Дисертації з теми "Lateral source"

A popular electrical exploration method is controlled source audiomagnetotellurics (CSAMT). Although the CSAMT method has had practical sucess, the theory used in CSAMT interpretation remains limited. The controlled source in CSAMT is a grounded electric dipole placed as far as is practical from the survey area. When the source-receiver separation is large enough, source fields can be adequately modeled by a single plane wave and conventional magnetotelluric interpretation methods can be used. Quite often however, data collected at lower frequencies can not be interpreted with magnetotelluric algorithms. If the electrical distance between source and receiver is too small, mathematical models must explicitly include the dipolar nature of the source fields to accurately model the data. Models which include a dipole source are limited. Most interpretation of CSAMT data is based upon horizontally layered models. In this dissertation, the model of plane layers excited by an arbitrary source is extended to include the effects of lateral changes in layer thickness. The method developed for modeling undulating interfaces mimics nature more accurately than plane-layered models, while preserving the utility of quick evaluation and insight lost to more general numerical methods. While the main emphasis here is on CSAMT applications, the theoretical results are valid for any source type. Topographic effects are modeled for plane wave and magnetic dipole sources. The effects of relief in basement topography on the response of a horizontal loop-loop system are computed and compared with scale-model measurements. CSAMT field data which can not be adequately explained by plane-layered models are analyzed using models with variable layer thickness. The correctness of the theoretical development is verified by comparison with physical measurements. The utility of the theory is illustrated by application to problems which can not be explained by plane-layered models.

Single-photon sources are essential building blocks in quantum photonic networks, where quantum-mechanical properties of photons are utilised to achieve quantum technologies such as quantum cryptography and quantum computing. In this thesis, a single-photon source driven by a surface acoustic wave (SAW) is developed and characterised. This single-photon source is based on a SAW-driven lateral n-i-p junction in a GaAs quantum-well structure. On this device, the lateral n-i-p junction is formed by gate-induced electrons and holes in two adjacent regions. The SAW potential minima create dynamic quantum dots in a 1D channel between these two regions, and are able to transport single electrons to the region of holes along the channel. Single-photon emission can therefore be generated as these electrons consecutively recombine with holes. After characterisation and optimisation in four batches of devices, clear SAW-driven charge transport and the corresponding electroluminescence (EL) can be observed on an optimised SAW-driven n-i-p junction. Time-resolved measurements have been carried out to study the dynamics of SAW-driven electrons. Time-resolved EL signals indicate that a packet of electrons is transported to the region of holes in each SAW minimum. In addition, the carrier lifetime of SAW-driven electrons in the region of holes is shown to be $\sim 100$ ps, which is much shorter than the SAW period of $860$ ps. Hence, it is promising to observe single-photon emission in the optimised device. In order to test single-photon emission, a Hanbury Brown-Twiss experimental setup has been employed to record an autocorrelation histogram of the SAW-driven EL signal at the single-electron regime. Suppression of autocorrelation coincidences at time delay $\Delta t = 0$ is evidence of photon antibunching. By fitting theoretical functions describing the SAW-driven EL signal, it is found that the second-order correlation function shows $g^{(2)}(0) = 0.39 \pm 0.05$, which is lower than the common criterion for a single-photon source $g^{(2)}(0) < 0.5$. Moreover, theoretical calculation and simulation suggest that, if a constant background signal can be filtered out, $\sim 80 \%$ of the SAW-driven EL is single-photon emission.

Les sites pollués (sol ou eaux souterraines) représentent un potentiel de risque pour la santé humaine et l’environnement. Il existe des outils d’aide à la gestion, en complément des mesures in-situ, qui permettent d’estimer rapidement et à moindre coût les risques sanitaires associés à l’exposition des polluants gazeux du sol dans les espaces intérieurs afin d’établir des mesures de prévention et/ou correction. Cependant, et malgré leur intérêt, il a été montré dans la littérature qu’il existe des différences importantes entre les concentrations intérieures mesurées et les estimations des outils existants. Ces incertitudes reposent principalement sur trois aspects : une mauvaise caractérisation du site, une modélisation incomplète des voies et mécanismes de transfert, ou bien du fait de négliger l’influence de certains paramètres sur le transfert. Par exemple, le fait de négliger la latéralité de la source reste une explication plausible des limites des modèles classiques de transfert. Les auteurs conviennent que la migration latérale joue un rôle important sur l’atténuation de la concentration intérieure en polluant, contrairement aux scénarios de source homogène ou continue, où les vapeurs migrent uniquement de manière verticale vers le bâtiment. Ainsi, lorsque la source est latéralement décalée vis-à-vis du bâtiment, les vapeurs vont migrer préférentiellement vers l’atmosphère et moins vers le bâtiment générant une atténuation plus importante de la concentration intérieure. Dans ce contexte, l’objectif principal de ces travaux de thèse est la contribution à l’amélioration des outils d’aide à la gestion afin d’élargir leur plage d’application. Pour ce faire, des nouveaux modèles ont été développés permettant de tenir compte de la latéralité de la source dans l’estimation de la concentration intérieure en polluant. Le développement de ces modèles est réalisé à partir de l’expérimentation numérique et l’analyse adimensionnelle sur la base des outils existants (modèles semi-empiriques construits en considérant une source continue). La combinaison de ces deux approches permet d’une part, de garder la capacité des modèles source continue de tenir compte des propriétés physiques du sol (perméabilité, coefficient de diffusion, …) et des caractéristiques du bâtiment (typologie de soubassement, dépression, volume, …), et d’une autre part, de mieux préciser la position de la source dans le sol en considérant l’influence de sa latéralité dans les estimations. Ces nouveaux modèles ont été issus d’une analyse comparative permettant de vérifier la cohérence et la précision des estimations vis-à-vis d’un modèle numérique (CFD), de données expérimentales et de modèles existants dans la littérature. Finalement, ces expressions ont été intégrées dans un code de ventilation (MATHIS-QAI) permettant de mieux préciser les caractéristiques des environnements intérieurs (système de ventilation, perméabilité à l’air de l’enveloppe, volume du bâtiment, …) et de réaliser des estimations des niveaux de concentration en fonction des variations temporelles (vitesse du vent, température extérieure, …) au cours du temps. À partir d’une étude paramétrique il a été montré que malgré l’impact non-négligeable des caractéristiques du bâtiment, l’influence de la latéralité de la source sur l’atténuation de la concentration intérieure en polluant reste prédominante (atténuation de plusieurs ordres de grandeur quand la source est décalée latéralement du bâtiment en comparaison à une source continue). Cependant, préciser les caractéristiques du bâtiment (soubassement, système de ventilation, perméabilité à l’air de l’enveloppe,…), ainsi que les conditions météorologiques uniques de chaque projet de construction, permet d’augmenter la précision des estimations en évitant la mise en œuvre de solutions extrêmes ou bien encore, de mesures inadaptées<br>Polluted sites and most precisely vapor intrusion represents a potential risk for human health and its environment. Various screening-level and analytical models have been proposed in order to evaluate vapor intrusion and provide assessment tools for exposure risk. However, some in situ investigations show significant differences between predicted and measured indoor concentrations leading eventually to misleading conclusions and inappropriate solution implementations. These uncertainties are mainly associated with a poor characterization of the site, an incomplete modeling of transfer pathways and mechanisms, or by neglecting certain influencing parameters on this transfer. For example, ignoring the lateral source/building separation may serve as possible explanation of the uncertainties presented by the conventional models based on a homogeneous source distribution assumption. The authors agree that lateral migration plays an important role in the attenuation of the indoor concentration. In homogeneous or continuous source scenarios vapors may migrate mainly vertically towards the building. However, lateral source may promote lateral migration to the atmosphere and less into the building generating a greater attenuation of the indoor concentration. In this context, the main objective of this thesis work is to contribute to the improvement of the assessment and management risk tools in order to improve the accuracy of their estimations and increase their range of application. To do this, new vapor intrusion models are developed considering the lateral source/building separation. These models are built on a numerical experimentation and dimensionless analysis based on existing models (semi-empirical models considering a homogeneous source distribution). The combination of these two approaches allows, on the one hand, to maintain the aptitude of the existing models to consider the physical properties of the soil (permeability, diffusion coefficient, …) and the characteristics of the building (type of construction, building depression, volume,…), and on the other hand, to better precise the position of the source in the soil taking into account the influence of the lateral source/building separation in the estimations. From a comparative analysis, the accuracy of these new expressions is verified comparing to the proposed numerical model (CFD), experimental data and existing models in the literature. Finally, the proposed expressions were coupled with a ventilation code (MATHIS-QAI) allowing to better specify indoor characteristics (ventilation system, air permeability of the envelope, volume of the building, …) and estimate indoor air concentration levels as a function of environmental variations (wind speed, outside temperature, …) over time. From a parametric study it was shown that despite the significant impact of the characteristics of the building, the influence of the lateral source/building separation remains predominant on the attenuation of the indoor concentration (attenuation of several orders of magnitude when the source is laterally offset of the building compared to a homogeneous source). However, specifying the characteristics of the building (construction type, ventilation system, air permeability, …) and weather conditions may increase the accuracy of the estimation avoiding the implementation of extreme solutions or insufficient actions

