Search

We address the line spectral estimation problem with multiple measurement corrupted vectors. Such scenarios appear in many practical applications such as radar, optics, and seismic imaging in which the measurements can be modeled as the sum of a spectrally sparse and a block-sparse signal known as outlier. Our aim is to demix the two components and for this purpose, we design a convex problem whose objective function promotes both of the structures. Using the Positive Trigonometric Polynomials (PTP) theory, we reformulate the dual problem as a Semidefinite Program (SDP). Our theoretical results state that for a fixed number of measurements N and constant number of outliers, up to O(N) spectral lines can be recovered using our SDP problem as long as a minimum frequency separation condition is satisfied. Our simulation results also show that increasing the number of samples per measurement vectors reduces the minimum required frequency separation for successful recovery.

Many studies show that mixed species stands can have higher gross growth, or so-called overyielding, compared with monocultures. However, much less is known about mortality in mixed stands. Knowledge is lacking, for example, of how much of the gross growth is retained in the standing stock and how much is lost due to mortality. Here, we addressed this knowledge gap of mixed stand dynamics by evaluating 23 middle-aged, unthinned triplets of monospecific and mixed plots of Scots pine (Pinus sylvestris L.) and European beech (Fagus sylvatica L.) repeatedly surveyed over 6-8 years throughout Europe. For explanation of technical terms in this abstract see Box 1. First, mixed stands produced more gross growth (+10%) but less net growth (MINUS SIGN 28%) compared with the weighted mean growth of monospecific stands. In monospecific stands, 73% of the gross growth was accumulated in the standing stock, whereas only 48% was accumulated in mixed stands. The gross overyielding of pine (2%) was lower than that of beech (18%). However, the net overyielding of beech was still 10%, whereas low growth and dropout of pine caused a substantial reduction from gross to net growth. Second, the mortality rates, the self- and alien-thinning strength, and the stem volume dropout were higher in mixed stands than monospecific stands. The main reason was the lower survival of pine, whereas beech persisted more similarly in mixed compared with monospecific stands. Third, we found a 10% higher stand density in mixed stands compared with monospecific stands at the first survey. This superiority decreased to 5% in the second survey. Fourth, the mixing proportion of Scots pine decreased from 46% to 44% between the first and second survey. The more than doubling of the segregation index (S) calculated by Pielou index (S increased from 0.2 to 0.5), indicated a strong tendency towards demixing due to pine. Fifth, we showed that with increasing water supply the dropout fraction of the gross growth in t

Many studies show that mixed species stands can have higher gross growth, or so-called overyielding, compared with monocultures. However, much less is known about mortality in mixed stands. Knowledge is lacking, for example, of how much of the gross growth is retained in the standing stock and how much is lost due to mortality. Here, we addressed this knowledge gap of mixed stand dynamics by evaluating 23 middle-aged, unthinned triplets of monospecific and mixed plots of Scots pine (Pinus sylvestris L.) and European beech (Fagus sylvatica L.) repeatedly surveyed over 6–8 years throughout Europe. For explanation of technical terms in this abstract see Box 1. First, mixed stands produced more gross growth (+10%) but less net growth (−28%) compared with the weighted mean growth of monospecific stands. In monospecific stands, 73% of the gross growth was accumulated in the standing stock, whereas only 48% was accumulated in mixed stands. The gross overyielding of pine (2%) was lower than that of beech (18%). However, the net overyielding of beech was still 10%, whereas low growth and dropout of pine caused a substantial reduction from gross to net growth. Second, the mortality rates, the self- and alien-thinning strength, and the stem volume dropout were higher in mixed stands than monospecific stands. The main reason was the lower survival of pine, whereas beech persisted more similarly in mixed compared with monospecific stands.Third, we found a 10% higher stand density in mixed stands compared with monospecific stands at the first survey. This superiority decreased to 5% in the second survey.Fourth, the mixing proportion of Scots pine decreased from 46% to 44% between the first and second survey. The more than doubling of the segregation index (S) calculated by Pielou index (S increased from 0.2 to 0.5), indicated a strong tendency towards demixing due to pine. Fifth, we showed that with increasing water supply the dropout fraction of the gross growth in the mixture s

