Tendo em vista que os amidos (catiônicos ou não) não são nocivos a saúde do ser humano, uma vez que são utilizados na indústria de alimentos, e observando-se a potencialidade do seu uso como auxiliares de floculação, foi proposto o presente trabalho. Consiste na realização de ensaios de coagulação (com sulfato alumínio), floculação e sedimentação em equipamento de reatores estáticos, objetivando verificar a influência do gradiente de velocidade e do tempo de agitação na floculação com polímero sintético catiônico e amidos de milho e mandioca catiônicos. Concluiu-se que as condições de aplicação exercem influência na eficiência de remoção de turbidez e cor aparente, sendo que cada polímero estudado apresentou uma condição ótima específica. Em todos os ensaios realizados, o amido de mandioca catiônico foi o mais eficiente, indicando que os amidos catiônicos podem ser substitutos em potencial dos polímeros sintéticos no tratamento de águas de abastecimento, quando utilizados como auxiliares de floculação. / This present work was based on the fact that native and cationic starches are not harmful to man\'s health, since they have been largely used in food processing industries, and that they may be potentially used as flocculation aids in water treatment. Jar Test assays were performed, including coagulation with aluminum sulphate, flocculation and sedimentation, aiming to study the influence of the velocity gradient and mixing time on flocculation, using a cationic synthetic polymer and cationic com and manioc starches as aids. In the main, it was concluded that the conditions under which the polymers were applied, affect the efficiency of turbidity and color removals and that each polymer studied presented an specific optimum condition. It was also observed that manioc cationic starch resulted more effective than the others polymers studied, when used as flocculation aids.

Nous avons étudié la dynamique de Glauber sur les pavages de domaines finies du plan par des losanges ou par des dominos de taille 2 × 1. Ces pavages sont naturellement associés à des surfaces de R^3, qui peuvent être vues comme des interfaces dans des modèles de physique statistique. En particulier les pavages par des losanges correspondent au modèle d'Ising tridimensionnel à température nulle. Plus précisément les pavages d'un domaine sont en bijection avec les configurations d'Ising vérifiant certaines conditions au bord (dépendant du domaine pavé). Ces conditions forcent la coexistence des phases + et - ainsi que la position du bord de l'interface. Dans la limite thermodynamique où L, la longueur caractéristique du système, tend vers l'infini, ces interfaces obéissent à une loi des grand nombre et convergent vers une forme limite déterministe ne dépendant que des conditions aux bord. Dans le cas où la forme limite est planaire et pour les losanges, Caputo, Martinelli et Toninelli [CMT12] ont montré que le temps de mélange Tmix de la dynamique est d'ordre O(L^{2+o(1)}) (scaling diffusif). Nous avons généralisé ce résultat aux pavages par des dominos, toujours dans le cas d'une forme limite planaire. Nous avons aussi prouvé une borne inférieure Tmix ≥ cL^2 qui améliore d'un facteur log le résultat de [CMT12]. Dans le cas où la forme limite n'est pas planaire, elle peut être analytique ou bien contenir des parties “gelées” où elle est en un sens dégénérée. Dans le cas où elle n'a pas de telle partie gelée, et pour les pavages par des losanges, nous avons montré que la dynamique de Glauber devient “macroscopiquement proche” de l'équilibre en un temps L^{2+o(1)} / We studied the Glauber dynamics on tilings of finite regions of the plane by lozenges or 2 × 1 dominoes. These tilings are naturally associated with surfaces of R^3, which can be seen as interfaces in statistical physics models. In particular, lozenge tilings correspond to three dimensional Ising model at zero temperature. More precisely, tilings of a finite regions are in bijection with Ising configurations with some boundary conditions (depending on the tiled domain). These boundary conditions impose the coexistence of the + and - phases, together with the position of the boundary of the interface. In the thermodynamic limit where L, the characteristic length of the system, tends toward infinity, these interface follow a law of large number and converge to a deterministic limit shape depending only on the boundary condition. When the limit shape is planar and for lozenge tilings, Caputo, Martinelli and Toninelli [CMT12] showed that the mixing time of the dynamics is of order (L^{2+o(1)}) (diffusive scaling). We generalized this result to domino tilings, always in the case of a planar limit shape. We also proved a lower bound Tmix ≥ cL^2 which improve on the result of [CMT12] by a log factor. When the limit shape is not planar, it can either be analytic or have some “frozen” domains where it is degenerated in a sense. When it does not have such frozen region, and for lozenge tilings, we showed that the Glauber dynamics becomes “macroscopically close” to equilibrium in a time L^{2+o(1)}

