High-strain composites and dual-matrix composite structures

Most space applications require deployable structures due to the limiting size of current launch vehicles. Specifically, payloads in nanosatellites such as CubeSats require very high compaction ratios due to the very limited space available in this typo of platform. Strain-energy-storing deployable structures can be suitable for these applications, but the curvature to which these structures can be folded is limited to the elastic range. Thanks to fiber microbuckling, high-strain composite materials can be folded into much higher curvatures without showing significant damage, which makes them suitable for very high compaction deployable structure applications. However, in applications that require carrying loads in compression, fiber microbuckling also dominates the strength of the material. A good understanding of the strength in compression of high-strain composites is then needed to determine how suitable they are for this type of application.

The goal of this thesis is to investigate, experimentally and numerically, the microbuckling in compression of high-strain composites. Particularly, the behavior in compression of unidirectional carbon fiber reinforced silicone rods (CFRS) is studied. Experimental testing of the compression failure of CFRS rods showed a higher strength in compression than the strength estimated by analytical models, which is unusual in standard polymer composites. This effect, first discovered in the present research, was attributed to the variation in random carbon fiber angles respect to the nominal direction. This is an important effect, as it implies that microbuckling strength might be increased by controlling the fiber angles. With a higher microbuckling strength, high-strain materials could carry loads in compression without reaching microbuckling and therefore be suitable for several space applications.

A finite element model was developed to predict the homogenized stiffness of the CFRS, and the homogenization results were used in another finite element model that simulated a homogenized rod under axial compression. A statistical representation of the fiber angles was implemented in the model. The presence of fiber angles increased the longitudinal shear stiffness of the material, resulting in a higher strength in compression. The simulations showed a large increase of the strength in compression for lower values of the standard deviation of the fiber angle, and a slight decrease of strength in compression for lower values of the mean fiber angle. The strength observed in the experiments was achieved with the minimum local angle standard deviation observed in the CFRS rods, whereas the shear stiffness measured in torsion tests was achieved with the overall fiber angle distribution observed in the CFRS rods.

High strain composites exhibit good bending capabilities, but they tend to be soft out-of-plane. To achieve a higher out-of-plane stiffness, the concept of dual-matrix composites is introduced. Dual-matrix composites are foldable composites which are soft in the crease regions and stiff elsewhere. Previous attempts to fabricate continuous dual-matrix fiber composite shells had limited performance due to excessive resin flow and matrix mixing. An alternative method, presented in this thesis uses UV-cure silicone and fiberglass to avoid these problems. Preliminary experiments on the effect of folding on the out-of-plane stiffness are presented. An application to a conical log-periodic antenna for CubeSats is proposed, using origami-inspired stowing schemes, that allow a conical dual-matrix composite shell to reach very high compaction ratios.

Death or taxes? The political economy of sanitation expenditure in nineteenth-century Britain

This thesis consists of three papers studying the relationship between democratic reform, expenditure on sanitation public goods and mortality in Britain in the second half of the nineteenth century. During this period decisions over spending on critical public goods such as water supply and sewer systems were made by locally elected town councils, leading to extensive variation in the level of spending across the country. This dissertation uses new historical data to examine the political factors determining that variation, and the consequences for mortality rates.

The first substantive chapter describes the spread of government sanitation expenditure, and analyzes the factors that determined towns' willingness to invest. The results show the importance of towns' financial constraints, both in terms of the available tax base and access to borrowing, in limiting the level of expenditure. This suggests that greater involvement by Westminster could have been very effective in expediting sanitary investment. There is little evidence, however, that democratic reform was an important driver of greater expenditure.

Chapter 3 analyzes the effect of extending voting rights to the poor on government public goods spending. A simple model predicts that the rich and the poor will desire lower levels of public goods expenditure than the middle class, and so extensions of the right to vote to the poor will be associated with lower spending. This prediction is tested using plausibly exogenous variation in the extent of the franchise. The results strongly support the theoretical prediction: expenditure increased following relatively small extensions of the franchise, but fell once more than approximately 50% of the adult male population held the right to vote.

Chapter 4 tests whether the sanitary expenditure was effective in combating the high mortality rates following the Industrial Revolution. The results show that increases in urban expenditure on sanitation-water supply, sewer systems and streets-was extremely effective in reducing mortality from cholera and diarrhea.

Energy loss of light ions channeling in silicon

Experimental measurements of rate of energy loss were made for protons of energy .5 to 1.6 MeV channeling through 1 μm thick silicon targets along the <110>, <111>, and <211> axial directions, and the {100}, {110}, {111}, and {211} planar directions. A .05% resolution automatically controlled magnetic spectrometer was used. The data are presented graphically along with an extensive summary of data in the literature. The data taken cover a wider range of channels than has previously been examined, and are in agreement with the data of F. Eisen, et al., Radd. Eff. 13, 93 (1972).

