Traditionell mätteknik innebär att instrument monteras i rör eller på andra sätt förs in i processen. Detta kan vara problematiskt vid svårhanterliga processfluider och i vissa fall ge missvisande analysresultat. Tekniken Aktiv Akustisk Spektroskopi, som bygger på att mäta en utsänd ljudsignals dämpning genom processfluider, kan dock kringgå detta. Mätinstrumentet med denna teknik kallas för Acospector och marknadsförs av företaget Acosense. Instrumentet är helt beröringsfritt och kan installeras och underhållas utan processtopp. Genom multivariat analys av ljudspektrum från instrumentet fås information om processen, fluiden och dess fysikaliska egenskaper så som viskositet, koncentration, torrsubstans, densitet m.m.   Projektet ”Realtidsanalys av kiselsol” genomfördes under 10 veckor som en del av Acosenses utvecklingsarbete för nya användningsområden för Acospector. Projektet utredde möjligheten till realtidsanalys av specifik yta för kiselsol vid en ny mätposition hos Akzo Nobel Pulp Performance Chemicals (tidigare Eka Chemicals).   Projektet har haft som mål att validera Acosense teknik för den nya mätpositionen mot den fysikaliska egenskapen specifik yta. Detta skulle göras genom att ta fram ett så bra kalibreringsunderlag som möjligt och skapa en kalibrerad modell som kan prediktera egenskaper hos ett valideringsset. Ett annat mål var även att granska befintliga laborationsmetoder hos Akzo Nobel för specifik yta.   Arbetet bestod till stor del av att samla in kalibreringsprover som analyserades med avseende på specifik yta, pH, Na2O-halt, turbiditet, densitet, viskositet och konduktivitet. För proverna erhölls även spektrum som behandlades med Direct Fourier Transform och analyserades samt korrelerades med laborationsresultaten med hjälp av multivariat analys.    Viktiga parametrar för att få så jämförbara resultat för specifik yta som möjligt är noggrant och konsekvent utförda laborationsanalyser, pH-elektrodens känslighet och tiden från det att prov tagits ut till dess att analys utförs. Granskning av Akzo Nobels laborationsmetoder gav inga konkreta förslag på förändringar i laborationsinstruktionen, även om projektets och fabriksoperatörers resultat skilde sig åt.     Resultaten från multivariat analys av spektrumen och laborationsresultaten visade efter den initiala valideringen med befintlig data ge en bra modell för att mäta specifik yta. Arbetet med att få fram modellen, genomfördes genom att börja med att identifiera ett fåtal observationer som avvek från övriga. De kunde efter fortsatt granskning uteslutas. Rekommendation är att genomföra ytterligare validering av modellen med ett externt dataset. När ett sådant test är gjort skulle det kunna bevisas hur applicerbar framtagen modell är för applikationen realtidsanalys av specifik yta hos kiselsol.   Inför nya installationer av Acospectorn verkar liknade projekt med riktade punktinsatser för insamling av större mängd laborationsprover genomförda på samma sätt, vara ett bra sätt att kunna säkerställa ett kalibreringsunderlag med bra kvalitet. Noggrant utförda analyser och bra analysmetoder begränsar spridningen i resultaten vilket kan ge bra resultat under en kort tidsperiod. / Traditional measurement techniques are often conducted by mounting instruments inside pipes or are in other ways introducing them to the process. This can be problematic when handling process fluids that are dangerous, cause precipitations or are in other ways hard to handle and might give misleading results. One way to avoid these problems are using the measuring technique Active Acoustic Spectroscopy. Active Acoustic Spectroscopy is conducted by measuring sound signals attenuation through process fluids. The instrument for this is called Acospector and is produced and marketed by the company Acosense. The Acospector is noninvasive and can be installed and maintained without needing to stop the manufacturing process. Through multivariate data analysis of the spectrums from the instrument, vital information is revealed about the process.   The project “Real Time Analysis of Silica Sol” was conducted during 10 weeks as a part of Acosense development plan for new application areas. The project evaluated the possibility of real time analysis of specific surface of silica sol at a new position at Akzo Nobel Pulp & Performance Chemicals (former known as Eka Chemicals).   The objective of the project was to validate Acosense technique for the new position and specific surface. This was to be done by collecting as good calibration data as possible and create a model which could predict properties of a validation data set. Another objective was to evaluate Akzo Nobel’s lab methods for analyzing specific surface.   A large part of the work was to collect calibration samples from the process and analyze them for specific surface, pH, Na2O-content, turbidity, density, viscosity and conductivity. A spectrum was also collected for each calibration sample with the Acospector. These spectrums were then treated with Direct Fourier Transform and analyzed and correlated with lab results by multivariate data analysis.   Important parameters for comparable results are carefully and consistent conducted analysis, the pH-electrodes sensitivity and the time that passes between the out take of the sample and analysis. Evaluation of Akzo Nobel’s lab methods for specific surface gave no suggestions to changes in the lab instruction even though the projects results and factory operators’ results deviated.   The results from the multivariate analysis of the spectrums and lab results showed after an initial validation with collected data to give a good model for prediction of specific surface. The work to develop the model was done by identifying some observations that deviated from the others. These could after further review be excluded from the data set. It is recommended to conduct further validation of the model with a new data set that hasn’t been used to build the model. When this is done, it would be proved how useful the model is for the application of real time analysis of specific surface at Akzo Nobel.    For new installations of the Acospector similar projects, with targeted operations for collecting large amounts of lab samples conducted in the same way, seems to be a good way to ensure good quality of the calibration data. Thoroughly and consistent conducted analyzing methods are crucial to limit the error in the calibration data and produce good results during a short period of time.