<p>Izmenama u konstrukciji vertikalne me&scaron;alice zamenom originalne obloge pužnice sa oblogama koje su imale 13,5%, 27% i 40% perforacija je pobolj&scaron;an proces me&scaron;anja. Ova pobolj&scaron;anja se ogledaju u skraćenju vremena me&scaron;anja 15%, 30% i 50% u odnosu na originalnu konstrukciju, smanjenju utro&scaron;ka rada i energije i otvaranja mogućnosti vraćanja ove vrste me&scaron;alica u masovniju upotrebu, jer su tro&scaron;kovi kupovine i održavanja ove vrste me&scaron;alica daleko niži u odnosu na druge. Izrađen je model sistem za cilindričnu me&scaron;alicu, na bazi primene teorije Markovljevog niza, kao pogodan način za predviđanje procesa me&scaron;anja u ovoj vrsti me&scaron;alica, čak i u industrijskim uslovima. Dobijeni rezultati<br />su pokazali slaganje sa eksperimentalno dobijenim rezultatima, pri čemu je utvrđeno vreme neophodno za postizanje homogenosti u cilindričnoj me&scaron;alici iznosilo 5 minuta za odnos me&scaron;anja 1:10.000, dok za odnos me&scaron;anja 1:100.000 i vreme od 5 minuta, homogenost nije postignuta Poređenjem rezultata programa za analizu slike, koji je prethodno razvijen u okviru metode Microtracer-a, sa ručno dobijenim rezultatima, utvrđen je visok stepen korelacije (R=0.995, 0.979 i 0.987) i ovaj program je ocenjen kao značajno pobolj&scaron;anje i skraćenje vremena potrebnog za dobijanje rezultata.</p> / <p>Changes in the construction of vertical mixers by replacement of the original screw barrel with barrel which had 13.5%, 27% and 40% perforation generally have improved process of mixing. These improvements are reflected in shortening the mixing time of 15%, 30% and 50% comparing to the original construction, reduction of labor and energy and create the possibility of returning these types of mixers in the more common, as the cost of purchasing and maintaining these types of mixers is far lower compared to others. A model system for the cylindrical mixer, based on the theory of Markov chain, as a convenient way to predict the mixing process in this type of mixer, even in industrial conditions. The results show agreement with experimental results, with a determined time necessary to achieve homogeneity in a cylindrical mixer was 5 minutes for mixing ratio 1:10,000, and more than 5 minutes for the mixing ratio 1:100,000. Comparing the results of program for image analysis, which was previously developed under a Microtracers methods with manually obtained results revealed a high degree of correlation (R = 0995, 0979 and 0987) and this program was rated as significantly improving and shortening the time needed to obtain results</p>

This study aims to investigate the possibility of using calcium-alginate as a coagulant in low turbidity waters. Jar tests were initially performed with synthetically prepared turbid waters to investigate the effect of alginate and calcium concentrations, alginate&rsquo / s molecular weight, rapid mixing time and speed (schedule), initial pH and alkalinity of synthetic water on turbidity removal efficiency of calcium-alginate system step by step. Alum as a coagulant was then used in jar tests conducted with synthetic water to compare with calcium-alginate in terms of its turbidity removal efficiency and produced sludge properties. Finally, raw water acquired from water treatment plant was tested for treatability by using calcium-alginate based on previously determined optimum parameters via jar tests. Experiments of calcium-alginate system with synthetic water showed that calcium was a key parameter in coagulation and high molecular weight alginate performed better in turbidity removal. Significant improvements in turbidity removal were observed when mixing schedules were rearranged / especially in case of increasing rapid mixing time following calcium dosing. Calcium-alginate system neither was notably affected by pH or alkalinity nor did significantly change the pH or alkalinity of the medium. Alum worked well in turbidity removal with additional adjustments of pH and alkalinity, however / alum produced higher quantities of sludge than calcium-alginate system. &ldquo / Raw water&ldquo / experiments with calcium-alginate did not result in desired level of turbidity removals due to the excessively different characteristics of this water compared to the synthetically prepared turbid water.