Biomolecular condensates are aggregates formed from liquid-liquid phase separation through the interactions between nucleic acids and multivalent proteins. These condensates are essential for many biochemical processes inside the cell. Therefore, there has been a large effort during the last decade to create models and methods to describe these systems. Mixtures of RNA and proteins are very common in nature, making it conceivable that RNA-protein interactions are important in many biomolecular condensates. In this thesis these biomolecules are modeled as simple chains of charged beads. First, one- and two-component protein systems are investigated, with results that are consistent with previous findings made by other groups. RNA was then added to few different one-component protein systems. It was found that the presence of a few RNA molecules increases the aggregation propensity in the sense that aggregation sets in at a higher temperature. When the amount of RNA in the system was increased past a certain threshold, this trend was reverted., Cellen är den fundamentala byggstenen för allt liv vi idag känner till. Inuti celler kompartmentaliseras viktiga funktioner till organeller, vilket är cellernas motsvarighet till organ. Klassiska organeller kan kontrollera sin biokemiska miljö med hjälp av ett membran som avgränsar organellens inre från cytoplasman. Under de senaste åren har forskare dock funnit att membran inte är universell för organeller. Biomolekyler, likt vattenmolekyler, kan attrahera och repellera varandra. Precis som vattenmolekyler har olika aggregationstillstånd, har många biomolekyler också det. Dessa biomolekyler kan under lämpliga förhållanden bildar droppar med mycket högre täthet inuti droppen än utanför. Dropparna har kommit att kallas för biomolekylära kondensat och kan ses som membranlösa organeller. Huvudkomponenter i biomolekulära kondensat är proteiner och RNA. De proteiner som ingår tillhör ofta den klass av proteiner som, istället för att vika sig till en särskild struktur, är strukturellt oordnande. Precis som ordinära proteiner är de strukturellt oordnade kedjemoleyler med aminosyror som byggstenar. De ingående aminosyrornas fysikaliska egenskaper avgör hur ett protein beter sig. Det är dock inte enbart vilka aminosyror som förekommer i ett protein som spelar roll, utan även deras inbördes ordning längs kedjan är viktig. Hos majoriteten av de funna biomolekylära kondensaten förekommer också en annan sorts biomolekyl - RNA. Istället för aminosyror är dessa uppbyggda av nukleotider. RNA-molekyler är starkt negativt laddade och attraherar därmed positivt laddade aminosyror som till exempel lysin. En möjlig hypotes är att RNA kan agera som aggregationsfrön, och hjälpa de strukturellt oordnade proteinerna att bilda kondensat. Att förutsäga en given biomolekyls benägenhet att fasseparera har visat sig vara en utmaning. Analytiska teorier kan ibland användas för att ge en viss förståelse av dessa system. Dock bygger analytiska teorier på grova approximationer. Genom att använda num

The effects of hydrocarbon dissociation and subsequent diamond precipitation on the internal structure and evolution of icy giant planets like Neptune and Uranus have been discussed for more than three decades. Inside these celestial bodies, gravity compresses mixtures of light elements to densities of several grams per cubic centimeter while the temperature reaches thousands of Kelvins resulting in thermal energies on the order of chemical bonding and above. Under these conditions, simple hydrocarbons like methane, which are highly abundant in the atmospheres of these planets, are believed to undergo structural transitions that release molecular hydrogen from deeper layers and may lead to compact stratified cores. Indeed, the isentropes of Uranus and Neptune intersect temperature-pressure conditions where first polymerization occurs, and then, in deeper layers, a phase separation into diamond and hydrogen may be possible. Here we show experimental evidence for this phase separation process obtained by in situ X-ray diffraction from polystyrene samples dynamically compressed to 150GPa and 5000 K, which resembles the environment ~10,000 km below the surfaces of Neptune and Uranus. Our findings demonstrate the necessity of high pressures for initiating carbon-hydrogen demixing and imply that diamond precipitation may require ~10x higher pressures than previously suggested by experiments investigating non-isolated hydrocarbons. Besides underlining the general importance of chemical processes inside giant planets, these results will inform evolutionary models of Uranus and Neptune, where carbon-hydrogen demixing can be a significant source for the convection necessary to explain their unusual magnetic fields. Additionally, our experiment demonstrates an alternative path for producing nanodiamonds for scientific and industrial applications that may be superior to current methods using oxygen-deficient explosives.

