Browsing by Issue Date

Sort by: Order: Results:

Now showing items 21-40 of 861
  • Jääskinen, Väinö (Helsingin yliopisto, 2015)
    In various fields of knowledge we can observe that the availability of potentially useful data is increasing fast. A prime example is the DNA sequence data. This increase is both an opportunity and a challenge as new methods are needed to benefit from the big data sets. This has sparked a fruitful line of research in statistics and computer science that can be called machine learning. In this thesis, we develop machine learning methods based on the Bayesian approach to statistics. We address a fairly general problem called clustering, i.e. dividing a set of objects to non-overlapping group based on their similarity, and apply it to models with Markovian dependence structures. We consider sequence data in a finite alphabet and present a model class called the Sparse Markov chain (SMC). It is a special case of a Markov chain (MC) model and offers a parsimonious description of the data generating mechanism. A Variable length Markov chain (VLMC) is a popular sparse model presented earlier in the literature and it has a representation as an SMC model. We develop Bayesian clustering methodology for learning the SMC and other Markovian models. Another problem that we study in this thesis is causal inference. We present a model and an algorithm for learning causal mechanisms from data. The model can be considered as a stochastic extension of the sufficient-component cause model that is popular in epidemiology. In our model there are several causal mechanisms each with its own parameters. A mixture distribution gives a probability that an outcome variable is associated with a mechanism. Applications that are considered in this thesis come mainly from computational biology. We cluster states of Markovian models estimated from DNA sequences. This gives an efficient description of the sequence data when comparing to methods reported in the literature. We also cluster DNA sequences with Markov chains, which results in a method that can be used for example in the estimation of bacterial community composition in a sample from which DNA is extracted. The causal model and the related learning algorithm are able to estimate mechanisms from fairly challenging data. We have developed the learning algorithms with big data sets in mind. Still, there is a need to develop them further to handle ever larger data sets.
  • Ruusuvuori, Kai (Helsingin yliopisto, 2015)
    New particle formation is an important process in the atmosphere. As ions are constantly produced in the atmosphere, the behaviour and role of charged particles in atmospheric processes needs to be understood. In order to gain insight on the role of charge in atmospheric new particle formation, the electron structure of the molecules taking part in this process needs to be taken into account using quantum chemical methods. Quantum chemical density functional theory was employed in an effort to reproduce an experimentally observed sign preference. While computational results on molecular structures agreed well with results obtained by other groups, the computationally obtained sign preference was opposite to the experimentally observed. Possible reasons for this discrepancy were found in both computational results and experiments. Simulations of clusters containing water, pyridine, ammonia and a proton were performed using density functional theory. The clusters were found to form a core consisting of ammonium ion and water with the pyridine molecule bonding to the ammonium ion. However, the solvation of the ammonium ion was observed to affect the possibility of proton transfer. Calculations of proton affinities and gas phase basicities of several compounds, which can be considered as candidates to form atmospheric ions in the boreal forest, were performed. The generally small differences between the calculated gas phase basicites and proton affinities implied only small entropy changes in the protonation reaction. Comparison with experiments resulted in the conclusion that the largest experimentally observed peaks of atmospheric ions most likely corresponded to pyridine and substituted pyridines. Furthermore, a combination of low proton affinity and high observed cation concentration was concluded to imply a high concentration of neutral parent molecules in the atmosphere. A combination of quantum chemistry and a code for modelling cluster dynamics was employed to study the use of protonated acetone monomers and dimers as the ionization reagent in a chemical ionization atmospheric pressure interface time-of-flight mass spectrometer (CI-APi-TOF). The results showed that the ionization reagents successfully charged dimethylamine monomers. However, there were discrepancies between the simulated and measured cluster distributions. Possible reasons for this discrepancy were found in both measurements and the modelling parameters.
  • Lindberg, Sauli (Helsingin yliopisto, 2015)
    The dissertation deals with the Jacobian equation in the plane. R.R. Coifman, J.-P. Lions, Y. Meyer and S. Semmes proved in their seminal paper from 1993 that when a mapping from the n-space to the n-space belongs to a suitable homogeneous Sobolev space, its Jacobian determinant belongs to a real-variable Hardy space. Coifman, Lions, Meyer and Semmes proceeded to ask the following famous open problem: can every function in the Hardy space be written as the Jacobian of some Sobolev mapping? It follows from the work of G. Cupini, B. Dacorogna and O. Kneuss that the range of the Jacobian operator is dense in the Hardy space. As a consequence of this, solving the Jacobian equation reduces to proving that every so-called energy-minimal solution satisfies certain natural a priori estimate. In the dissertation we use Lagrange multipliers in Banach spaces to prove the sought after a priori estimate for a large class of energy-minimal solutions. It remains unclear whether the class is large enough to imply the surjectivity of the Jacobian operator, but we present many partial results on the properties of the class. To cite an example, when the Hardy space is endowed with a particular norm that is well suited to the study of the Jacobian equation, all the extreme points of the unit ball are Jacobians. Furthermore, the energy-minimal solutions for the extreme points satisfy the wanted a priori estimate. As one of the main results of the dissertation we reduce solving the Jacobian equation to a fairly concrete finite-dimensional problem. As the main tools of the dissertation we use Banach space geometry, harmonic analysis in the plane and methods from the theory of incompressible elasticity.