The present thesis is devoted to theoretical studies of resonant X-ray scattering and propagation of strong X-ray pulses. In the first part of the thesis the nuclear dynamics of different molecules is studied using resonant X-ray Raman and resonant Auger scattering techniques. We show that the shortening of the scattering duration by the detuning results in a purification of the Raman spectra from overtones and soft vibrational modes. The simulations are in a good agreement with measurements, performed at the MAX-II and the Swiss Light Source with vibrational resolution. We explain why the scattering to the ground state nicely displays the vibrational structure of liquid acetone in contrast to excited final state. Theory of resonant X-ray scattering by liquids is developed. We show that, contrary to aqueous acetone, the environmental broadening in pure liquid acetone is twice smaller than the broadening by soft vibrational modes significantly populated at room temperature. Similar to acetone, the "elastic" band of X-ray Raman spectra of molecular oxygen is strongly affected by the Thomson scattering. The Raman spectrum demonstrates spatial quantum beats caused by two interfering wave packets with different momenta as the oxygen atoms separate. It is found that the vibrational scattering anisotropy caused by the interference of the "inelastic" Thomson and resonant scattering channels in O2. A new spin selection rule is established in inelastic X-ray Raman spectra of O2. It is shown that the breakdown of the symmetry selection rule based on the parity of the core hole, as the core hole and excited electron swap parity. Multimode calculations explain the two thresholds of formation of the resonant Auger spectra of the ethene molecule by the double-edge structure of absorption spectrum caused by the out-of- and in-plane modes. We predict the rotational Doppler effect and related broadening of X-ray photoelectron and resonant Auger spectra, which has the same magnitude as its counterpart-the translational Doppler effect. The second part of the thesis explores the interaction of the medium with strong X-ray free-electron laser (XFEL) fields. We perform simulations of nonlinear propagation of femtosecond XFEL pulses in atomic vapors by solving coupled Maxwell's and density matrix equations. We show that self-seeded stimulated X-ray Raman scattering strongly influences the temporal and spectral structure of the XFEL pulse. The generation of Stokes and four-wave mixing fields starts from the seed field created during pulse propagation due to the formation of extensive ringing pattern with long spectral tail. We demonstrate a compression into the attosecond region and a slowdown of the XFEL pulse up to two orders of magnitude. In the course of pulse propagation, the Auger yield is strongly suppressed due to the competitive channel of stimulated emission. We predict a strong X-ray fluorescence from the two-core-hole states of Ne created in the course of the two-photon X-ray absorption. / QC 20110426

Purpose of this study was to investigate the possibilities of using Near-infrared (NIR) spectroscopy to detect glass in refuse derived fuel (RDF) as well as what on-line data of glass content could be used for in terms of boiler operation and performance determination. Sample configurations were done with dried RDF (to prevent mass loss due to moisture and spectroscopic disturbance) and increasing concentrations of colored soda-lime glass, total of 100 samples. Glass was randomly scattered among the RDF by shaking the added glass and RDF in a bucket to generate representative samples of real life conditions. NIR-spectra acquisition was done between 12000 and 4000 cm-1, at 8 cm-1 resolution and average of 32 scans. The determination of boiler performance was done in accordance to Swedish standards for acceptance testing and heat loss due to glass was treated as slag. Resulting performance calculations showed boiler efficiency via indirect method matching efficiency calculated via direct method (deviating at maximum 2 %) which validates the summarized losses (including due to glass). The heat loss due to glass was calculated to 0,068 MW/%glass, which equated to average of 0,16 MW for 2,37 % glass. Total heat loss was amounted to an average of 11,53 MW. The developed models were not satisfactory in their quality of regression prediction. Although some had, through pre-processing, good development of explained variance at increasing factors, but still had a “Not Applicable” coefficient of determination by regression prediction. The poor quality of models can be explained by poor glass detection (poor representation) of the spectroscopic instrument due to a combination of glass being randomly scattered in the background material and sometimes covered by RDF as well as that the NIR-spectroscopy light beam only hits a small area. By increasing the number of samples upwards 300 -500, the effect of random scatter of glass can be mitigated and acceptable models could be acquired. / FUDIPO