&nbsp;Acorus calamus L. је вишегодишња зељаста биљка влажних подручја, чија су лековита својства одавно позната. Дрогу иђирота чини ризом који се употребљава као чај, прах, сок, гел, уље или крема. Због прекомерне експлоатације и високог степена угрожености ове дивље лековите биљне врсте у Србији је уведена забрана његовог сакупљања из природе. Да би се задовољила повећана потражња од стране индустрија које га користе, гајење иђирота се намеће као једно од најпрагматичнијих решења. За потребе утврђивање начина гајења, прво се приступило истраживању услова успевања и варирање својстава иђирота са пет локалитета природних станишта: Обедска бара, Делиблатска пешчара, Засавица, Раковац и Дубовац. Потом је заснован двогодишњи пољски оглед на ком је испитиван утицај примене основних агротехничких мера, односно густине садње и ђубрења азотом, на својства биљака. Оглед је постављен 2013. године, на подручју засеока Ћумуране у насељу Рипањ. Код биљка са природних станишта и са огледног поља праћена су следећа својства: висина биљака, дужина ризома, број и дужина бочних грана на ризому, број пупољака на ризому и бочним гранама, број нодуса на ризому и бочним гранама, проценат суве материје ризома, маса свежег и сувог ризома, однос масе свежег и сувог ризома. Код гајених биљака праћен је и принос свежег и сувог ризома. Код свих узорака иђирота испитиван је садржај и састав етарског уља из ризома. Идентификоване су, такође, најзаступљеније компоненте и утврђен је садржаја &beta;-азарона. За утврђивање нивоа плоидије популација из Србије рађено је пребројавање хромозома. Испитивањем биљака са природних станишта, констатовано је да на већину њихових морфолошких одлика, локалитет није имао утицаја. Утицај локалитета испољен је само на: број нодуса на ризому (највећи је био у Дубовцу - 58 а најмањи на Обедској бари - 15), проценат суве материје (највећи је био у Раковцу - 50,2%, а најмањи на Делиблатској пешчари - 37,9%) и однос масе свежег и сувог ризома (највећи је на Делиблатској пешчари - 2,64, а најмањи у Раковцу - 1,90). У пољском огледу добијени су следећи резултати: висина биљака је била највећа при најмањој густини усева (35000 биљака/ha) и најмањој дози азота (60 kg/ha) дужина ризома, као и број и дужина њихових бочних грана су највећи при средњој густини садње (48000 биљака/ha) и најмањој дози азота; број пупољака на ризому има највећу вредност при најмањој густини садње и најмањој дози азота. Број пупољака прве бочне гране је био највећи на највећој густини садње (62000 биљака/ha) и при најмањој дози азота, а број нодуса на ризому није зависио од густине садње, као ни од доза ђубрења. Највећа маса свежег и сувог ризома остварена је при средњој густини садње (48000 биљака/ha) и најмањој (62000 биљака/ha) дози азота, док је проценат суве материје био највећи при највећој густини и најмањој дози азота (60 kg/ha). На однос масе свежег и сувог ризома утицај нису имали ни ђубрење ни густинасадње усева, док су приноси свежег и сувог ризома били највећи на највећој густини садње и при најмањој дози азота. Садржај етарског уља иђирота из природе није показао значајна одступања између локалитета. Садржај етарског уља ризома гајеног иђирота опадао је са порастом густине садње, док различите дозе азота нису утицале на ову особину. Максимална концентрација &beta;-азарона у етарском уљу иђирота са природног станишта била је 17,07 % (локалитет Раковац), док је највећи садржај &beta;-азарона код гајеног иђирота био 21,41 %. Повећана концентрација &beta;- азарона се објашњава интензивнијим метаболизмом азота, због повећане количине из ђубрива. обијени резултати потврђују да је концентрација &beta;-азарона у етарском уљу ризома иђирота слична резултатима из других земаља Европе. Бројањем хромозома утврђено је да иђирот припада триплоидном, европском варијетету Аcоrus cаlаmus vаr. cаlаmus.<br>&nbsp;Acorus calamus L. je višegodišnja zeljasta biljka vlažnih područja, čija su lekovita svojstva odavno poznata. Drogu iđirota čini rizom koji se upotrebljava kao čaj, prah, sok, gel, ulje ili krema. Zbog prekomerne eksploatacije i visokog stepena ugroženosti ove divlje lekovite biljne vrste u Srbiji je uvedena zabrana njegovog sakupljanja iz prirode. Da bi se zadovoljila povećana potražnja od strane industrija koje ga koriste, gajenje iđirota se nameće kao jedno od najpragmatičnijih rešenja. Za potrebe utvrđivanje načina gajenja, prvo se pristupilo istraživanju uslova uspevanja i variranje svojstava iđirota sa pet lokaliteta prirodnih staništa: Obedska bara, Deliblatska peščara, Zasavica, Rakovac i Dubovac. Potom je zasnovan dvogodišnji poljski ogled na kom je ispitivan uticaj primene osnovnih agrotehničkih mera, odnosno gustine sadnje i đubrenja azotom, na svojstva biljaka. Ogled je postavljen 2013. godine, na području zaseoka Ćumurane u naselju Ripanj. Kod biljka sa prirodnih staništa i sa oglednog polja praćena su sledeća svojstva: visina biljaka, dužina rizoma, broj i dužina bočnih grana na rizomu, broj pupoljaka na rizomu i bočnim granama, broj nodusa na rizomu i bočnim granama, procenat suve materije rizoma, masa svežeg i suvog rizoma, odnos mase svežeg i suvog rizoma. Kod gajenih biljaka praćen je i prinos svežeg i suvog rizoma. Kod svih uzoraka iđirota ispitivan je sadržaj i sastav etarskog ulja iz rizoma. Identifikovane su, takođe, najzastupljenije komponente i utvrđen je sadržaja &beta;-azarona. Za utvrđivanje nivoa ploidije populacija iz Srbije rađeno je prebrojavanje hromozoma. Ispitivanjem biljaka sa prirodnih staništa, konstatovano je da na većinu njihovih morfoloških odlika, lokalitet nije imao uticaja. Uticaj lokaliteta ispoljen je samo na: broj nodusa na rizomu (najveći je bio u Dubovcu - 58 a najmanji na Obedskoj bari - 15), procenat suve materije (najveći je bio u Rakovcu - 50,2%, a najmanji na Deliblatskoj peščari - 37,9%) i odnos mase svežeg i suvog rizoma (najveći je na Deliblatskoj peščari - 2,64, a najmanji u Rakovcu - 1,90). U poljskom ogledu dobijeni su sledeći rezultati: visina biljaka je bila najveća pri najmanjoj gustini useva (35000 biljaka/ha) i najmanjoj dozi azota (60 kg/ha) dužina rizoma, kao i broj i dužina njihovih bočnih grana su najveći pri srednjoj gustini sadnje (48000 biljaka/ha) i najmanjoj dozi azota; broj pupoljaka na rizomu ima najveću vrednost pri