  • Leinonen, Lasse (Helsingin yliopisto, 2015)
    Supersymmetry is a proposed new symmetry that relates bosons and fermions. If supersymmetry is realized in nature, it could provide a solution to the hierarchy problem, and one of the new particles it predicts could explain dark matter. In this thesis, I study supersymmetric models in which the lightest supersymmetric particle can be responsible for dark matter. I discuss a scenario in which the supersymmetric partner of the top quark called stop is the next-to-lightest supersymmetric particle in the constrained Minimal Supersymmetric Standard Model. Mass limits and various decay branching fractions are considered when the allowed parameter space for the scenario is determined. If the mass of stop is close to the mass of the lightest supersymmetric particle, one can obtain the observed dark matter density. The scenario leads to a novel experimental signature consisting of high transverse momentum top jets and large missing energy, which can be used to probe the model at the LHC. I also discuss an extended supersymmetric model with spontaneous charge-parity (CP) violation and a right-handed neutrino. When CP is spontaneously violated, a light singlet scalar appears in the particle spectrum, which provides new annihilation channels for the lightest supersymmetric particle. In the model, a neutralino or a right-handed sneutrino can produce the observed dark matter density. Dark matter direct detection limits are found to be especially constraining for right-handed sneutrinos.
  • Backman, John (Helsingin yliopisto, 2015)
    Aerosol particles are part of the Earth's climatic system. Aerosol particles can significantly impact the climate. The ability of aerosol particles to do so depends mainly on the size, concentration and chemical composition of the particles. Aerosol particles can act as cloud condensation nuclei (CCN) and can therefore mediate cloud properties. Aerosol particles can thus perturb the energy balance of the Earth through clouds. Aerosol particles can also directly interact with solar radiation through scattering, absorption, or both. The climatic implications of aerosol radiation interactions depend on the Earth s surface properties and the amount of light scattering in relation to light absorption. Light absorbing aerosol particles, in particular, can alter the vertical temperature structure of the atmosphere and inhibit the formation of convective clouds. The net change in the energy balance imposed by perturbing agents, such as aerosol particles, results in a radiative forcing. Globally, aerosol particles have a net cooling effect on the climate, but, not necessarily on a local scale. Accurate measurements of the optical properties of aerosol particles are needed to estimate the climatic effects of aerosols. A widely used means of measuring light absorption by aerosol particles is to use a filter-based measurement technique. The technique is based on light-transmission measurements through the filter when the aerosol sample is drawn through the filter and particles deposit onto the filter. As the sample deposits, it will inevitably interact with the fibres of the filter and the interactions needs to be taken into account. This thesis investigates different approaches to dealing with filter-induced artefacts and how they affect aerosol light absorption using this technique. In addition, the articles included in the thesis report aerosol optical properties at sites that have not been reported in the literature before. The locations range from an urban environment in the city of São Paulo, Brazil, an industrialised region of the South African Highveld, to a rural station in Hyytiälä in Finland. In general, it can be said that sites that are distant from urban areas tend to scatter more light in relation to light absorption. In urban areas, the aerosol particle optical properties show the aerosol particles to be darker.
  • Aalto, Juha (Helsingin yliopisto, 2015)
    Climate, Earth surface processes and soil thermal hydrological conditions drive landscape development, ecosystem functioning and human activities in high latitude regions. These systems are at the focal point of concurrent global change studies as the ongoing shifts in climate regimes has already changed the dynamics of fragile and highly specialized environments across pan Arctic. This thesis aimed to 1) analyze and model extreme air temperatures, soil thermal and hydrological conditions, and the main Earth surface processes (ESP) (cryoturbation, solifluction, nivation and palsa mires) controlling the functioning of high latitude systems in current and future climate conditions; 2) identify the key environmental factors driving the spatial variation of the studied phenomena; and 3) develop methodology for producing novel high quality datasets. To accomplish these objectives, spatial analyses were conducted throughout geographical scales by utilizing multiple statistical modelling approaches, such as regression, machine learning techniques and ensemble forecasting. This thesis was based on unique datasets from the northern Fennoscandia; climate station records from Finland, Sweden and Norway, state of the art climate model simulations, fine scale field measurements collected in arctic alpine tundra and remotely sensed geospatial data. In paper I, accurate extreme air temperature maps were produced, which were notably improved after incorporating the influence of local factors such as topography and water bodies into the spatial models. In paper II, the results showed extreme variation in soil temperature and moisture over very short distances, while revealing the factors controlling the heterogeneity of ground thermal and hydrological conditions. Finally, the modelling outputs in papers III and IV provided new insights into the determination of geomorphic activity patterns across arctic alpine landscapes, while stressing the need for accurate climate data for predictive geomorphological distribution mapping. Importantly, Earth surface processes were found to be extremely climatic sensitivity, and drastic changes in geomorphic systems towards the end of 21st century can be expected. The increase of current temperature conditions by 2 ˚C was projected to cause a near complete loss of active ESPs in the high latitude study area. This thesis demonstrated the applicability of spatial modelling techniques as a useful framework in multiple key challenges of contemporary physical geography. Moreover, with the utilized model ensemble approach, the modelling uncertainty can be reduced while presenting the local trends in response variables more robustly. In future Earth system studies, it is essential to further assess the dynamics of arctic alpine landscapes under changing climatic conditions and identify potential tipping points of these sensitive systems.