The theory in the literature for channeling energy loss due to interaction with local electrons, core electrons, and distant valence electrons of the crystal atoms is summarized. Straggling is analyzed, and a computer program which calculates energy loss and straggling using this theory and the Moliere approximation to the Thomas Fermi potential, V_TF, and the detailed silicon crystal structure is described. Values for the local electron density Z_loc in each of the channels listed above are extracted from the data by graphical matching of the experimental and computer results.

Zeroth and second order contributions to Z_loc as a function of distance from the center of the channel were computed from ∇²V_TF = 4πρ for various channels in silicon. For data taken in this work and data of F. Eisen, et al., Rad. Eff. 13, 93 (1972), the calculated zeroth order contribution to Z_loc lies between the experimentally extracted Z_loc values obtained by using the peak and the leading edge of the transmission spectra, suggesting that the observed straggling is due both to statistical fluctuations and to path variation.

A tool for better management of water quality during periods of algal blooms: the case of the River Oise

Like other rivers in the Paris area, the Oise is subject to important seasonal algal blooms. This eutrophication generates notable problems for the production of drinking-water from a treatment plant on the river at Méry. A mathematical model has been developed to simulate variation in water quality in a pre-treatment storage basin, and another model is currently being adapted to model the River Oise. Integration of the two models should provide a comprehensive tool for predicting variations of phytoplankton and water-quality parameters associated with algal blooms. This will be a decision-aid for optimizing control of the treatment process for providing potable water.

双菱体λ/4消色差相位延迟器的设计

根据菲涅尔全内反射相变理论,给出了双菱体λ/4消色差器的结构设计、性能分析和测量方法.由有效通光孔径和光线追迹设计出BK7玻璃在波长532 nm时相位延迟λ/4的双菱体的结构,用作532 nm至1 064 nm波长范围的标准λ/4相位延迟器.理论分析了入射角变化和波长变化对双菱体相位延迟的影响,当入射角变化限制在±4.3°以内时,其影响得到补偿；波长从532 nm到1 064 nm产生的误差为－0.65°.采用椭偏法中的消光技术,分别实测了双菱体在532 nm和1 064 nm波长下的相位延迟为：90.08±0.14°和88.99±0.1°,可知两不同波长产生的相位延迟误差为－1.09°.

Dieta do boto-cinza (Sotalia guianensis Van Benédén, 1864) (Cetacea: Delphinidae) na Baía de Sepetiba (RJ)

O boto-cinza (Sotalia guianensis Van Benédén, 1864) é um pequeno cetáceo da família Delphinidae. Ocorre em águas costeiras da América do Sul e Central, associado à ambientes estuarinos, às baías e áreas protegidas. Estudos sobre hábitos alimentares são importantes para avaliar os padrões de relação entre presa e predador. Desta forma, o conhecimento da composição da dieta do predador pode fornecer informações a respeito de sua distribuição, padrões de migração e de seu comportamento, além de contribuir com informações sobre a biologia e comportamento de suas presas. Neste trabalho, são abordados aspectos da ecologia alimentar do boto-cinza da Baía de Sepetiba, com o objetivo de caracterizar a dieta e compará-la em relação ao sexo, classe etária e estações do ano, a partir da análise do conteúdo estomacal de 76 botos-cinza encalhados entre 2005 e 2011. Os itens alimentares recuperados dos estômagos como otólitos e ossos de peixes, bicos de cefalópodes, carapaças de crustáceos, foram identificados baseando-se em trabalhos de identificação e em uma coleção de referência. O comprimento e biomassa das presas foram estimados com o uso de equações de regressão encontradas na literatura para estas espécies. As amostras foram separadas em machos adultos, fêmeas adultas e juvenis, e em estações quente/chuvosa (out-abr) e fria/seca (mai-set), em que os estômagos foram recuperados. Um total de 1800 presas foi identificado, relativo a 23 espécies de teleósteo, quatro espécies de cefalópode e três de crustáceo. O boto-cinza da Baía de Sepetiba apresentou uma dieta tipicamente piscívora, com um hábito alimentar no qual, poucas espécies foram consumidas em alta frequencia. Todas as presas identificadas têm distribuição costeira sendo a maioria estuarina de pequeno porte ou juvenil. As presas mais importantes na dieta do boto-cinza, segundo o índice de importância relativa (IIR), foram Cetengraulis edentulus; Micropogonias furnieri; Mugil spp.; Chloroscombrus chrysurus; Cynoscion jamaicensis; Stellifer sp.; e Sciadeichthys luniscutis. Dentre os cefalópodes, a lula Doryteuthis plei foi à presa mais importante. Este estudo indica que o boto-cinza apresenta variações intra-específicas no seu hábito alimentar entre fêmeas adultas, machos adultos e juvenis, além de variações sazonais na composição de sua dieta. A partir do conhecimento do comportamento e hábito de suas presas, pode-se concluir que, Sotalia guianensis da Baía de Sepetiba se alimenta ao longo de toda a Baía, além de utilizar áreas costeiras próximas para atividades de alimentação e forrageio.