najmanjoj gustini sadnje i najmanjoj dozi azota. Broj pupoljaka prve bočne grane je bio najveći na najvećoj gustini sadnje (62000 biljaka/ha) i pri najmanjoj dozi azota, a broj nodusa na rizomu nije zavisio od gustine sadnje, kao ni od doza đubrenja. Najveća masa svežeg i suvog rizoma ostvarena je pri srednjoj gustini sadnje (48000 biljaka/ha) i najmanjoj (62000 biljaka/ha) dozi azota, dok je procenat suve materije bio najveći pri najvećoj gustini i najmanjoj dozi azota (60 kg/ha). Na odnos mase svežeg i suvog rizoma uticaj nisu imali ni đubrenje ni gustinasadnje useva, dok su prinosi svežeg i suvog rizoma bili najveći na najvećoj gustini sadnje i pri najmanjoj dozi azota. Sadržaj etarskog ulja iđirota iz prirode nije pokazao značajna odstupanja između lokaliteta. Sadržaj etarskog ulja rizoma gajenog iđirota opadao je sa porastom gustine sadnje, dok različite doze azota nisu uticale na ovu osobinu. Maksimalna koncentracija &beta;-azarona u etarskom ulju iđirota sa prirodnog staništa bila je 17,07 % (lokalitet Rakovac), dok je najveći sadržaj &beta;-azarona kod gajenog iđirota bio 21,41 %. Povećana koncentracija &beta;- azarona se objašnjava intenzivnijim metabolizmom azota, zbog povećane količine iz đubriva. obijeni rezultati potvrđuju da je koncentracija &beta;-azarona u etarskom ulju rizoma iđirota slična rezultatima iz drugih zemalja Evrope. Brojanjem hromozoma utvrđeno je da iđirot pripada triploidnom, evropskom varijetetu Acorus calamus var. calamus.<br>&nbsp;&nbsp;Acorus calamus L. is a perennial herbaceous plant found in wet areas, whose medicinal properties have been long known. The drug of sweet flag is made from the rhizome which is used as tea, powder, juice, gel, oil or cream. Because of overexploitation and the high degree of endangerment of this wild medicinal plant species in Serbia collecting was banned. In order to meet the increased demand for this plant by industries that use it, cultivation of sweet flag is emerging as one of the most pragmatic solutions. For the purpose of determining the method for plant growth, the first approach was studying the growing conditions and the variation of properties of sweet flag from five natural habitats: Obedska pond, Deliblato Sands, Zasavica, Rakovac and Dubovac.After that a two-year field experiment was designed in which the effect of application of basic agrotechnical measures, i.e., different planting density and doses of nitrogen fertilization on the plants, was studied. The experiment was set up in 2013, in the area of the hamlet Ćumurana in the settlement Ripanj. In plants from natural habitats and the experiment, the following properties were measured: the height of the plants, the length of the rhizome, the number and length of lateral branches on the rhizome, the number of buds on the rhizome and lateral branches, the number of leaf scars on the rhizome and lateral branches, the percentage of dry matter of the rhizome, the mass of the fresh and the dry rhizome, the ratio of mass between the fresh and the dry rhizome. In the experiment with cultivated plants, the yield of the fresh and the dry rhizome was also measured. For all samples of sweet flag, the content and composition of the essential oil from the rhizome was measured. Also, the main components were identified and the content of the &beta;-asarone was determined. For the purposes of identifying the ploidy, i.e., which varieties the sweet flag from Serbia belongs to, counting of chromosomes was done. By examining plants from natural habitats, it was noted that the site had no influence on the majority of morphological features of plants. The impact of the sites was manifested only in the number of leaf scars (the largest number was in Dubovac-58,0 and the smallest number was in Obedska pond-15,0), percentage of dry matter (the largest percentage was in Rakovac-50.2% and the smallest percentage was in Deliblato Sands-37,9 %) and the ratio of mass between a fresh and a dry rhizome (the largest ratio was in Deliblato Sands-2,6 and the smallest ratio was in Rakovac-2,2). In the field experiment, the following results were obtained: the height of the plants is the largest at the lowest planting density (35000 plants/ha) and lowest nitrogen dosage (60 kg/ha), the length of the rhizome, as well as number and length of lateral branches of the rhizome are largest at medium planting density (48000 plants/ha) and the lowest dosage of nitrogen (60 kg/ha), the number of buds on the rhizome has the highest value at the lowest planting density (35000 plants/ha) and the lowest dosage of nitrogen (60 kg/ha). The number of buds on the first lateral branch was largest at the largest planting density (62000 plants/ha) and the lowest dosage of nitrog (60 kg/ha) , and the number of leaf scars on the rhizome doesnot depend on the planting dosage or the dosage of fertilizing. The largest mass of the fresh and the dry rhizome is accomplished at medium planting density and the lowest dosage of nitrogen, while the percentage of dry matter was largest at the largest density and the lowest dosage of nitrogen. The fertilization and the crop density had no effect on the ratio of mass between the fresh and the dry rhizome, while the yields of the fresh and the dry rhizome were largest at the largest planting density and the lowest dosage of nitrogen. The content of essential oils of natural sweet flag showed no significant discrepancies between the sites. The content of essential oil of cultivated sweet flag rhizomes declined with the increase of planting density, while different nitrogen doses had no effect on this property. The maximum concentration of &beta;-asarones in the essential oil of natural sweet flag was 17,07 % (Rakovac), while the largest content of &beta;-asarones in cultivated sweet flag was 21,41 %. The larger concentration of &beta;-asarones is explained through increased nitrogen metabolism, because of increased amounts of fertilizer. The results confirm that the concentration of &beta;-asarone in the essential oil of sweet flag rhizomes is similar to the concentration of essential oil of sweet flag rhizomes from Europe. By counting the chromosomes it is established that sweet flag belongs to the triploid, European variety of Аcоrus cаlаmus vаr. cаlаmus.