  • Karesoja, Mikko (Helsingin yliopisto, 2015)
    In this study several inorganic-organic hybrids and multiresponsive hybrid polymers were prepared and characterised in detail. Especially the focus has been on stimuli responsive materials but also on nanocomposites based on modified montmorillonite clay. Furthermore thin SiO2-capillaries were modified for electrophoretic separations. In all cases different controlled radical polymerisation techniques have been used. The modification of montmorillonite clay was conducted by surface initiated atom transfer polymerisation. Clay was grafted with random copolymer of butyl acrylate and methyl methacrylate and the modified clay was further mixed with a matrix polymer with the same chemical composition to create nanocomposite films. The relation of the nanocomposite structure to its mechanical properties was in the main focus. The extent of exfoliation of the clay in the composite films clearly affected mechanical properties. Montmorillonite clay was also grafted with pH- and thermoresponsive poly(2-dimethylaminoethyl methacrylate). The thermoresponsive properties of the resulting hybrid materials were compared to similar homopolymer. The inner walls of thin silica capillaries were grafted with a cationic polymer, poly([2-(methacryloyl)oxyethyl]trimethylammonium chloride) (PMOTAC). These capillaries were further used in capillary electrophoresis to separate standard proteins, different β-blockers and low-density as well as high density lipoproteins. The separation of the analytes was not possible with bare SiO2-capillaries but with polymer coated capillaries good separation of the analytes was achieved. Hybrid materials based on mesoporous silica particles grafted with poly(N-vinylcaprolactam-b-polyethylene oxide) (PVCL-b-PEO) were synthesised. The challenging synthesis of these hybrids was performed as a combination of surface initiated atom transfer polymerisation and click reactions. Thermal behaviour and the colloidal stability of these hybrid particles were studied. The role of the PEO block in the colloidal stability of the particles was crucial. Finally, multiresponsive hybrid block copolymers based on N-vinylcaprolactam and 2-dimethylaminoethyl methacrylate was prepared. The thermal properties of these block copolymers can be tuned by varying the chain length of PVCL block. On the other hand the thermal behaviour of PDMAEMA block is highly dependent on the environmental conditions like pH and ionic strength.
  • Pohjola, Valter (2014)
    This thesis deals with various aspects of the inverse boundary value problem for the magnetic Schrödinger operator. The first paper extends earlier uniqueness results to the case where the domain is a half space. The two main features of this problem is that the domain is not a bounded set and that the DN-maps are known only on parts of the boundary. The results in this paper extend known results for the slab geometry to the half space case and moreover give some improvements on the conditions for the measurement sets. The second paper (which is joint work with Pedro Caro) deals with the problem of stability. The main aim of the paper is to show that log-type stability holds for rougher classes of potentials A and q, than were previously known. We prove stable determination of an inverse boundary value problem associated to a magnetic Schrödinger operator assuming that the magnetic and electric potentials are essentially bounded and the magnetic potentials admit a certain type of Hölder-type modulus of continuity. The third paper deals with the convection-diffusion equation, which is another first order perturbation of the Laplacian. This equation is closely related to the magnetic Schrödinger equation. Here we use this relationship to show that one can recover a certain scale of Hölder continuous velocity fields from the DN-map. A common theme in the second and third papers is that of lowering regularity requirements, i.e. extending known results so that they apply to larger and more irregular classes of potentials. This is a central research topic in inverse problems.