Submitted by Luciana Sebin (lusebin@ufscar.br) on 2016-09-14T18:52:44Z No. of bitstreams: 1 TeseLJC.pdf: 3758058 bytes, checksum: 4551fb3e08abee5796d1f863288fd5ae (MD5) / Approved for entry into archive by Marina Freitas (marinapf@ufscar.br) on 2016-09-15T14:23:21Z (GMT) No. of bitstreams: 1 TeseLJC.pdf: 3758058 bytes, checksum: 4551fb3e08abee5796d1f863288fd5ae (MD5) / Approved for entry into archive by Marina Freitas (marinapf@ufscar.br) on 2016-09-15T14:23:30Z (GMT) No. of bitstreams: 1 TeseLJC.pdf: 3758058 bytes, checksum: 4551fb3e08abee5796d1f863288fd5ae (MD5) / Made available in DSpace on 2016-09-15T14:23:43Z (GMT). No. of bitstreams: 1 TeseLJC.pdf: 3758058 bytes, checksum: 4551fb3e08abee5796d1f863288fd5ae (MD5) Previous issue date: 2016-02-04 / Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) / In this work it was evaluated, firstly, the performance of four impellers configurations in the enzymatic hydrolysis of sugarcane bagasse. The configurations evaluated were: (1): Rushton turbine - Rushton turbine; (2): Elephant ear down-pumping and Rushton turbine; (3): Rushton turbine and Elephant ear up-pumping; (4): Elephant ear downpumping and Elephant ear up-pumping. The choice of the best impeller configuration was based in mixing efficiency, characterized by the mixing time. The configurations were also evaluated considering the conversion of cellulose to glucose, power consumption as well as the rheological behavior during hydrolysis. The hydrolysis experiments were carried out in batch stirred tank reactor (3 L) using 10% w/v of solids (pH 4.8; 50°C; 470 rpm), 10 FPU· g-1 biomassa for 96 h. The configuration (4) showed the lowest mixing time and energy efficiency values (ratio of conversion of cellulose to glucose and total energy consumption) of 78.9%·MJ-1. Further, to get a high concentration of glucose associated with low power consumption, it was investigated two operating modes: batch and fed-batch. The strategies evaluated were: E1 [20%]; E2 [10(E)+5+5%]; E3 [5(E)+5+5+5%]; E4 [5(E)+5+5+5%], and E5 5(E)+5(E)+5(E)+5(E)%]. The best energy efficiency was obtained for the E5 strategy in which substrate and enzyme were added simultaneously (0.35 kgglicose·kWh-1). This value was 52% higher than that obtained in the single batch operation (E1). In continuation of the work were carried out enzymatic hydrolysis of exploded and hydrothermal bagasse and cane straw submitted to hydrothermal pretreatment. The experiments were carried out under the conditions: solids loading of 10 (w/v), pH 4.8; 50 ° C; 470 rpm and 10 FPU·g-1 biomass for 96 h. The efficiency obtained in the enzymatic hydrolysis of steam explosion sugarcane bagasse proved to be 41 and 46% higher than the hydrolysis of hydrothermally pretreated sugarcane straw and bagasse, respectively. Finally, a scale-up protocol with a scale factor equal to 1000 was proposed. It was analyzed the maintenance of two parameters on larger scale: the constancy of the mixing time (tm) and the constancy of the power consumption per unit volume (P/V). In turn, maintenance P/V parameter constant, the mixing time and the new scale power consumption (3000L) were approximately 4 and 1000 times higher, respectively, than those values obtained in the smaller scale (3L). / Neste trabalho avaliou-se, primeiramente, o desempenho de quatro configurações de impelidores na hidrólise enzimática do bagaço explodido de cana-de-açúcar. As configurações avaliadas foram: (1): turbina Rushton – turbina Rushton; (2): Elephant ear down-pumping e turbina Rushton; (3): turbina Rushton e Elephant ear up-pumping; (4): Elephant ear down-pumping e Elephant ear up-pumping. A escolha da melhor configuração de impelidores foi baseada na eficiência de mistura, caracterizada pelo tempo de mistura. As configurações