BACKGROUND:Laterally transferred genes have often been identified on the basis of compositional features that distinguish them from ancestral genes in the genome. These genes are usually A+T-rich, arguing either that there is a bias towards acquiring genes from donor organisms having low G+C contents or that genes acquired from organisms of similar genomic base compositions go undetected in these analyses.RESULTS:By examining the genome contents of closely related, fully sequenced bacteria, we uncovered genes confined to a single genome and examined the sequence features of these acquired genes. The analysis shows that few transfer events are overlooked by compositional analyses. Most observed lateral gene transfers do not correspond to free exchange of regular genes among bacterial genomes, but more probably represent the constituents of phages or other selfish elements.CONCLUSIONS:Although bacteria tend to acquire large amounts of DNA, the origin of these genes remains obscure. We have shown that contrary to what is often supposed, their composition cannot be explained by a previous genomic context. In contrast, these genes fit the description of recently described genes in lambdoid phages, named 'morons'. Therefore, results from genome content and compositional approaches to detect lateral transfers should not be cited as evidence for genetic exchange between distantly related bacteria.

The region between the infrared and microwave region in the electromagnetic spectrum, the Terahertz (THz) gap, provides an exciting opportunity for future wireless communications as this band has been under utilised. This doctoral work takes a two-pronged approach into closing the THz gap with low-dimensional materials. The first attempt addresses the need for a compact THz source that can operate at room temperature. The second approach addresses the need to build optical elements such as filters and modulators in the THz spectrum. Terahertz quantum cascade lasers (THz QCLs) are one of the most compact, powerful sources of coherent radiation that bridge the terahertz gap. However, their cryogenic requirements for operation limit the scope of the applications. This is because of the electron-electron scattering and heating of the 2-dimensional free electron gas which leads to significant optical phonon scattering of the hot electrons. Theoretical studies in laterally confined QCL structures have predicted enhanced lifetime of the upper state through suppression of the non-radiative intersubband relaxation of carriers, which leads to lower threshold, and higher temperature performance. Lithographically defined vertical nanopillar arrays with electrostatic radius less than tens of nm offer a possible route to achieve lateral confinement, which can be integrated into QCL structures. A typical gain medium in a QCL consists of at least 100 repeat periods, with a thickness of 6-14 micron. For practical implementation of the top-down approach, restrictions are imposed by aspect ratios that can be achieved in present dry-etching systems. Typically, for sub-200 nm radius pillars, the thickness ranges from 1-3.5 micron. It is therefore necessary to work with THz QCLs based on 3-4 quantum well active regions, so as to maximise the number of repeat periods (hence gain) within an ultra-thin active region. After an introductory chapter, Chapter 2 presents a theoretical treatise on the realistic electrostatic potential in a lithographically defined nanopillar by scaling from a single quantum well (resonant tunnelling diode) to a THz QCL. Chapter 2 also discusses, the effect of lateral confinement on the intersubband states and the plasmonic mode in a THz QCL. One of the key experimental challenges in scaling down from QCLs to quantum-dot cascade lasers is the electrical injection into the nanopillars. This involves insulation and planarisation of the high aspect-ratio nanopillar arrays. Furthermore, the choice of the planarising layer is critical since it determines the loss of any optical mode. This experimental challenge is solved in Chapter 3. Chapter 4 presents the electro-optic performance of low-repeat period QCLs with an active region thickness that is less than 3.5 micron. Another topic of recent interest in the THz optics community is plasmonics in graphene. This is because the bound electromagnetic modes (plasmons) are tightly confined to the surface and can also be tuned with carrier concentration. Plasmonic resonance at terahertz frequencies can be achieved by gating graphene grown via chemical vapour deposition (CVD) to a high carrier concentration. THz time domain spectroscopy of such gated monolayer graphene shows resonance features around 1.6 THz superimposed on the Drude-like frequency response of graphene which may be related to the inherent poly-crystallinity of CVD graphene. Chapter 5 discusses these results, as an understanding of these features is necessary for the development of future THz optical elements based on CVD graphene. Chapter 5 finally describes how the gate tunability of THz transmission through graphene can be exploited to indirectly modulate a THz QCL. Chapter 6 presents ideas from this doctoral work, which can be developed in future to address the issues of enhanced temperature performance of THz QCLs and to realise realistic THz devices based on graphene.