  • Saarinen, Juha (Helsingin yliopisto, 2014)
    The climatic cooling during the Cenozoic (65 Ma present) culminated in the Pleistocene Ice Ages (ca. 2.6 Ma 10 000 BP) during which the global climate oscillated between relatively warm climatic phases and very cold and dry glacial phases when extensive continental glaciers formed in the Northern hemisphere. The oscillation between the cold and warm climatic stages caused dramatic cyclic changes in the structure of vegetation varying at its extreme between relatively humid forests and very dry and cold mammoth steppes in Europe. These constantly changing and harsh climatic and environmental conditions caused strong extinction and evolution pressures on mammal species. In this thesis I will discuss how two major ecometric variables, body size and diet, of large herbivorous land mammals have varied during the Pleistocene and how these patterns are connected with climate, environmental conditions and competing mammal species. Mammals diversified and started to occupy the niches of large vertebrates after the Late Cretaceous mass extinction which caused the extinction of large non-avian dinosaurs. The frequency of maximum body size in archaic mammal orders shows a significant global peak in the Middle Eocene (ca. 40 Ma) as a result of the diversification and niche filling after the Late Cretaceous mass extinction, but after that maximum size frequency in mammal orders was low until it peaked significantly again the Pleistocene Ice Ages. This indicates that the Pleistocene climatic and environmental conditions favoured particularly large body sizes in mammals. The overall harshness of the Ice Age climate (seasonal, mostly cold and dry conditions and often rapid climatic changes) could have favoured large body sizes in large terrestrial mammals through mechanisms which are more complicated than the often cited benefit of large size for heat conservation (Bergmann s rule). Large size increases the ability to survive over seasonal shortages of resources such as food and water and enables long-distance migrations to areas of better resource availability. On the other hand, strong erosional processes caused by glaciers produced fertile soils and harsh climates reduced the chemical defences of plants, which resulted in seasonally high primary production and plant quality, which would have enabled herbivorous mammals to grow into large sizes during seasons of high productivity. The main factor driving fine-scale body size variations in ungulate populations has been shown by several studies to be resource availability, which is regulated by primary productivity, plant quality, population densities of the ungulate species (intraspecific resource competition) and interspecific resource competition. The comparisons of ungulate body sizes from Middle and Late Pleistocene of Britain and Germany with vegetation openness (percentages of non-arboreal pollen from associated pollen records) show that species with different ecological strategies have different body size patterns in relation to the vegetation structure. The connection between body size patterns and ecological strategies could explain the different responses of body size to vegetation openness. Species which tend to have relatively small group sizes (e.g. deer) show on average larger body sizes in environments where the vegetation structure is open, whereas gregarious, open adapted species (e.g. horses) tend to have smaller average body sizes in open habitats. I suggest this is because open habitats favour large body size in ecologically flexible species with small group sizes due to high resource availability and quality per an individual (relatively low population densities), less size-restricted manoeuvrability and enhanced capability to escape predators, whereas resource limitations for each individual caused by high population densities can become a limiting factor for individual body size in open-adapted, gregarious species which are efficient open-vegetation feeders and form large groups in open habitats. In closed environments, the body size of the open-adapted, gregarious species is not limited by high population density which enables them to attain larger individual sizes. Dietary signals of the key ungulate species in Middle and Late Pleistocene Europe based on mesowear analyses are on average significantly positively correlated with vegetation openness (non-arboreal pollen percentages) at locality-level. However, there are significant interspecific differences. While most of the species show positive correlations between their mesowear signal and non-arboreal vegetation, others, especially the red deer (Cervus elaphus), do not show any correlation. Instead, the mesowear signal of the red deer is significantly more abrasive dominated when other browse-dominated feeders, especially the roe deer (Capreolus capreolus) are present. This indicates that interspecific competition can obscure the effect of available plant material in the diet of ecologically flexible species. This should be taken into account when interpreting the feeding ecology of the key species in palaeocommunities, and especially when attempting to reconstruct palaeoenvironmental conditions from dietary proxies of mammals. Such attempts should ideally be based on as complete dietary analyses of fossil herbivore faunas as possible. In order to extend the palaeodietary and palaeoecological analyses based on mesowear signals of herbivorous mammas, a new tooth wear -based dietary analysis method was developed for elephants and other lamellar toothed proboscideans, based on measuring occlusal relief of their molar teeth as angles. The benefits of that approach compared with other available methods are that it is easy-to-do, fast and robust, and it gives consistent and comparable results for species with different dental morphologies. The preliminary results from that study indicate that the angle measurement method is a powerful tool for reconstructing proboscidean diets from the fossil record.
  • Liao, Li (Helsingin yliopisto, 2014)
    Atmospheric aerosol particles influence the Earth's climate system, affect air visibility, and harm human health. Aerosol particles originate from both anthropogenic and biogenic sources, either from direct emissions or secondary particle formation. Secondary particle formation from gas phase precursors constitutes the largest fraction of global aerosol budget, yet large uncertainties remain in its mechanisms. This thesis attempted to study the source, the formation mechanisms, and the sink of secondary particles based on data analysis of field measurements and chamber experiments. In addition, numerical simulations were performed to model the processes of secondary particle formation observed in the chamber experiments. We summarized our findings into five main conclusions: 1) Monoterpenes originated from anthropogenic sources (e.g. forest industry) can significantly elevate the local average concentrations and result in a corresponding increase in local aerosol loading; 2) Monoterpenes from biogenic emissions show direct link to secondary particle production: the secondary aerosol masses correlate well with the accumulated monoterpene emissions; 3) Temperature influences biogenic monoterpene emissions, resulting in an indirect effect on the biogenic secondary particle production and corresponding cloud condensation nuclei (CCN) formation; 4) Both data analysis and numerical simulation suggested that nucleation involving the oxidation products of biogenic volatile organic compounds (VOCs) and H2SO4 better explains the nucleation mechanism, yet the specific VOCs participating in the nucleation process remains uncertain; 5) The numerical simulation showed evidence of vapor wall loss effect on the yield of secondary particles from the chamber experiments; a reversible gas-wall partitioning had to be considered to properly capture the observed temporal evolution of particle number size distribution during the chamber experiments. The results of this thesis contribute to the understanding on the role of monoterpenes to secondary particle formation. This thesis raises caution on the parameterization of the temperature dependence of biogenic secondary particle formation in predicting the aerosol production potential due to rising temperatures in the future. This work also points out a way for improving the comprehensive numerical models to better understand the secondary particle formation processes and related climatic effects.