também foram avaliadas considerando a conversão de celulose em glicose, o consumo de potência, bem como o comportamento reológico durante a hidrólise. Os experimentos de hidrólise em batelada foram realizados em reator tipo tanque agitado (3 L) utilizando 10% m/v de sólidos (pH 4,8; 50°C; 470 rpm), 10 FPU·g-1 biomassa por 96 h. A configuração (4) apresentou os menores valores de tempo de mistura e uma eficiência energética de 78,9 %·MJ-1. Com o intuito de obter-se alta concentração de glicose associada a um baixo consumo de potência, investigou-se dois modos de operação: batelada e batelada alimentada. As estratégias avaliadas foram: E1 [20%]; E2 [10(E)+5+5%]; E3 [5(E)+5+5+5%]; E4 [5(E)+5+5+5%] e E5 [5(E)+5(E)+5(E)+5(E)%]. Os melhores resultados foi obtido na estratégia E5, obtendo uma eficiência energética de 0,35 kgglicose∙kWh-1. Este valor foi 52% maior do que o obtido na operação em batelada simples (E1). Na continuação do trabalho, foi realizada a hidrólise enzimática do bagaço explodido e hidrotérmico e a palha de cana submetida ao pré-tratamento hidrotérmico. Os experimentos foram realizados em um reator de 3 L nas seguintes condições: carga de sólidos de 10% m/v (pH 4,8; 50°C, 470rpm ) e 10 FPU·g-1 biomassa por 96 h. A eficiência energética obtida na hidrólise do bagaço explodido mostrou-se 41 e 46 % superior aos pré-tratados de bagaço e palha hidrotermicamente, respectivamente. Finalmente, foi proposto um protocolo de aumento de escala com um fator de escala igual a 1000. Analisou a manutenção de dois parâmetros na escala maior: tempo de mistura (tm) e do consumo de potência por unidade de volume (P/V). Com a manutenção do parâmetro P/V constante, o tempo de mistura e consumo de potência da nova escala (3000L) foram aproximadamente 4 e 1000 vezes maiores, respectivamente, do que o obtido na escala menor escala (3L). / FAPESP: 2011/23807-1

Tendo em vista que os amidos (catiônicos ou não) não são nocivos a saúde do ser humano, uma vez que são utilizados na indústria de alimentos, e observando-se a potencialidade do seu uso como auxiliares de floculação, foi proposto o presente trabalho. Consiste na realização de ensaios de coagulação (com sulfato alumínio), floculação e sedimentação em equipamento de reatores estáticos, objetivando verificar a influência do gradiente de velocidade e do tempo de agitação na floculação com polímero sintético catiônico e amidos de milho e mandioca catiônicos. Concluiu-se que as condições de aplicação exercem influência na eficiência de remoção de turbidez e cor aparente, sendo que cada polímero estudado apresentou uma condição ótima específica. Em todos os ensaios realizados, o amido de mandioca catiônico foi o mais eficiente, indicando que os amidos catiônicos podem ser substitutos em potencial dos polímeros sintéticos no tratamento de águas de abastecimento, quando utilizados como auxiliares de floculação. / This present work was based on the fact that native and cationic starches are not harmful to man\'s health, since they have been largely used in food processing industries, and that they may be potentially used as flocculation aids in water treatment. Jar Test assays were performed, including coagulation with aluminum sulphate, flocculation and sedimentation, aiming to study the influence of the velocity gradient and mixing time on flocculation, using a cationic synthetic polymer and cationic com and manioc starches as aids. In the main, it was concluded that the conditions under which the polymers were applied, affect the efficiency of turbidity and color removals and that each polymer studied presented an specific optimum condition. It was also observed that manioc cationic starch resulted more effective than the others polymers studied, when used as flocculation aids.