Carbon Dioxide (CO2) emissions from streams are a significant component of the global carbon cycle.Terrestrial export of CO2 through runoff is increasingly recognized as a major source of CO2 in boreal headwater streams. However, the spatial and temporal distribution of soil water CO2 within theterrestrial landscape remains poorly quantified, contributing to large uncertainties about the origin of CO2 in headwater streams. The riparian zone (i.e. the area with fine sediments and organic rich soils closest to the stream) is accepted as a main contributor of organic carbon to streams, but its importanceas a source of CO2 is less evident. Here I evaluate the riparian zone as a main source by quantifying the contribution of lateral CO2 export from the riparian and hillslope zones to a headwater stream in a Swedish boreal catchment. Hourly measurements of CO2 concentration, conductivity, soil temperature and water table levels were taken in the riparian zone and the hillslope from June 2014 to October 2015. The riparian zone accounted for 58-89 % (August 2014 and March respectively) of the total terrestrial CO2 export from the slope to the stream. The hillslope, in turn, became a progressively larger source of CO2 to the stream during high flow events. To identify the drivers behind these zone-dependent and seasonal patterns in CO2 export, the CO2 production dissolved in the groundwater (groundwater- absorbed carbon) was estimated by taking the temporarily stored CO2  into account. The highest groundwater-absorbed carbon was observed during April and May (5.0 and 7.1 g C-CO2 m-2 month-1 respectively) which is the period with the highest discharge due to snow melt and the initiation of spring production. As such, conventional methods (gas chambers and the gradient method) may underestimate the soil respiration up to 50% during periods of high flow, as they exclude groundwater-absorbed carbon. CO2 consumption was observed in September 2014 and October 2015 (-0.2 and -0.7 g C-CO2 m-2 month-1 respectively) and may be explained by a major amount of the soil respiration being emitted instead of diluted in the groundwater during periods of low groundwater levels. It can be concludedthat, regardless of season, the riparian zone is a major source of CO2 to the headwater stream.<br>En signifikant mängd koldioxid (CO2) är lagrad i skog och marken. Marken i barrskogsregionernaförvarar en signifikant mängd CO2 där det partiella trycket av CO2 varierar mellan ~10 000 – 50 000 ppm i jämförelse med atmosfären (400 ppm). Mättnaden av CO2 gör att mycket avdunstar tillbaka till atmosfären. Dock absorberas en del CO2 av grundvattnet; vilket resulterar i en naturlig transport av CO2 vidare till ytvattnen där det kapillära nätverket av bäckar är största recipienten. Det är fortfarande oklart hur transporten av CO2 är distribuerad i ett vattenavrinningsområde vilket medför brister i förståelsen av en viktig processväg som kan komma att spela en större roll i framtidens kolkretslopp på grund av den globala uppvärmningen. Därför är en kvantifiering av olika områdens bidrag av CO2 till bäckarna nödvändig. Två betydande zoner i ett vattenavrinningsområde som troligen bidrar olika är: the riparian zone som är närmast bäcken och består av fina sediment med hög organisk halt och, the hillslope som är resterande område och består av grovkorniga jordar med låg organisk halt. Den förstnämnda misstänks transportera mer CO2 via grundvattnet på grund av dess närhet till bäcken, höga halter av CO2 och höga vattenmättnad men detta är ännu inte verifierat. Jag evaluerar the riparian zone som en viktig källa till CO2 i ett vattenavrinningsområde genom att kvantifiera transporten av CO2 från de två zonerna. För att förklara varför transporten varierar presenterar jag en ny modell (GVR) som beräknar den månatliga fluktuationen av den del av CO2-produktionen som absorberas i grundvattnet i the riparian zone. Mätningar av data utfördes i Västrabäcken, ett mindre vattenavrinningsområde i ett större vid namn Krycklan, i norra Sverige. En transekt av tre mätstationer (i bäcken, the riparian zone och the hillslope) installerades i den förmodade grundvattenströmningsriktningen. Resultaten visar på en hög produktion av CO2 under vårfloden (maj) då en hög grundvattenyta troligen absorberar en signifikant mängd CO2. Detta kan betyda att jordrespiration under våren underskattas då dagens mätmetoder är begränsade till mätningar i jorden av CO2 ovan grundvattenytan. Fortsatta studier rekommenderas där GVR-modellen och andra mätmetoder utförs samtidigt för att vidare utröna den kvantitativa underskattningen under perioder med hög grundvattenyta (speciellt under våren). Bidraget från the riparian zone till den totala laterala transporten av CO2 till bäcken under ett år varierar mellan 58-89 % och det månatliga transportmönstret kunde förklaras med resultaten från GVR-modellen. Resultaten verifierar att oberoende av säsong så är the riparian zone den huvudsakliga laterala koltransporten från landvegetationen; medan the hillslope procentuellt bidrar med mer CO2 under höga grundvattenflöden.

La sclérose latérale amyotrophique (SLA) est une pathologie neurodégénérative caractérisée par une perte progressive des neurones moteurs et par de l’atrophie musculaire menant à de la paralysie. Bien que plusieurs mécanismes pathologiques aient été élucidés, la SLA reste un mystère médical puisqu’il n’existe toujours pas de traitement efficace. Nous avons développé deux stratégies dans le but de traiter la SLA. La première stratégie fut de cibler la protéine SOD1 mal repliée (chapitre 2) et la deuxième consistait à traiter la neuroinflammation présente dans la maladie (chapitre 3). Dans le chapitre 2, nous avons tenté de diminuer le niveau de la protéine SOD1 mal repliée présent dans le système nerveux de souris transgéniques développant un phénotype de SLA. Pour ce faire, nous avons testé une nouvelle approche thérapeutique basée sur l’utilisation d’ « adeno-associated virus (AAV) ». Ce virus contient une séquence d’ADN qui encode pour un anticorps à chaîne unique variable (scFv). Cet anticorps est composé d’une des deux chaînes légères et lourdes de l’anticorps D3H5 ciblant de manière spécifique la protéine SOD1 mal repliée. Une injection intra-thécale unique de l’AAV encodant l’anticorps à chaîne unique dans des souris SOD1G93A repousse le début de la maladie et augmente la survie des souris de 28%. Nous avons démontré que la Withaferin A (WA), un inhibiteur du facteur NF-κB, diminuait le phénotype neurologique retrouvé chez le modèle de souris transgénique TDP-43. Donc, nous avons testé la Withaferin A sur deux autres lignées de souris transgéniques exprimant des mutations dans la protéine SOD1, soit la lignée de souris SOD1G93A et la lignée de souris SOD1G37R. L’effet bénéfique de la WA chez les souris SOD1G93A était accompagné d’un soulagement de la neuroinflammation, d’une diminution du niveau de protéine SOD1 mal repliée dans la moelle épinière et d’une baisse de la mortalité des neurones moteurs. En considérant ces résultats, l’utilisation d’AAV encodant un anticorps à chaîne unique contre la protéine SOD1 mal repliée ainsi que l’utilisation de Withaferin A devrait toutes les deux être considérées comme des approches pour traiter la SLA.<br>Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disease associated with motor neuron degeneration, muscle atrophy and paralysis. Although numerous pathological mechanisms have been elucidated, ALS still remains a medical mystery in the absence of any effective therapy. Riluzole is the only therapeutic drug approved for ALS with regard to prolonging survival. Here, we have developed two strategies for treatment of ALS, first targeting the misfolded SOD1 (chapter 2) and other targeting neuroinflammation (chapter 3). In chapter 2, we aimed to reduce the level of misfolded SOD1 species in the nervous system. We tested a novel therapeutic approach based on adeno-associated virus (AAV)-mediated tonic expression of a DNA construct encoding a secretable single chain fragment variable (scFv) antibody composed of the variable heavy and light chain regions of a monoclonal antibody (D3H5) binding specifically to misfolded SOD1. A single intrathecal injection of the adeno-associated virus encoding the single chain antibody in SOD1G93A mice delayed disease onset and extended the life span by up to 28%, in direct correlation with scFv titers in the spinal cord. Our second treatment strategy which is aimed to target neuroinflammation is based on previous reports from our lab where it has been shown that Withaferin A (WA), an inhibitor of NF-κB activity was efficient in reducing disease phenotype in TDP-43 transgenic mouse model of ALS. We tested WA in mice from two transgenic lines expressing different ALS-linked SOD1 mutations, SOD1G93A and SOD1G37R. The beneficial effects of WA in SOD1G93A mice model was accompanied by alleviation of neuroinflammation, decrease in level of misfolded SOD1 species in spinal cord, a reduction in loss of motor neurons, resulting in delayed disease progression and mortality. Based on these evidences, AAV encoding a secretable scFv against misfolded SOD1 and WA should be considered as a potential treatment for ALS.