  • Sjöholm, Elina (Helsingin yliopisto, 2014)
    Adsorption and clustering induce changes in the vibrational spectra. The main focus of this work is to study some of these changes in a controlled manner for the systems of water and ammonia molecules. The systems that I study are water molecule adsorbed on the Cu(110) surface, ammonia molecule adsorbed on the (111) fcc transition metal surfaces, and the hydrogen-bonded water-ammonia complex. The metal surface is the perturbing environment in adsorption studies. The water and ammonia molecules serve as the perturbing environment for each other in the water-ammonia complex. Density functional theory and coupled cluster calculations have been used to obtain the potential energy surfaces for the adsorption systems and the water-ammonia complex, respectively. The anharmonic vibrational Hamiltonians have been obtained by combining the exact gas phase kinetic energy operators with the potential energy surfaces which are calculated in the presence of the perturbing environment. The vibrational energy levels have been calculated variationally. The procedure explained above gives us the energy levels for the high-frequency vibrations that are mainly localized on the water and ammonia molecules. These vibrations are also present in the isolated molecules. For the adsorption system the largest approximation in the procedure explained above is the use of density functional theory. This error is successfully corrected by calculating the adsorption induced shifts instead of the absolute vibrational energy levels. The adiabatic approximation that separates the vibrational motion of the molecule and the perturbing environment from each other is the biggest approximation for the water-ammonia complex. This error is corrected by including one crucial intermolecular mode into the vibrational model. Finally, the systematic trends upon adsorption are studied for the ammonia molecule adsorbed on several metal surfaces.
  • Suuronen, Jussi-Petteri (Helsingin yliopisto, 2014)
    X-rays are an extremely versatile probe for materials characterization: while conventional medical x-ray imaging is used to visualize structures with macroscopic dimensions, x-ray diffraction and spectroscopy provide information on phenomena at atomic length scales. A fairly recently introduced intermediate-scale method is x-ray microtomography, which is used to image the internal structure of millimeter-sized samples at a resolution of approximately one micrometer. Especially in the case of hierarchical materials, a thorough description of the bulk properties depends on understanding the interplay of differently scaled effects. Multimodal studies characterizing material structures at different length scales are often crucial in achieving this goal. In this thesis, x-ray diffraction, wide-angle x-ray scattering and microtomography were used to analyze the correlations between nanoscale and microscale structure, and microstructural features affecting macroscopic properties in representative model systems. A novel experimental setup was constructed that adds the capability for in situ x-ray scattering experiments to a state-of-the-art x-ray microtomography scanner. Using the microtomography reconstruction to target the x-ray beam in the scattering experiment enables mapping selected crystallographic properties with 200 micrometer resolution. One of the first sets of samples analyzed with the new setup were a series of submillimeter-sized micrometeorites, whose volume and porosity are indicative of their atmospheric entry velocity. Using the combined setup, the microtomography results could be complemented by information on the micrometeorites' mineralogical composition and degree of crystallite orientation obtained with x-ray diffraction. Consisting of stacked platelets with a high aspect ratio, clays and clay-based materials are prime examples of anisotropic materials, where the alignment of nanometer-thick particles produces discernible features also in the micrometer length scale. This was studied by combining microtomography with wide-angle scattering and transmission electron microscopy observations of clay-polystyrene nanocomposites with and without alignment of the clay particles by an external electric field. Compared with pure polystyrene, addition of small amounts of surface modified hectorite clay was found to improve the thermal resilience without seriously degrading the mechanical properties. Another clay material where the nanoscale orientation produces effect in a longer length scale is bentonite. Due to its exceptional swelling and water retention properties, compacted bentonite is used in many waste management applications, including its planned use as a buffer material in repositories for spent nuclear fuel. In this work, combining small-angle x-ray diffraction and microtomography with controlled humidity conditions allowed near simultaneous measurement of both the local clay platelet orientation and platelet spacing, as well as the orientation of microcracks developed in the drying sample. The anisotropic effects were found to be significantly weaker in natural bentonite compared with a purified montmorillonite sample. Even with samples unsuitable for the associated scattering experiments, the three-dimensional information provided by microtomography can produce new insights into the microscopic features that influence the bulk properties of a material or biological system. In this work, a new three dimensional image analysis method was developed for quantifying the orientation distribution of steel fibers from tomography data of steel fiber reinforced concrete. As the orientation distribution plays a fundamental role in determining the bulk mechanical properties of the concrete, controlling the orientation is an open research question with significant economic importance. The results of the experiment showed a significant alignment of the fibers with the edge of the formwork and illustrated the utility of x-ray tomography for measuring the orientation distribution. Water transport in trees is a second example where a phenomenon occurring over the length of the tree is easily disrupted by the development of micrometer-scaled embolisms within the xylem conduits. In this case, the utility of microtomography was demonstrated by non-destructively imaging the contents of individual xylem conduits within a living tree sapling under varying environmental conditions. This enabled following the same sample plants over an extended period of time, which has not been possible with conventional, destructive methods for measuring xylem embolism.