Community Turnover of Wood-Inhabiting Fungi across Hierarchical Spatial Scales

For efficient use of conservation resources it is important to determine how species diversity changes across spatial scales. In many poorly known species groups little is known about at which spatial scales the conservation efforts should be focused. Here we examined how the community turnover of wood-inhabiting fungi is realised at three hierarchical levels, and how much of community variation is explained by variation in resource composition and spatial proximity. The hierarchical study design consisted of management type (fixed factor), forest site (random factor, nested within management type) and study plots (randomly placed plots within each study site). To examine how species richness varied across the three hierarchical scales, randomized species accumulation curves and additive partitioning of species richness were applied. To analyse variation in wood-inhabiting species and dead wood composition at each scale, linear and Permanova modelling approaches were used. Wood-inhabiting fungal communities were dominated by rare and infrequent species. The similarity of fungal communities was higher within sites and within management categories than among sites or between the two management categories, and it decreased with increasing distance among the sampling plots and with decreasing similarity of dead wood resources. However, only a small part of community variation could be explained by these factors. The species present in managed forests were in a large extent a subset of those species present in natural forests. Our results suggest that in particular the protection of rare species requires a large total area. As managed forests have only little additional value complementing the diversity of natural forests, the conservation of natural forests is the key to ecologically effective conservation. As the dissimilarity of fungal communities increases with distance, the conserved natural forest sites should be broadly distributed in space, yet the individual conserved areas should be large enough to ensure local persistence.

Estimating relative abundance from catch and effort data, using neural networks

We develop and test a method to estimate relative abundance from catch and effort data using neural networks. Most stock assessment models use time series of relative abundance as their major source of information on abundance levels. These time series of relative abundance are frequently derived from catch-per-unit-of-effort (CPUE) data, using general linearized models (GLMs). GLMs are used to attempt to remove variation in CPUE that is not related to the abundance of the population. However, GLMs are restricted in the types of relationships between the CPUE and the explanatory variables. An alternative approach is to use structural models based on scientific understanding to develop complex non-linear relationships between CPUE and the explanatory variables. Unfortunately, the scientific understanding required to develop these models may not be available. In contrast to structural models, neural networks uses the data to estimate the structure of the non-linear relationship between CPUE and the explanatory variables. Therefore neural networks may provide a better alternative when the structure of the relationship is uncertain. We use simulated data based on a habitat based-method to test the neural network approach and to compare it to the GLM approach. Cross validation and simulation tests show that the neural network performed better than nominal effort and the GLM approach. However, the improvement over GLMs is not substantial. We applied the neural network model to CPUE data for bigeye tuna (Thunnus obesus) in the Pacific Ocean.

Energy efficiency in trawling operations

Diurnal variation in trawl catches and its influence on energy efficiency of trawler operations are discussed in this paper, based on data on landings of a Japanese factory trawler which operated in the Indian waters during 1992-93. The factory vessel equipped for stern trawling had a length overall of 110 m, GT of 5460 and installed engine power of 5700 hp. Operations were conducted off west coast of India between 31 and 278 m depth contours, using a 80.4 m high opening bottom trawl with an adjusted vertical opening of 7.60.9 m. The catch data was grouped according to the median towing hour, by the time of the day. CPUE obtained was 3713.4 kg.h-1 for day time operations and 1536.6 kg.h-1 for night-time operations. Mean daily catches were 31367 kg.day-1 (SE: 2743) for day time operations and 9430 kg.day-1 (SE: 966) for night-time operations. Fuel consumption were 0.399 and 0.982 kg fuel.kg fish-1, respectively for day and night-time operations. Total catch and catch components such as threadfin bream, bulls eye, hairtails, trevelly, lizard fish showed significant improvement during day-time operations while swarming crabs showed a significant improvement in the night-time operations. The difference in catch rates between day and night could be attributed to diurnal variation in the spatial distribution and schooling behaviour of the catch categories, their differential behaviour in the vicinity of trawl systems under varying light levels of day and night and consequent effect on catching efficiency and size selectivity at different stages in the capture process. The results obtained in addition to its importance in the operational planning of trawling in order to realise objectives of maximising catch per unit effort and minimising fuel consumption per unit volume of fish caught, has added significance in the use of bottom trawl surveys in stock abundance estimates.