We investigated various aspects of colloidal dispersions. In this study we not only emphasized the tunability of the particle properties (rods versus spheres, charging properties of the particle surface), but also the properties of the medium in which they reside (oil and/or water). These properties can impact the phase behaviour of colloidal dispersions, and the effective colloid-colloid or colloid-interface interactions. We applied concepts like (long-range) ionic screening, charge regulation, many-body effects, and anisotropy in particle shape, to various dispersions of charged colloidal spheres and hard needles. For the phase behaviour, we studied reentrant melting of colloidal Wigner crystals as function of colloid density, the spontaneaous demixing in binary mixtures of positively charged colloidal spheres, and the formation of string-like structures consisting of charged colloidal particles. Charge regulation, the way how colloidal particles acquire their charge by adsorption or desorption of ions, turned out to be essential not only on the level of the macroscopic phase diagram (such as the occurrence of reentrant fluid phases), but also on the type of structures that could be formed (strings or clusters). Colloid-interface interactions were studied in the context of colloidal particles in a medium that consists of two immiscible solvents, namely oil and water. We showed that so-called non-touching colloids experience an effective interaction potential with the oil-water interface, that is tunable by the Donnan potential and colloidal charge. These colloid-ion forces can not only be tuned in range and strength, but also in being repulsive or attractive at sufficiently large distances from the interface, and should be added to the well-known image-charge and van der Waals potential. Furthermore, we investigated lateral colloid-colloid interactions for colloids that can penetrate the oil-water interface. We suggested that the colloid-colloid latera

Partikkelisysteemien segregaatio eli erottuminen on ilmiö, jossa tasalaatuisen jauheseoksen komponenteilla on taipumus erota toisistaan. Jauheen erottumistaipumus riippuu partikkelien ominaisuuksista, ympäröivistä olosuhteista ja partikkelien välisistä vuorovaikutuksista. Segregaatiomekanismeja on esitetty kirjallisuudessa valtava määrä ja pienetkin erot partikkelien välisissä ominaisuuksissa ja vuorovaikutuksissa voivat johtaa täysin eri segregaatiomekanismeihin. Segregaatioilmiö on lääketeollisuuden näkökulmasta hyvin keskeinen, eikä sitä tunneta vielä riittävän hyvin, jotta siltä osattaisiin systemaattisesti välttyä. Nykyinen segregaatiotutkimus perustuu suurelta osin yrityksen ja erehdyksen kautta tapahtuvaan oppimiseen. Todellisen segregaatioilmiön ymmärtämiseen tarvittaisiin innovatiivisia tutkimusmenetelmiä. Kokeellisen osan tarkoituksena oli kehittää ja perustestata menetelmä, jolla voidaan tutkia erilaisten partikkelisysteemien erottumiskäyttäytymistä, ja käyttää tätä menetelmää farmaseuttisten rae- ja pellettiseosten segregaation tutkimiseen. Tavoitteena oli todistaa kehitetyn Babel-laitteen toimintaperiaatteen soveltuvuus partikkelisysteemien erottumiskäyttäytymisen tutkimiseen, mutta suoritetut kokeet olivat lähinnä menetelmän ja laitteen testausta. Ongelmiksi muodostuivat Babel-laitteen asettamat rajoitukset, partikkelien sähköistyminen ja partikkelien väliset vuorovaikutukset. Käytetyt suoraviivaiset lähestymistavat eivät riittäneet segregaation aiheuttamiseen Babel-laitteella. Vertikaalisen ravistelun seurauksena syntynyt konvektiopyörre esti segregaation syntymisen. Johtopäätöksenä voidaan sanoa, että Babel-laite mittaa hyvin ja toistettavasti sekä se kykenee erottamaan erikokoiset partikkelit ja erilaiset kokojakaumat toisistaan. Laitteen kehittämistavoitteena olisi saada segregaatio paremmin näkyviin jauheseoksissa ravistelun seurauksena. Tällöin voitaisiin tehdä päätelmiä jauheseoksen erottumistaipumuksesta ja systeemissä vallitsevista erottumisme, Segregation or demixing of particle systems is a phenomenon where one component of a homogeneous powder mixture tends to separate from the other components. The segregation tendency of powder depends on the characteristics of particles, environmental conditions and interactions between particles. A huge number of segregation mechanisms are presented in literature and even small differences between the properties of particles and particle interactions can lead to a completely different segregation mechanism. The segregation phenomenon is very essential from the perspective of pharmaceutical industry. However, the phenomenon is not yet sufficiently well known in order for segregation to be systematically avoided. Current research on segregation is largely based on learning through trial and error. Therefore, innovative research methods are needed to understand the true segregation phenomenon. The purpose of the experimental part was to develop and basic test the method of testing the segregation behavior of different particle systems and use this method to examine the segregation behavior of pharmaceutical mixtures of granules and pellets. The aim was to prove that the operating principle of the developed Babel-device is suitable for examining the segregation behavior of particle systems, but the trials carried out mainly consisted of method and device testing. The problems were composed of the limitations imposed by the Babel-device, particle electrification and particle interactions. Linear approaches used were insufficient for creating segregation by the Babel-device. Convection resulting from vertical shaking prevented the generation of segregation. In conclusion, we can say that the Babel-device measures well and reproducibly, and it is able to distinguish different particle sizes and different particle size distributions from each other. The development aim of the device would be to obtain a more visible segregation in the powder mixture as a result of shaking.