  • Åhlgren, Elina (Helsingin yliopisto, 2014)
    Graphene is a two-dimensional one-atomic-layer-thick carbon nanomaterial. The first reported samples were obtained only in 2004. The properties of graphene have proved to be extreme in many ways. To name a few, it is one of the mechanically strongest materials known, one square meter of it weights less than one milligram, and it has very high electronic conductivity. For a material to be suitable for specific applications in industry, one should be able to modify the structure and the properties of the material to fit the purpose in hand. In the case of graphene, the used method has to be precise enough to tune the structure one atom at a time. Irradiation with energetic particles fits this purpose and it is used by the industry for e.g. to create thin films. The novelty is, that it has not been used in such a precision before. In this thesis the irradiation response of graphene is studied in detail, providing further understanding of the prospects of this material for device applications. The studied systems vary from freestanding graphene to metal supported membrane. The effect of the irradiation changes by the type and energy of the ion. Atomistic simulations provide the needed theoretical tool for modeling the atomic scale behavior of the systems during the irradiation process. This detailed information would be impossible to achieve with existing experimental tools. By choosing the right energy for the bombarding ion, the lattice modification can be predicted and chosen to fit the purpose of the experiment. At the low energy regime (keV's), the interactions are mainly ionic and produced defects are small. Choosing an ion comparable to the lattice atoms in size, namely B or N in the case of graphene, and a low energy, the ion can be used to substitute a lattice atom. This is called doping, and it affects the electronic properties of graphene. The results indicate that doping of graphene is possible with ion beams, and the most suitable energy for this is 50 eV. Other defect types have separate energy ranges where they are the most likely configurations after irradiation. Therefore, by choosing the ion and the energy carefully, the defect type can be controlled. For continuous irradiation, a simple model is constructed to evaluate the amount of atoms leaving the target and to predict the amount of defects graphene can withstand. The calculations show that even with high vacancy concentrations up to 35%, the graphene membrane remains stable. As the graphene sheet is placed on top of a metal substrate, the irradiation response changes. Depending on the energy of the ion, the defect production decreases or increases compared to the freestanding membrane. After the irradiation, there can be more than two carbon atoms stuck at the area between the graphene and the metal for each incoming ion. The amount is substantial, and a new quantity called trapping yield is introduced to describes the amount of trapped atoms. Upon annealing these trapped atoms start to form small graphene platelets under the continuous graphene network in the experiments, creating new nanostructures observed with scanning tunneling microscopy. With high bombarding energies in the range of MeV's, the defect production mechanism differs from the lower energy range. At this energy range the interactions take place mostly via the electronic system. The inelastic scattering of electrons increases the thermal energy of the lattice locally at the path of the ion, creating hole-like defects in a freestanding membrane. The simulations show that the diameter of the hole changes from few nm's up to tens of nm's, and can be varied with the sopping power of the ion. The results show that carefully selected irradiation parameters are in the central part of controlling the type of the modification. The precisions changes from single atoms to tens of nm's, giving multiple options for patterning graphene.
  • Lindborg, Marjaana (Helsingin yliopisto, 2014)
    All stars that have outer convection zones show magnetic activity. This activity strengthens with angular velocity and depth of the convection zone. Cool starspots are believed to be caused by local magnetic field concentrations on the surfaces of stars. They show up as dark regions against a bright photosphere and are observable manifestations of the internal dynamo activity. Therefore the solar dynamo is not unique, only one example of a cyclic dynamo. This makes studies of magnetic activity in other stars important. The convection zone is believed to be the source area of the solar dynamo. The distributed dynamo paradigm is relying on magnetic field generation throughout the convection zone, while the flux-transport paradigm is based on the Babcock-Leighton -effect near the surface in which the mean poloidal field is produced by the merging of twisted magnetic loops. Mean-field dynamo models can produce many observed features both of the solar cycle and some of the features seen in active rapid rotators, but it remains yet under debate which dynamo paradigm is correct. Especially in the case of the Sun, the kinematic mean-field models of different types can lead to a satisfactory reproduction of the solar cycle main properties. These two prevailing dynamo paradigms are briefly introduced in this thesis, as well as their current challenges. For II Peg our time series covers both states of high and low activity. Furthermore we discover a drift of the active region. This drift is also confirmed from photometry, with the carrier fit analysis. The most natural explanation for it would be an azimuthal dynamo wave. DI Psc is a rapidly rotating single giant, which has an interestingly high lithium concentration. We examine the spot behaviour for a two years time-span, and retrieve changes in the activity, which could indicate a fast activity cycle of only a few years.