The detection of Earth-like exoplanets and the characterization of their atmospheres is a challenge one needs to solve to assess their habitability and the presence of life in the universe. If this challenge is still unresolved today, even in the era of giant telescopes, it is mainly because of the very high contrast between these exoplanets and their host star and also their proximity. To overcome both of these constraints, a new method combining high-dispersion spectroscopy and nulling interferometry has been imagined. The idea is to use the nulling interferometry to attenuate the star light emission and detect the inner rocky planets with a high angular resolution. The high-dispersion spectroscopy is increasing the exoplanet detectability significantly which enables to relax the star attenuation requirement for an Earth-like observation. Our simulation made for an exoplanet similar to the Earth orbiting Proxima Centauri is giving a condition for the star attenuation ∼10−4 to detect it. Given this condition, we are able to evaluate the unability of a photonic device at our disposal to achieve such performance without dealing with its limitations. If a future project manages to overcome these limitations, this device could be part of a precursor instrument at IPAG to demonstrate experimentally the performance of high-dispersion nulling interferometry. / Upptäckten av jordliknande exoplaneter och karakteriseringen av deras atmosfärer är en utmaning man behöver lösa för att bedöma deras beboelighet och närvaron av liv i universum. Om denna utmaning fortfarande inte är löst idag, även i jätteteleskopens tid, beror det främst på den mycket höga kontrasten mellan dessa exoplaneter och deras värdstjärna och också deras närhet. För att övervinna båda dessa begränsningar har en ny metod som kombinerar högdispersionsspektroskopi och nullingsinterferometri föreställts. Idéen är att använda nullingsinterferometrin för att minska stjärnljusemissionen och upptäcka de inre steniga planeterna med hög vinkelupplösning. Spektroskopin med hög dispersion ökar exoplanetens detekterbarhet betydligt vilket gör det möjligt att minska stjärndämpningsbehovet för en jordliknande observation. Vår simulering för en exoplanet som liknar jorden som kretsar omkring Proxima Centauri ger ett tillstånd för stjärndämpningen att ∼10−4 för att upptäcka den. Med tanke på detta villkor kan vi utvärdera oförmågan hos en fotonisk enhet till vårt förfogande för att uppnå sådan prestanda utan att hantera dess begränsningar. Om ett framtida projekt lyckas övervinna dessa begränsningar kan den här enheten vara en del av ett föregångarinstrument på IPAG för att experimentellt visa prestanda för högdispersionsnullningsinterferometri.

Technology is rapidly becoming more compact and engineers are investigating new applications that are possible with this compact technology. For example photography has recently become an obvious part in our lives due to the shrinking of cameras. If there was a way to implement a light spectrum sensor into our phones there are a lot of applications this could be used for. For example the visual light spectrum can be used to detect the difference between materials in ways that our eyes can not. It could also be used to detect the amount of UV-light when outdoors or be used as an infrared camera. In order for this to be possible the spectrum sensor can’t be too big which makes the common spectrometer a bad choice for the application. This thesis compares two methods of detecting visual light spectrum suitable for On-Chip applications. One method utilize well-defined optical filters while the other works by incorporating an algorithm that boosts the performance of less well-defined optical filters. A simulation was created to assess the performance of the methods and one of these spectrometers were then designed into a circuit board to test the performance in real environments. The results concluded that the algorithmic method currently has better performance to the same price but that the purely optically filtered method is set to be stronger in the future. / Dagens teknik blir snabbt mer kompakt och ingenjörer undersöker nya tillämpningar som är möjliga med den här kompakta tekniken. Till exempel fotografi har nyligen fått en självklar plats i våra liv när kameratekniken blivit mindre. Om det fanns ett sätt att implementera en ljusspektrumsensor i våra telefoner så hade det funnits en mängd applikationer för detta. Till exempel så kan ljusspektrumet användas för att märka skillnad på material som våra ögon inte kan. Det kan också användas för att mäta mängden UV-ljus när man är ute, eller användas som en infraröd kamera. För att det här ska vara möjligt så kan inte spektrumsensorn vara alltför stor vilket gör en traditionell spektrometer blir olämplig. Den här avhandlingen jämför två metoder för att mäta det synliga ljusspektrumet som är passande för att placeras direkt på ett kretskort. En metod använder specifika optiska filter medan den andra metoden använder en algoritm för att förbättra utslaget hos mindre specifika optiska filter. En simulation skapades för att bedöma de två metoderna och ett kretskort designades sedan för att implementera en av spektrometrarna på och testa beteendet i verkliga miljöer. Resultaten visade att den algoritmiska metoden just nu har bättre prestanda men att den rent optiskt filtrerade metoden troligtvis kommer vara starkare i framtiden.