Continuous milli-scale and micro-scale structures such as FlowPlate® microreactors have emerged as a promising element of process intensification due to their inherently effective rates of mass and heat transfer. These microfluidic devices have proven to be a preferred solution in place of energy-intensive batch processes for certain pathways of fine chemical and pharmaceutical synthesis, most notably fast reactions taking place on the scale of milliseconds to seconds. Computational fluid dynamics (CFD) has become an increasingly valuable tool in the field of microreactor design and optimization for its ability to locally map complex fluid flow patterns and resolve microscopic scales of reactive mixing that are challenging to characterize experimentally. The primary objective of this research was thus to develop and validate a mathematical model for the simulation of chaotic flow and homogeneous mixing in continuous microreactors. The model needed to be versatile enough to handle transition between flow regimes within a given reactor as well as the coexistence of both chaotic and laminar flow patterns in the micromixing elements that comprise said reactors. This was successfully achieved through the implementation of a k-ω SST (shear-stress transport) turbulence model that accounts for the impact of small-scale temporal and spatial fluctuations generated in the micromixer geometries studied herein; be it a liquid-liquid mixer (LLM), a serpentine (SZ) or a tangential (TG) mixer. In a first CFD study, the computational predictions were validated based on excellent agreement with experimental pressure loss (R^2 > 0.997) and residence time distribution (RTD) data (R^2 > 0.97) in several LL microreactors at Reynolds numbers ranging from 210 to 2140. Furthermore, the local velocity distribution and streamlines were mapped across the 3D domain of these reactors and it was discovered, based on the emergence of advective recirculation zones and turbulent dispersion, that a drastic change in flow behaviour occurred in these mixing elements at a Reynolds number of about 640. The interspacing of LLM elements with straight microchannels proved to be a suitable approach to modulating pressure loss while concurrently maintaining the chaotic secondary flow patterns generated from the mixers. In a second CFD study, the impact of micromixer geometry on the local velocity fields and advective transport performance was investigated both from a macromixing and micromixing perspective. Like the LLM, the SZ and TG mixers conferred chaotic secondary flow patterns at characteristic Reynolds numbers between 500 and 1000. As such, it was concluded that it would be ideal to operate these mixers at water flow rates of at least 30 ml/min. Contour plots of the velocity magnitude coupled with the computation of RTD showed that the SZ virtually mimics a plug-flow profile over a volume of 77 mm3 or greater at 50 g/min. The RTD of the LLM and TG resembles that of a mixed flow pattern given that approximately 65-80% of their fluid volume is occupied by recirculation zones. As such, it required 65 LLMs in series (3105 mm3) and 80 TGs (1142 mm3) to approach the same pattern as 10 SZs (77 mm3) from a macromixing perspective. Micromixing time distributions (MTD) were also characterized by locally computing the decay time of small-scale segregation (t_SSS) as a function of flow rate, wherein higher flow rates generated lower characteristic mixing times. The TG and LLM conferred the broadest range of mixing times, spanning nearly four orders of magnitude in the range of [0.02 ms, 10 ms], whereas the SZ generated a much narrower MTD ranging between [0.024 ms, 0.69 ms]. Finally, the impact of geometry and flow conditions on reaction yield was assessed by characterizing the extent of a finite-rate reaction relative to an infinitely fast reaction taking place in parallel. The calculated yield for the competitive-parallel reaction scheme showed that the second Damköhler number (Dall) computed based on the mean tSSS provides useful information about whether the process will be limited by the intrinsic rate of reaction or by the rate of mass transfer, even though the reaction process is controlled by a combination of the RTD as well as loss of LSS and SSS. It was concluded that the change in MTD as a function of power dissipation should coincide with the reaction yield response, and that any deviation in that relationship is because of macroscopic blending of reactants in the entrance region.

La première partie de cette thèse concerne l'inférence de modèles statistiques non normalisés. Nous étudions deux méthodes d'inférence basées sur de l'échantillonnage aléatoire : Monte-Carlo MLE (Geyer, 1994), et Noise Contrastive Estimation (Gutmann et Hyvarinen, 2010). Cette dernière méthode fut soutenue par une justification numérique d'une meilleure stabilité, mais aucun résultat théorique n'avait encore été prouvé. Nous prouvons que Noise Contrastive Estimation est plus robuste au choix de la distribution d'échantillonnage. Nous évaluons le gain de précision en fonction du budget computationnel. La deuxième partie de cette thèse concerne l'échantillonnage aléatoire approché pour les distributions de grande dimension. La performance de la plupart des méthodes d’échantillonnage se détériore rapidement lorsque la dimension augmente, mais plusieurs méthodes ont prouvé leur efficacité (e.g. Hamiltonian Monte Carlo, Langevin Monte Carlo). Dans la continuité de certains travaux récents (Eberle et al., 2017 ; Cheng et al., 2018), nous étudions certaines discrétisations d’un processus connu sous le nom de kinetic Langevin diffusion. Nous établissons des vitesses de convergence explicites vers la distribution d'échantillonnage, qui ont une dépendance polynomiale en la dimension. Notre travail améliore et étend les résultats de Cheng et al. pour les densités log-concaves. / The first part of this thesis concerns the inference of un-normalized statistical models. We study two methods of inference based on sampling, known as Monte-Carlo MLE (Geyer, 1994), and Noise Contrastive Estimation (Gutmann and Hyvarinen, 2010). The latter method was supported by numerical evidence of improved stability, but no theoretical results had yet been proven. We prove that Noise Contrastive Estimation is more robust to the choice of the sampling distribution. We assess the gain of accuracy depending on the computational budget. The second part of this thesis concerns approximate sampling for high dimensional distributions. The performance of most samplers deteriorates fast when the dimension increases, but several methods have proven their effectiveness (e.g. Hamiltonian Monte Carlo, Langevin Monte Carlo). In the continuity of some recent works (Eberle et al., 2017; Cheng et al., 2018), we study some discretizations of the kinetic Langevin diffusion process and establish explicit rates of convergence towards the sampling distribution, that scales polynomially fast when the dimension increases. Our work improves and extends the results established by Cheng et al. for log-concave densities.

