizadas. Os dados obtidos foram submetidos à análise de variância e, quando significativo, ao teste de médias Tukey (p<=0,05). Posteriormente, foram feitas correlações entre os atributos físicos utilizados como indicadores da estabilidade dos agregados e os demais atributos avaliados. A rizosfera apresentou agregados mais estáveis em água do que o solo fora dela, sendo a argila mecanicamente dispersa em água o indicador mais sensível na detecção desse resultado. Foi evidênciado que o sistema radicular de diferentes culturas influenciaram de forma diferente a agregação das partículas do solo. As correlações mais significativas foram para a argila mecanicamente dispersa em água. / Considering the fact that the crops root system modifies the soil properties around it and in the almost absence of scientific works to evaluate the structural differences between soil and rhizosphere, the objective of this study was to increase the knowledge of these differences by evaluating soil aggregation close to the root and out of its influence, at different crop rotations. Sampling was performed in the soil surface layer (0.00 - 0.05 m) in each rotation system, considering as rhizosphere the soil adhered to the roots and as soil without influence of the roots, that not adhered to them. To evaluate the structural stability of the soil, analyzes of clay readily and mechanically dispersed in water and of the percolation stability were made. The granulometric and chemical analyzes were also made. Data obtained were analysed by the analysis of variance and, when significant, by Tukey test (p<=0.05). Then, the correlations were made between the physical attributes used as indicators of the aggregates stability and the other attributes evaluated. Rhizosphere presented more stable aggregates in water than the soil outside it, being the mechanically clay dispersed in water the most sensitive indicator in the detection of this outcome. It was evidenced that the root system of different crops influenced of different way the soil particles aggregation. The more significant correlations were those for clay mechanically dispersed in water.