L’utilisation du cannabis peut précipiter la schizophrénie, en particulier chez les individus qui présentent une vulnérabilité génétique aux désordres mentaux. Des recherches humaines et animales indiquent que la neuréguline 1 (Nrg1) est un gène de susceptibilité à la schizophrénie. L’objectif de cette thèse est d’examiner si une modification du gène Nrg1 chez des souris mutante module les effets neuronaux et comportementaux des cannabinoïdes après traitement aiguë et chronique. De plus cette thèse examine les effets d'un pré-traitement au delta9-tétrahydrocannabinol, le principal composant psychotropique du cannabis, sur un modèle de flexibilité cognitive chez la souris<br>Cannabis use may precipitate schizophrenia, especially in individuals who have a genetic vulnerability to the disorder. Human and animal researches indicate that neuregulin 1 (Nrg1) is a susceptibility gene for schizophrenia. This thesis aim at investigating if partial deletion of Nrg1 in mutant mice modulate the neuronal and behavioural effects cannabinoids after acute or chronic treatment. In addition, this thesis examine the effects of a pre-treatment with delta9-tetrahydrocannabinol, the main psychoactive constituent of cannabis, in a model of cognitive flexibility in the mice

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Ocean Engineering, 1989.<br>GRSN 405710<br>Title as it appears in the M.I.T. Graduate List, June 1989: Radiation from directional seismic sources in horizontally stratified media with application to Arctic ice cracking noise.<br>Includes bibliographical references (leaves 168-171).<br>by Jea Soo Kim.<br>Ph.D.

There has been little interest in eighteenth century urban history in England and particularly in the significance of patterns of urban social structure during the transition from a traditional to a modern society. One reason for this is the intractable and fragmentary nature of the sources for this precensus period. In this study three types of source, a town directory, a Parliamentary Poll Book and the city rate and national tax returns for Bristol in 1774/5, were collated using nominal record linkage techniques to give a body of information which covered 80% of the city's heads of household. With the use of this database and various computer techniques occupation, sex, wealth, place of residence and voting allegiance were analysed. The results suggest that a professional or leisured suburban group was by this date well established in distinct areas of the city. The supremacy of the traditional élite, the overseas merchants, was challenged by this group, although the merchants themselves were in part joining the suburban dwellers. Poorer Bristolians still concentrated in dockside parishes and in parts of the city which were becoming increasingly unfashionable and homogeneous as the richer men moved out, though this process was not very far advanced and there was still a degree of mixing in the older city parishes. The economic structure of the city was changing with increased emphasis on services, professions and distribution. This increased disparities in wealth within the city and between the city and its hinterland and gave the ability to the rich to further their isolation from the poor by moving to the suburbs. The 1774 election pointed to the continuing importance of traditional influences (here of religion) In society, but also confirmed suggestions that the professions and distributors were drawing away from the mass of the populace. A revision of previous interpretations of the nature of Bristol society is necessary to accommodate this growing and important group - the emergent middle class. The thesis shows that a comprehensive computer-based study can make usable dubious sources (in particular fiscal records) and use them to revise interpretations of English urban communities at this date.

This thesis takes as its starting point the use of extracts from the works of historical authors who wrote in England in the early to mid twelfth-century. It focuses upon the ways in which their works began to be incorporated into three particular texts in the later twelfth and early thirteenth centuries. Through the medium of individual case studies – De Abbatibus (Abingdon), CCCC 139 (Durham) and The London Collection three elements are explored (i) how mediaeval writers used extracts from the works of others in ways which differed from modern practices with their concerns over charges of plagiarism and unoriginality (ii) how the structural and narrative roles which the use of extracts played within the presentation of these texts (iii) how the application of approaches developed in the twentieth century, which transformed how texts are now analysed, enabled a re-evaluation and re-interpretation of their use of source material with greater sensitivity to their original purposes This analysis casts fresh light upon the how and why these texts were produced and the means by which they fulfilled their purposes and reveals that despite their disparate origins and individual perspectives these three texts share two common features: (i) they follow a common three stage pattern of development (ii) they deal with similar issues: factional insecurities and concerns about the quality of those in power over them – using an historical perspective The analysis also reveals the range of techniques which were at the disposal of the composers of these texts, dispelling any notion that they were either unsophisticated or naïve in their handling of their source materials. Together these texts demonstrate how mediaeval authors used combinations of extracts as a means of responding quickly and flexibly to address particular concerns. Such texts were not regarded as being set in stone but rather as fluid entities which could be recombined at will in order to produce new works as required.

Les modèles murins de la Sclérose Latérale Amyotrophique (SLA) ont permis des avancées dans la compréhension des mécanismes pouvant conduire à la mort sélective et progressive des motoneurones (Mns) mais ils présentent des disparités dans la sévérité et le décours temporel de la maladie. Parmi les hypothèses avancées figurent des modifications des propriétés intrinsèques des motoneurones conduisant à des modifications de l’excitabilité et de l’homéostasie du calcium intracellulaire et à la mort du motoneurone.Nous avons donc étudié les propriétés électrophysiologiques des Mns lombaires de souris SOD1G85R et SOD1G93A, deux modèles à faible nombre de copies du gène humain muté, durant les deux premières semaines postnatales afin d’identifier d’éventuelles anomalies pré-symptomatiques précoces. Nos travaux ont été réalisés sur deux préparations in vitro de moelle entière isolée et de tranches de moelle épinière. Les Mns mutants présentent, sur les deux types de préparations, une altération des propriétés du potentiel d’action se traduisant par un allongement de la durée associée à une diminution des vitesses maximales de dépolarisation et repolarisation et une réduction d’amplitude. Ces altérations apparaissent entre P2-P5 dans les Mns SOD1G85R et entre P6-P10 dans les Mns SOD1G93A et suggèrent une diminution de la densité des canaux sodiques et potassiques associés au potentiel d’action. Nous avons aussi observé sur des tranches de moelle épinière entre P6-P10 que le gain de fréquence des Mns SOD1G85R diminue et celui des SOD1G93A augmente sans aucune modification des densités des courants entrants persistants sodiques et calciques. On note également que, sur tranches de moelle épinière, les Mns SOD1G93A présentent un potentiel de repos diminué. En présence d’une surcharge calcique extracellulaire, les propriétés membranaires des Mns SOD1G85R entre P6-P10 sont moins affectées que celles des Mns témoins. Les effets différentiels de cette surcharge peuvent être dus à des modifications différentes de la dépendance au voltage des canaux voltage-dépendants et/ou à la modulation de certains types de canaux activés par le calcium extracellulaire. Une arborisation dendritique plus ramifiée que celle de Mns témoins, comparable à celle précédemment décrite dans les Mns SOD1G85R, a été observée dans les Mns SOD1G93A à P8-P9 avec des altérations du potentiel d’action citées plus haut et une réduction de la rhéobase. Ces altérations morphologiques et électriques pourraient indiquer des modifications de cinétiques et/ou de densités de canaux sur des sites différents dans ces Mns. Nos travaux montrent donc, d’une part que les mutations SOD1G85R et SOD1G93A induisent dans ces deux modèles murins des altérations des propriétés des Mns lombaires comparables mais décalées dans le temps et d’autre part que certaines altérations semblent être spécifiques à une mutation SOD1 donnée<br>The SOD1 murine models of Amyotrophic Lateral Sclerosis (ALS) allowed major progress in the understanding of mechanisms which could lead to a selective loss of motoneurons (Mns), but these models display differences in the severity and time course of the disease. Changes in intrinsic properties of motoneurons may induce changes in excitability and intracellular calcium homeostasis leading to motoneuron death.Therefore, we studied electrophysiological properties of lumbar Mns from SOD1G85R and SOD1G93A mice, low expressor lines, during the first two postnatal weeks in order to identify possible early presymptomatic abnormalities. Our studies were carried out on two in vitro preparations: the whole isolated spinal cord and acute spinal cord slices. Mutant Mns display, in the two preparations, a modified action potential characterized by an increased duration due to a decrease of the maximal speeds of depolarisation and repolarisation and a reduction of the spike amplitude. These alterations appeared between P2-P5 in SOD1G85R Mns and between P6-P10 in SOD1G93A Mns and suggest a decrease of the density of sodium and potassium channels related to action potential. We also showed on spinal cord slices between P6-P10 that the gain of frequency decreases for SOD1G85R Mns and increases for SOD1G93A Mns without any change in the density of persistent inward sodium or calcium currents in these different mutant Mns. We observed also that the resting membrane potential of SOD1G93A Mns on spinal cord slices is decreased. The membrane properties of SOD1G85R Mns between P6-P10 were less susceptible to changes in presence of an extracellular calcium overload. Differential effects of this extracellular calcium overload on membrane properties of WT and SOD1G85R Mns could be due to different alterations of the potential dependence of voltage-gated channels and/or to the modulation of some types of channels sensitive to extracellular calcium. An over-branching of dendritic arborization, similar to that previously described in SOD1G85R Mns, was observed in SOD1G93A at P8-P9 with the above-mentioned action potential alterations and a weak rheobasic current. These morphogical and electrical changes could indicate together alterations of kinetics and/or density of channels on different sites on these Mns. In conclusion, our work shows on one hand that SOD1G85R and SOD1G93A mutations induce similar alterations of lumbar Mns properties but time-shifted in these two murine models and on the other hand that some alterations seem to be specific to a given SOD1 mutation