  • Sundström, Anu-Maija (Helsingin yliopisto, 2014)
    Atmospheric aerosol particles affect public health, environment, weather and climate in various ways, and therefore the importance on obtaining information about their spatial and temporal variation is evident. Remote sensing measurements have particular capability to provide broad horizontal and/or vertical view on the ambient aerosol field from local to global scales. They also can provide observations over remote areas where carrying out in situ measurements is not possible. The aim of this Thesis, is to explore both ground-based and spaceborne remote sensing measurement techniques for monitoring aerosol particles, and their applications on air quality as well as climate studies. In the first part of this Thesis the potential of a ground-based ceilometer-type lidar to be used as an aerosol measurement device is investigated. Ceilometers are originally designed for observing cloud heights, and at the time of the study they were not commonly used to monitor aerosols. The results obtained in this study indicate that the absolute accuracy of a ceilometer-type lidar is sufficient for quantitative aerosol measurements in some applications. The first study using an improved version of the AATSR (Advanced Along-Track Scanning Radiometer) satellite algorithm shows that aerosol optical depth (AOD) can be retrieved with sufficient accuracy over Eastern China, where the aerosol conditions are highly variable and therefore challenging from the satellite remote sensing point of view. In addition, the improved version of the algorithm provides also valuable information about the fine mode particle contribution to the total AOD. The satellite based AOD data is also used to evaluate the performance of a coupled climate-aerosol model. The comparison of ECHAM5-HAM model and satellite-based AOD (from MODerate Imaging Spectroradiometer) showed that, with few exceptions, the model reproduced relatively well the spatiotemporal variation of AOD over India and China. In this Thesis it is also shown that satellite data can be used to derive such climatically relevant quantities that are not directly available in common retrieval products (such as e.g. AOD). By combining coincident observations from two different satellite instruments, an observation-based estimate of the clear-sky shortwave aerosol direct radiative effect ADRE (at the top of the atmosphere) can be established. Results of the case study over Eastern China show that, overall, the satellite-based estimates of ADRE, aerosol-free fluxes, and their spatial variation are in agreement with model-based values.
  • Parkkinen, Pauli (Helsingin yliopisto, 2014)
    Recently, there has been a lot of interest to computationally study ice and icelike systems. Simulating such structures, using ab initio methods require atomic level models. However, creating adequate models for ice is challenging, as there are a huge number of isomers for any ice formation. Two factors contribute to the number of isomers: (i) the positions of oxygens that give rise to various crystal structures, and (ii) the possibilities for hydrogen atom positions for each crystal structure. The second factor means that many nearly degenerate ``proton configurations'' exist for ice. Consequently, large entropy can be observed for ice at normal temperatures. However, because of the large energetic barriers between different proton configurations, 'residual' entropy can be measured even at 0 K. In this thesis, the proton configurational disorder in ice and icelike systems is studied computationally using ab initio methods. In addition, a computer program is developed to aid in the generation and analysis of proton configurations. The systems discussed in the four articles of this thesis are the normal ice (ice Ih), protonated and neutral ice clusters, and icelike systems on surfaces. In the case of ice Ih, the ordering to the lowest energy configuration (ice XI) and the claimed ferroelectricity of this configuration are studied using ice films with and without natrium hydroxide dopants. Calculations with ice films on platinum surfaces were also performed. According to my results, ice XI is not ferroelectric, although NaOH and the Pt(111) substrate are observed to cause some ordering. In the context of ice clusters, phenomena behind the abundance of the 'magic number clusters', observed in experiments, are studied. The results indicate that the protonated cluster energetics are dominated by the local geometry of the H9O4+ Eigen complex, which prefers a near-planar orientation. In addition, the number of hydrogen bonds in a cluster is seen to affect the energetics. However, a proton configurational analysis indicates that the cluster energetics are also affected by the entropy caused by the number of proton configurations. The icelike systems containing less than four-fold coordinated water molecules are studied, and relationships are established between 'hydrogen bond connectivity parameters' and proton configuration energetics.
  • Rämänen, Pirita (Helsingin yliopisto, 2014)
    Due to increased awareness of environmental issues and tightened legislation, bio-based substitutes for traditional petroleum-based polymers are being increasingly sought. Tall oil fatty acid (TOFA) is an attractive material for that purpose being a by-product of kraft pulping. Thus, it is abundant year-round, the price is reasonable, and it does not compete with foodstuff materials. In this study, the preparation and properties of TOFA-based waterborne materials for various coating and barrier applications were examined. Alkyd-acrylic copolymers were synthesized from conjugated and nonconjugated fatty acid-based alkyd resins, as well from rapeseed oil-based alkyd resins for comparison. The polymerization was performed in a miniemulsion, because of the stability and copolymer formation issues. The ratio between the alkyd resin and acrylate monomers was varied and the effect on copolymerization and the copolymer binder properties, such as monomer conversion and grafting of acrylate to the alkyd resin was studied. It was observed that the monomers butyl acrylate (BA) and methyl methacrylate (MMA) showed dissimilar affinity for the grafting site. The steric hindrances prevented MMA from reacting with the double bonds of the fatty acids as readily as BA. The allylic, especially the bis-allylic sites, were the principal grafting sites of MMA, for energetic reasons. However, this effectively retarded the polymerization and increased the homopolymerization of the acrylates. Limiting monomer conversion was overcome, using post-initiation. This research showed that it is possible to prepare stable dispersions of TOFA-based alkyd-acrylate copolymers with varied chemical composition. Self-standing films of these dispersions can be prepared and the dispersions applied effortlessly on paperboard and utilized as barrier material. An increased amount of alkyd resin made the copolymer films more brittle and increased their hydrophobicity. Oxygen barrier performance of the materials was not adequate, but was improved with cellulose. Various cellulose types were modified with TOFA to improve the compatibility between cellulose and polymer matrix. Modified cellulose was added to the copolymer dispersion to improve the mechanical and barrier performance of the copolymer films and coatings. Enhanced strength as well as increased oxygen barrier properties were clearly observed when cellulose was used as filler. The water barrier of the coatings was favorable despite the material composition.