We present results of computational simulations of tungsten-inert-gas and metal-inert-gas welding. The arc plasma and the electrodes (including the molten weld pool when necessary) are included self-consistently in the computational domain. It is shown, using three examples, that it would be impossible to accurately estimate the boundary conditions on the weld-pool surface without including the arc plasma in the computational domain. First, we show that the shielding gas composition strongly affects the properties of the arc that influence the weld pool: heat flux density, current density, shear stress and arc pressure at the weld-pool surface. Demixing is found to be important in some cases. Second, the vaporization of the weld-pool metal and the diffusion of the metal vapour into the arc plasma are found to decrease the heat flux density and current density to the weld pool. Finally, we show that the shape of the wire electrode in metal-inert-gas welding has a strong influence on flow velocities in the arc and the pressure and shear stress at the weld-pool surface. In each case, we present evidence that the geometry and depth of the weld pool depend strongly on the properties of the arc.

We study the demixing transition of mixtures of equal size hard spheres and dipolar hard spheres using computer simulation and integral equation theories. Calculations are carried out at constant pressure, and it is found that there is a strong correlation between the total density and the composition. The critical temperature and the critical total density are found to increase with pressure. The critical mole fraction of the dipolar component on the contrary decreases as pressure is augmented. These qualitative trends are reproduced by the theoretical approaches that on the other hand overestimate by far the value of the critical temperature. Interestingly, the critical parameters for the liquid-vapor equilibrium extrapolated from the mixture results in the limit of vanishing neutral hard sphere concentration agree rather well with recent estimates based on the extrapolation of charged hard dumbbell phase equilibria when dumbbell elongation shrinks to zero [G. Ganzenmüller and P. J. Camp, J. Chem. Phys. 126, 191104 (2007)]

This thesis presents the derivation of the theory leading to the determination of the governing equations of chemically reacting flows under local thermodynamic equilibrium, which rigorously takes into account effects of elemental (de)mixing. As a result, new transport coefficients appear in the equations allowing a quantitative predictions and helping to gain deeper insight into the physics of chemically reacting flows at and near local equilibrium. These transport coefficients have been computed for both air and carbon dioxide mixtures allowing the application of this theory to both Earth and Mars entry problems in the framework of the methodology for the determination of the catalytic activity of Thermal Protections Systems (TPS) materials.Firstly, we analyze the influence of elemental fraction variations on the computation of thermochemical equilibrium flows for both air and carbon dioxide mixtures. To this end, the equilibrium computations are compared with several chemical regimes to better analyze the influence of chemistry on wall heat flux and to observe the elemental fractions behavior along a stagnation line. The results of several computations are presented to highlight the effects of elemental demixing on the stagnation point heat flux and chemical equilibrium composition for air and carbon dioxide mixtures. Moreover, in the chemical nonequilibrium computations, the characteristic time of chemistry is artificially decreased and in the limit the chemical equilibrium regime, with variable elemental fractions, is achieved. Then, we apply the closed form of the equations governing the behavior of local thermodynamic equilibrium flows, accounting for the variation in local elemental concentrations in a rigorous manner, to simulate heat and mass transfer in CO2/N2 mixtures. This allows for the analysis of the boundary layer near the stagnation point of a hypersonic vehicle entering the true Martian atmosphere. The results obtained using this formulation ar, Doctorat en sciences appliquées, info:eu-repo/semantics/nonPublished