L'objectif de ce travail est de contribuer à la compréhension du cycle atmosphérique du soufre en Antarctique par l'utilisation d'un Modèle de Circulation Générale Atmosphérique (MCGA). Les versions "Antarctique" et "soufre" du MCGA LMD-ZT ("Laboratoire de Météorologie Dynamique- Zoom Traceurs") ont été fusionnées pour l'étude à haute résolution Antarctique du cycle du soufre dans les moyennes et hautes latitudes Sud. Une méthode de forçage "latéral" de la circulation atmosphérique antarctique par des analyses du Centre Européen de Prévision Météorologique à Moyen Terme (CEPMMT) a été spécifiquement développée et appliquée. Le modèle a d'abord été évalué. Il représente correctement le cycle saisonnier des espéces soufrées aux sites d'observation en Antarctique. Plusieurs défauts ont cependant été identifiés , discutés, et certains ont été étudiés par des expériences numériques spécifiques. Ceci nous a permis, dans la suite de ce travail, de décrire de manière critique le cycle du soufre en Antarctique à partir des résultats du modèle. Le modèle a été utilisé pour trois applications. La première a été une mise en oeuvre directe du modèle, tel qu'il a été évalué, dans le but d'estimer les distributions spatiales, le cycle saisonnier dans les régions centrales, et le bilan annuel des espèces soufrées en Antarctique. La deuxième se présente sous forme d'études de sensibilité à la formulation des émissions de sulfure de diméthyle (DMS) océanique, que la phase d'évaluation a révélé comme déterminante pour la modélisation du cycle du soufre. Pour la troisième, l'adjoint du modèle de transport (qui, en première approximation, permet de remonter le temps), complété par l'adjoint du module chimique spécifiquement développé pour ce travail, a été mis en oeuvre pour une recherche quantitative de l'origine géographlque et de l'âge des espèces soufrées en Antarctique.

La sclérose latérale amyotrophique est une maladie neurodégénérative associée à un dysfonctionnement métabolique. Des altérations du métabolisme des lipides, décrites chez les patients SLA et les animaux modèles, pourraient participer à la mise en place des premières étapes de la maladie. L’objectif de cette thèse était d’étudier le rôle de la stéaroyl-coenzyme A désaturase 1 (SCD1), une enzyme clé du métabolisme des lipides, dans la SLA. En étudiant le profil d’acides gras périphériques dans un modèle de souris SLA, les souris SOD1m, nous avons vu une diminution de l’activité de la SCD1 dès les stades précoces (subcliniques) de la maladie. Cette diminution pourrait expliquer, à elle seule, les altérations du métabolisme des lipides caractéristiques de la SLA. La répercussion de la perte de l’activité de la SCD1 sur l’axe moteur a été étudiée. Une délétion du gène ou une inhibition pharmacologique de la SCD1 améliore la récupération fonctionnelle après lésion du nerf sciatique chez la souris sauvage. Nous avons cherché à voir si la perte d’activité de la SCD1 trouvée chez les souris SOD1m est un mécanisme de protection mis en place pour lutter contre l’évolution de la SLA. Nous avons traité des souris SOD1m avec un inhibiteur de l’activité de la SCD1. Le traitement a conduit à une augmentation du métabolisme oxydatif, une préservation de l’intégrité neuromusculaire ainsi qu’une amélioration de la survie des motoneurones. Nousconcluons que l’inhibition de la SCD1 représente une cible thérapeutique prometteuse dans la SLA<br>Amyotrophic lateral sclerosis is a neurodegenerative disease, associated with metabolic dysfunction. Alteration of lipid metabolism has been documented in ALS patients and animal models, and could participate to the first pathological steps of the disease. The objective of this thesis was to study the role of stearoyl-CoA desaturase 1 (SCD1), a key enzyme of lipid metabolism, in ALS. By studying the profile of peripheral fatty acids in an animal model of ALS, the SOD1 mice, we found that SCD1 activity was strongly reduced at early (sub-clinical) disease stage, and that this reduction could explain in itself the alteration of lipid metabolism characteristic of ALS. The impact of loss of SCD1 activity for the motor axis was then studied. Genetic deletion or pharmacological inhibition of SCD1 enhanced functional recovery after sciatic nerve injury in mice. Wefurther explored if the loss of SCD1 activity found in SOD1 mice is a protective mechanism elicited in response to ALS. We treated SOD1 mice with an inhibitor of SCD1 activity. The treatment resulted in exacerbated muscular oxidative metabolism,preservation of neuromuscular integrity and enhanced motor neuron survival. We conclude that inhibition of SCD1 represents a promising therapeutic target for ALS