  • Martin, Jussi (Helsingin yliopisto, 2014)
    In this thesis we study the spectrum of a certain boundary value problem in cuspidal domains. This boundary value problem originates from the linear theory of water-waves. One particular motivation for studying the problem in cuspidal domains, is that in some cases the cuspidal shape could act as a wave absorbing structure. Furthermore, the problem itself has also mathematical interest on its own, due to its unusual form where the spectral parameter appears in the boundary condition. The thesis consists of an introductory part and three articles. In the first article we study the problem in a domain that represents a lake or a pond, in which parameter describing the sharpness of the cusp is two. In the article we show that in this case the essential spectrum of the problem contains real numbers which are greater or equal to a certain positive number. In the second article the setting is similar, but the sharpness parameter is allowed to be any real number greater than two. We show that in this case the continuous spectrum of the problem is the set of all non-negative real numbers and this is also the whole spectrum. In addition, we improve the result of the first article by showing that values found in the essential spectrum also belong to the continuous spectrum. In the third article we consider a case where the domain is a canal, filled with two liquid layers of different densities. The cuspidal form is in one of the layers and its sharpness parameter is two. We show that in this case the continuous spectrum of the problem contains real numbers which are greater or equal to a certain positive number, and that the interval from zero to this number is contained in the discrete spectrum.
  • Korpelainen, Virpi (Helsingin yliopisto, 2014)
    Reliability of measurement is a crucial element of both research and industry. Metrological traceability to the SI unit metre guarantees commensurate units, also at nanometre range. In this thesis, a traceability chain is established for nanometre scale measurements. Measurement instruments and methods were developed for accurate measurements, calibration of instruments and transfer standards, and uncertainty estimations. A metrological atomic force microscope (MAFM) was developed and characterized. The MAFM can be used in the calibration of transfer standards and in accurate AFM measurements. Calibration methods for commercial AFMs were developed. A laser diffractometer was also developed for accurate calibration of 1-D and 2-D gratings with a standard uncertainty of several tens of picometres. Laser interferometric position measurement with a calibrated vacuum wavelength is directly traceable to the realization of the metre if measuring full interferometer fringes, but there is small nonlinearity in sub-fringe measurements. Therefore, in sub-nanometre measurements the nonlinearity of the interferometer needs to be corrected. A method for this correction was developed. Laser diffraction measurement is a very accurate method for characterization of grating pitch. One of the main uncertainty sources is the uncertainty of the measured diffraction angle. Therefore, a method for calibration of the rotary table of the laser diffraction setup was developed. The method can be used also in the realization of angle scale. Methods for transfer standard calibration were developed for both pitch and step height calibration by MAFM. An acoustic method was developed for compensation of the refractive index of air in interferometric measurements. Sub-nanometre uncertainty can be reached with this method. Characterization of instruments, validation of methods and uncertainty estimations are a crucial part of traceability. Therefore, uncertainty estimates based on the characterization of the instruments are given for all measurements in this thesis. Comparisons between laboratories are the best way to ensure commensurate measurements. International comparison results between national metrology institutes for pitch and step height transfer standards are listed.
  • Lagerspetz, Eemil (Helsingin yliopisto, 2014)
    We have created a mobile energy measurement application and gathered energy measurement data from over 725,000 devices, running over 300,000 applications, in heterogeneous environments, and constructed models of what is normal in each context for each application. We have used this data to find energy abnormalities in the wild, and provide users of our application advice on how to deal with them. These abnormalities cannot be discovered in laboratory conditions due to the rich interaction of the smartphone and its operating environment. Employing a collaborative mobile energy awareness application with thousands of users allows us to gather a large amount of data in a short time. Such a large and diverse dataset has helped us answer many research questions. Our work is the first collaborative approach in the area of mobile energy debugging. Information received from each device running our application improves the advice given to other users running the same applications. The author has developed a context data gathering hub for smartphones, discovered the need for a common API that unifies network connectivity, energy awareness, and user experience, and investigated the impact of mobile collaborative energy awareness applications, to find previously unknown energy bugs on smartphones, and to improve users' knowledge of smartphone energy behavior.