The theoretical part of the diploma thesis recapitulates the existing literature in the field of recycling asphalt mixtures with a higher content of RAP (Reclaimed asphalt pavement). The practical part of the diploma thesis examines changes in the properties of asphalt mixtures with a higher content of RAP, due to changes in mixing time. Five mixtures have been proposed for this research. It is an asphalt mixture intended for the road abrasive layer (ACO 11+). Reference mixture "A" without RAP, with a standard mixing time of 25 seconds. Mixture "B" with 40% RAP, without rejuvenator, with an extended mixing time of 40 seconds. Mixture "C" with 40% RAP, with rejuvenator and standard mixing time. Mixture "D" with 40% RAP with rejuvenator and extended mixing time of 40 seconds. Mixture "E" with 40% RAP, with rejuvenator and extended mixing time of 55 seconds.

L’objectif de ce travail est une approche pluridisciplinaire dans le but de comprendre l’impact des phénomènes microscopiques, l’interaction charge-matrice avec ou sans agent de couplage et la dispersion des charges sur les propriétés macroscopiques. La caractérisation de l’impact de la stratégie de mélange avec ou sans un agent de couplage, la cinétique, l’état final et les différents scénarii de dispersion possibles sont ainsi étudiés. L’incorporation de microperles de silice dans une matrice SBR est réalisée dans un mélangeur interne. Finalement, la silice est dispersée à l’échelle nanométrique (10-100 nm). Les nanocomposites obtenus sont alors caractérisés par des techniques complémentaires (spectroscopie mécanique, MET, mesure du taux d’élastomère lié à la charge) dans le but de caractériser quantitativement les interactions charge-matrice et charge-charge. L’outil rhéologique est un outil très sensible pour caractériser l’évolution de la dispersion de charges dans une matrice élastomère. En particulier, le module de conservation G’ montre un plateau significatif pour les faibles fréquences de déformation. Ce plateau est très sensible à l’état de dispersion ainsi qu’à la nature des interactions entre les charges. En couplant les mesures rhéologiques, l'analyse d'image faite sur des photos MET et la mesure du taux d’élastomère lié, il est possible d’établir un scénario de dispersion de la silice dans l'élastomère en fonction des conditions de mélange et de mettre en évidence les paramètres élémentaires de la dispersion impliqués. En complément, la modélisation de certains modules montre toute sa pertinence dans la caractérisation de la rupture des agglomérats ou de l’évolution des interactions charge-charge au cours du mélangeage. Enfin, la dimension fractale des réseaux de charges obtenus est déterminée à partir de nos descripteurs de la dispersion / Filler dispersion in an elastomeric matrix, states and mechanisms of dispersion had to be investigated throughout the mixing process. This work focuses on a multidisciplinary approach to understand how microscopic phenomena, like rubber-filler interaction or filler dispersions, affect macroscopic properties such as rheological behavior. The incorporation of silica is realized in an internal mixer under temperature control. Finally, silica is dispersed at the nanoscale (10-100 nm). Afterwards, nanocomposites are characterized using complementary techniques in order to discuss quantitatively the nature of rubber-filler and filler-filler interactions and their effect on rheological properties. Thus, the global evolution of dispersion during the mixing is understood through these various tests. Different mechanisms in the dispersion have been observed. First, intense particle size reduction occurs at the earliest mixing times. Then, the aggregate size does not change while the amount of physically bound rubber at the surface of aggregate increases and levels off. For some silica, a second dispersion stage has been observed after the diffusion of the elastomer to the core of the aggregates. Rheology has showed to be a very sensitive tool to characterize the evolution of the dispersion in the system. Particularly, the complex shear modulus exhibits a significant plateau (Ge), at low frequency, which is very sensitive to the dispersion state and the nature of the interaction between the fillers. There is a striking correlation between the value of plateau Ge and the bound rubber content. Finally, a dispersion scenario has been established and fundamental interaction parameters have been identified