La Sclérose Latérale Amyotrophique (SLA) et la Démence FrontoTemporale (DFT) sont deux maladies neurodégénératives fatales pour lesquelles il n’existe aucun traitement curatif. Ces deux maladies sont liées par un continuum clinique que corroborent des arguments génétiques et histologiques. Aujourd’hui, environ 15% des patients atteints de SLA développent par la suite une DFT et inversement. En 2006, des mutations dans le gène CHMP2B ont été découvertes chez des patients atteints de SLA et de DFT. Douze ans plus tard, les mécanismes pathologiques associés aux mutations de ce gène dans le syndrome SLA-DFT sont encore peu compris. Pour pallier cela, nous avons étudié par plusieurs approches l’impact du mutant CHMP2Bintron5 dans un modèle murin. Les analyses transcriptomiques réalisées au niveau de la moelle épinière lombaire ont mis en lumière une dérégulation de plusieurs processus cellulaires, notamment la réponse inflammatoire et le métabolisme lipidique. De plus, nous avons mis en évidence au niveau neuronal une perturbation très précoce de la macroautophagie, avec un blocage dans l’étape finale de dégradation lysosomale associé à une répression de l’initiation de l’autophagie. Enfin, nous avons montré que l’expression neuronale du mutant CHMP2Bintron5 entraine au niveau périphérique une atrophie musculaire et une altération structurale et fonctionnelle de la jonction neuromusculaire<br>Amyotrophic Lateral Sclerosis (ALS) and FrontoTemporal Dementia (FTD) are two fatal and incurable neurodegenerative diseases. These two diseases share a clinical continuum supported by genetic and histological arguments. Today, about 15% of ALS patients develop later FTD symptoms and reversely. In 2006, mutations in the CHMP2B gene were discovered in ALS-FTD patients. Twelve years later, pathological mechanisms associated with mutations of this gene in the ALS-FTD syndrome are poorly understood. For a better understanding of CHMP2Bintron5-related pathological mechanisms, we studied by different approaches the impact of this mutation in a mouse model. Transcriptomic analysis on lumbar spinal cord highlighted a panel of deregulated cellular pathways, including the inflammatory response and the lipid metabolism. In addition, we showed an early disturbance of macroautophagy, with a blockage of final degradation step associated with a repression of autophagy initiation. Finally, our results show that neuron specific expression of CHMP2Bintron5 leads to muscular atrophy and structural and functional alterations of neuromuscular junctions

碩士<br>國立臺灣科技大學<br>機械工程系<br>99<br>Hot vapors such as oil fumes are often found in a food factory. For example, deep frying which is a common way to cook food generates tremendous oil fumes in a very hot and wide surface oil tank. In order to capture those hot oil fumes, a novel ventilation system which is called the lateral push-pull ventilation hood for hot toxic vapors from a wide area is proposed in this study. The comparison between the conventional hood and the lateral push-pull ventilation hood on capture performance is undertaken using the finite volume method. The hydrogen sulfide (H2S) which is one of toxic substances in oil fumes is chosen as a pollutant in this study. Moreover, the influence of a manikin standing in front of hood on the capture performance of those two hoods is also explored. According to the numerical results without a manikin presence, a primary vortex occurs in the fully open conventional hood and the entire space inside the hood is filled with H2S. Meanwhile, leakage of H2S to the indoor environment is found in the edge between the side walls and the environment. For the fully open lateral push-pull ventilation hood, three dominant flow modes which are called dispersion, transition, and encapsulation are determined by the variation of speed ratio of the pull flow to the push flow. In the encapsulation mode, H2S is captured completely by the push-pull capture flow and the concentration values at selected twelve points are far below the criterion of 10 p.p.m.. When a manikin stands in front of the conventional hood, the lifting airflow comes from the gap between the hood and the manikin carries H2S to the breathing zone of the manikin. In contrast, the capture performance of the lateral push-pull hood is not reduced due to a manikin presence but enhanced in the encapsulation mode. H2S also can be expelled and the concentration values at twelve selected points on the sash opening are close to null. In conclusion, the lateral push-pull ventilation hood can effectively prevent the leakage of H2S and protect an operator from those pollutants.

碩士<br>國立成功大學<br>測量及空間資訊學系<br>107<br>The 2009 Mw 7.6 Padang earthquake has been inferred as an intraslab event because it locates at approximately 80 km depth which is within the oceanic slab with maximum curvature. However, the major trench-parallel-striking normal faulting event usually occurs at this tectonic environment but not the trench-normal-striking reverse event like this 2009 Padang earthquake. To solve this enigma, the coseismic displacements were estimated based on the analysis of daily coordinate time series calculated by GAMIT software from 15 continuous GPS stations along the Sumatra region. The maximum horizontal displacement is approximately 40 mm toward SW at the MSAI station while the maximum vertical displacement reaches 13 mm. The optimized geometry parameters of the source fault were determined by the Markov Chain Monte Carlo using the uniform-slip dislocation model. The optimized strike and dip of fault plane are 72° and 54°, respectively. The coseismic slip distribution was then estimated using the distributed-slip dislocation model in terms of optimized fault geometry. The geodetic moment of 1.35 × 1027 dyne-cm in our best-fit model is equivalent to Mw 7.39. The coseismic slip mainly ranges 40 - 75 km in depth with the maximum slip of 2000 mm. The optimized source fault plane is also comparable to the relocated aftershock distribution. Comparing to the location of interface, our source fault is mainly located at the place above the interface. We therefore proposed that this 2009 Padang earthquake occurred in the deep part of accretionary wedge, but not with the slab. In addition, we also proposed this 2009 event as a lateral ramp event because its strike is normal to the trench, such as the 2010 Jaishian earthquake in Taiwan. Finally, we proposed that a thick-skinned deformation may be also represented in the prism of Sumatra subduction zone.

Organic semiconductors show promise to yield a novel class of bendable electronic devices, and much research efforts have focused on the optimization of these films for device performance. It is well known that the structure of organic films has a large influence over the electronic properties. In particular, the carrier mobility is often highly anisotropic, and domain boundaries have a detrimental effect on charge transport. Therefore the domain structure and lattice orientation are of particular interest. However, little is known about the domain structure of organic films, and techniques to study these properties have only begun to emerge in recent years. In this thesis, we apply two experimental techniques, Grazing-Incidence X-ray Diffraction (GIXD) and Lateral Force Microscopy (LFM), toward studying the lattice and domain structure of tetracene films grown on the silicon(001)-monohydride surface. We describe the necessary steps toward optimizing the sensitivity of these techniques to the domain structure. Results show that the crystalline tetracene films form a layered morphology in which the a-b plane lies parallel to the substrate surface. The film lattice structure is similar to bulk tetracene, and the lattice is confined to two orthogonal orientations, forming a partially-commensurate relationship with the substrate surface lattice along the film 'a' axis. LFM images reveal two types of polycrystalline domains. The first type ("major domains") are tens of microns in size, and are classified by their lattice orientation. They are subdivided into the second type ("sub-domains"), which range from 0.1 to 5um in size, and are argued to represent regions of uniform molecular tilt direction. The GIXD data show that the single-crystal domains which comprise these two larger domain types are anisotropic in size, being up to two times longer along the film 'b' axis than along 'a'. The single-crystal domains range from 0.05 to 0.2um in size, depending on lattice orientation and film thickness. The mathematical basis for these single-crystal domain size calculations is presented. The single-crystal domain sizes are thickness-dependent, and are two orders of magnitude smaller than a typical surface island observed in atomic-force microscopy (AFM) topographs. Substrate steps can also significantly influence the film structure by inducing boundaries in the single-crystal domains and sub-domains, but not in the major domains. This detailed knowledge of the domain structure of organic thin-films may assist in our understanding of the factors which affect charge transport in thin films, and may help to direct research efforts in optimizing the film structure for device performance.<br>Natural Sciences and Engineering Research Council (NSERC), Canadian Foundation for Innovation (CFI), Ontario Innovation Trust (OIT).