Vattenhalten i ett frystorkat proteinläkemedel bestämdes med KF och LOD som är två primära och traditionella metoder som används för bestämning av vattenhalten i produkter inom flera industrier. Istället för KF och LOD, som är tidskrävande och destruktiva, kan den sekundära, snabba och icke-destruktiva metoden NIR användas. För att kunna implementera NIR måste en kalibrering gentemot en primär metod ske. Syftet med detta projekt var att med en kvantitativ analys finna ett samband mellan NIR och KF samt mellan NIR och LOD för den frystorkade produkten. Det initiala steget var att upprätta en kalibreringsuppsättning och samla in spektrum med NIR. Därefter gjordes försök att bestämma vattenhalten med KF och LOD för den frystorkade proteinprodukten med inledande tester som verifierades genom att spetsa prover med vatten. LOD utfördes i en exsickator och KF-analyserna utfördes kolometriskt på två olika sätt, med och utan extraktion av proverna med metanol. Det slutgiltiga steget var att korrelera insamlat spektrum från NIR med bestämda referensvärden genom att skapa en kalibreringsmodell. Data från insamlat spektrum förbehandlades med multiple scatter correction (MSC), andra derivatan och Savitzy Golay, innan en kalibreringsmodell skapades med partial least square (PLS). Resultatet från LOD-analyserna med exsickator tydde på att tillvägagångssättet som användes inte var lämpligt för produkten och en kalibreringsmodell upprättades inte. Vattenhalten bestämdes med KF med extraktion av åtta prover till 0,144 – 0,558 % [w/w], och med KF utan extraktion till 2,34– 2,972 % [w/w] för åtta prover. Det förväntade värdet var 0,62-1,61 % [w/w]. En kalibreringsmodell upprättades för KF med och utan extraktion samt för det förväntade referensintervallet. Detta resulterade i en korrelationsfaktor på 0,9496 för KF med extraktion, 0,97418 för KF utan extraktion och 0,9932 för de förväntade värdena. Metoderna KF med och utan extraktion var inte lämpliga för produkten i detta projekt, utan fler försök behöver utföras med större kalibreringsuppsättningar för att kunna dra en slutgiltig slutsats. / The water content of a lyophilized protein drug is determined using KF and LOD which are two primary and traditional methods used by several industries for determination of water content in products. Instead of KF and LOD, which are time consuming and destructive, the secondary, fast and non-destructive method near infrared spectroscopy (NIR), can be used. To implement NIR, a calibration towards a primary method must be performed. The purpose of this study was to find a relation between NIR and KF and between NIR and LOD for the lyophilized product with a quantitative analysis. The initial step was to establish a calibration set and collect a spectrum with NIR. Attempts to determine the water content of KF and LOD for the lyophilized protein product were made with initial tests which later were verified by spiking samples with water. LOD was performed in a desiccator and the KF analyses were performed calorimetrically in two different ways: with and without extraction of the samples with methanol. The last step was to correlate the collected spectrum from NIR with determined reference values by creating a calibration model with partial least square (PLS). Prior to model development, data from collected spectrum was pre-treated with multiple scatter correction (MSC), second derivative and Savitzy-Golay filter. The result of the LOD analyses with a desiccator indicated that the method used was not suitable for the product and a calibration model was not established. The water content was determined by KF with extraction of eight samples to 0.144 - 0.558 % [w/w], and with KF without extraction of eight samples to 2.34-2.972 [w/w]. The expected value was 0.62-1.61 % [w/w]. A calibration model was established for KF with extraction of the samples, KF without extraction of the samples and for the expected reference interval. The developed calibration models resulted in a correlation factor of 0.9496 for KF with extraction, 0.97418 for KF without extraction and 0.9932 for the expected values. KF with and without extraction of the samples was not suitable for the product in this project. More experiments are needed with bigger calibration sets to be able to make a conclusion.