Evolução temporal e fatores climáticos relacionados à leishmaniose visceral em municípios endêmicos brasileiros no período de 2001 a 2008

Este é um estudo ecológico misto onde o perfil de distribuição de chuvas e das variações térmicas foi relacionado à variação sazonal das taxas de incidência de leishmaniose visceral em vinte municípios brasileiros de transmissão intensa da doença, no período de 2001 a 2008. O objetivo foi identificar similaridades e diferenças entre os municípios estudados quanto à tendência temporal e sazonalidade da doença e à possível relação entre variações climáticas e a distribuição sazonal da doença. Os dados de incidência de leishmaniose visceral foram obtidos do Sistema de Informação de Agravos de Notificação (Sinan), os dados demográficos foram obtidos do Departamento de Informática do Sistema Único de Saúde (DATASUS), os dados pluviométricos e de temperatura foram obtidos do Sistema de Monitoramento Agrometeorológico do Ministério da Agricultura (AGRITEMPO). Os resultados são apresentados graficamente e mostram que a distribuição sazonal da incidência e períodos prováveis de transmissão são diferentes em vários municípios e acompanham as diferenças climáticas, sugerindo que as intervenções que visem diminuir a incidência devem ser pontuais, obedecendo às características de cada município.

ZrO2／SiO2多层膜中膜厚组合周期数及基底材料对残余应力的影响

ZrO2／SiO2多层膜由相同沉积条件下的电子束蒸发方法制备而成，通过改变多层膜中高(ZrO2)、低(SiO2)折射率材料膜厚组合周期数的方法，研究了沉积在熔石英和BK7玻璃基底上多层膜中残余应力的变化．用ZYGO光学干涉仪测量了基底镀膜前后曲率半径的变化，并确定了薄膜中的残余应力．结果发现，该多层膜中的残余应力为压应力，随着薄膜中膜厚组合周期数的增加，压应力值逐渐减小．而且在相同条件下，石英基底上所沉积多层膜中的压应力值要小于BK7玻璃基底上所沉积多层膜中的压应力值．用X射线衍射技术测量分析了膜厚组合周

沉积参量及时效时间对SiO2薄膜残余应力的影响

SiO2薄膜由电子束蒸发方法沉积而成。用GPI数字波面光学干涉仪测量了不同沉积条件下玻璃基底镀膜前后曲率半径的变化，并确定了SiO2薄膜中的残余应力。在其他条件相同的情况下，当沉积温度由190℃升高到350℃时，SiO2薄膜中的压应力由一156MPa增大为-289MPa。氧分压由3．0×10^-3Pa升高到13．0×10^-3Pa时，SiO2薄膜中的应力由-223．5MPa变为20．4MPa。通过对薄膜折射率的测量，发现薄膜的堆积密度随沉积条件的改变也发生了规律性的变化。应力的变化主要是由于沉积时蒸发粒子

沉积温度对HfO2薄膜残余应力的影响

用电子柬蒸发方法制备了HfO2薄膜，根据镀膜前后基片曲率半径的变化，用Stoney公式计算了薄膜应力。讨论了沉积温度对薄膜残余应力的影响。结果发现，HfO2薄膜的残余应力均为张应力，应力值随沉积温度的升高先增大后减小，在280℃左右出现极大值。对样品进行了XRD测试，从微观结构上对实验结果进行了分析，发现微结构演变引起的内应力变化是引起薄膜残余应力改变的主要因素，HfO2薄膜在所选沉积温度60～350℃内出现了晶态转变，堆积密度随温度升高而增大。

薄膜厚度对HfO2薄膜残余应力的影响

HfO2薄膜是用电子束蒸发方法制备的，利用ZYGO干涉仪测量了基片镀膜前后曲率半径的变化，计算了薄膜应力。对样品进行了XRD测试，讨论了膜厚对薄膜残余应力的影响。结果发现不同厚度HfO2薄膜的残余应力均为张应力，应力值随薄膜厚度的增加而减小，当薄膜厚度达到一定值后，应力值趋于稳定。从微观结构变化对实验结果进行了分析，发现微结构演变引起的本征应力变化是引起薄膜残余应力改变的主要因素。

SiO_2薄膜折射率的准确拟合分析

精确的光学常数对于设计和制备高品质的光学薄膜非常重要,尤其是那些光学性能对折射率变化敏感的薄膜。SiO_2是一种常用的低折射率材料,因与常用基底折射率相近使其准确拟合有一定难度。实验通过离子束溅射制备了SiO_2单层膜。考虑测量时的误差和基底折射率的影响,采用透射率包络和反射率包络得到了SiO_2的折射率,并用所得折射率进行反演来对这两种途径在实际测量拟合过程中的准确性进行比对。分析表明,剩余反射率在实际的测量过程中误差更小,直接用测量镀膜前后基片的剩余反射率值可以更简便更准确地得到SiO_2的折射率,能达到10~(-2)的精度。