The main purpose for this paper is to study iron artefacts excavated in the Black Earth harbour district of Birka. This locale, positioned on the island Björkö in Adelsö par. Sweden, is one of the mayor sites in regard to the Scandinavian Viking age. The how and why of the corrosive process will be touched upon. Specifically how archaeological iron interacts with the soil in which it is found, and what can be done to prevent unnecessary decay after excavation. This, in turn, leads to an observation of how methods have changed in regard to the conservation of archaeological iron, from the processes used in the early days of the field, to the more present day with the method utilized in this study. The conservation of said artefacts is for this study accomplished through the use of a conventional chemical method called EDTA. The usability of this method with waterlogged metallic artefacts is tested. The product of the conservation work revealed the objects to be mostly, iron rivets, bolts or nails. Not too surprising when the articles came from an area widely believed to be a harbour. This specific chemical solution proved not to be a sufficient conservation method for the waterlogged metal objects examined in this paper. This was due to the notably hard crust that covered the objects and the structural fragility of the artefacts. Specific analyses of some of the artefacts were undertaken due to the emergence of substantial amounts of depositions in the final steps of the process. Using XRD, XRF and IR-spectroscopy, the depositions were analysed and the result showed that the depositions most likely consists of a blend of non-water-soluble iron phosphates.

Im Rahmen dieser Arbeit wurde ein synthetischer Membrantransporter für acetylierte Aminosäurecarboxylate entworfen und hergestellt. Als Bindungsstelle für die Carboxylate wurde das Guanidiniocarbonylpyrrol-Motiv von Schmuck verwendet. In den Seitenarm des Pyrrols wurde ein L-Valinamid-Rest eingebracht, um die Möglichkeit zu zusätzlichen Wasserstoffbrückenbindungen zu bieten und gegebenenfalls Substrat- und Enantioselektivität zu erreichen. Zur Herstellung der Löslichkeit in unpolaren Medien wie dem Inneren der Zellmembran musste eine lipophile Gruppe eingebracht werden. Als löslichkeitsvermittelnder Rest wurde Tris-(Dodecyloxy)phenylmethylen ausgewählt, das drei lange unpolare Alkylreste trägt. Zusammengenommen ergab sich so ein Rezeptor für Oxo-Anionen und speziell für Aminosäurecarboxylate mit erhöhter Löslichkeit in organischen Medien. Somit war die Fähigkeit zu Membrantransport gegeben. In Kraftfeldrechnungen erhielt man die vermutliche Struktur des Rezeptor-Substrat-Komplexes, der eine Kombination aus einer Salzbrücke, Wasserstoffbrückenbindungen und einer Stapelwechselwirkung von Guanidinum-Kation, Benzylgruppe und ggf. aromatischem Rest des Aminosäuresubstrates aufweist. Nach erfolgreicher Synthese wurde in Extraktionsexperimenten die Fähigkeit des Rezeptors erprobt, Aminosäurecarboxylate aus einer wässrigen in eine organische Phase aus zu überführen. Man erhielt das beste Extraktionsvermögen für Ac-Trp-OH, gefolgt von Ac Phe OH und Ac Tyr OH. Es wurde eine neue Formel aufgestellt, mit der aus den pKS-Werten der Substrate und den Extraktionsdaten mit und ohne Rezeptor die Bindungskonstanten der Rezeptor-Substrat-Komplexe berechnet werden konnten. Die Größe der Bindungkonstanten entsprach der Reihenfolge Trp > Tyr > Phe ~ Val mit den höchsten Bindungskonstanten für das Tryptophanderivat mit 1.5*10E4 1/M. Zur Bestätigung der Bindungskonstanten wurden ITC-Messungen durchgeführt. Es wurden Messungen des Rezeptors in Chloroform mit den tert-Butylammoniumsalzen der acetylierten Aminosäuren Phenylalanin, Tyrosin und Valin durchgeführt. Für die Werte von Enthalpie und Entropie konnten bei dieser Auswertung konsistente Werte ermittelt werden. Die höchsten Werte der Enthalpie erhielt man für das Tyrosinderivat mit 3.7*10E3 cal/mol, gefolgt vom Phenylalaninderivat mit 2.8*10E3 cal/mol und Valinderivat mit 1.3*10E3 cal/mol. Diese Abstufung entspricht dem Einfluss des aromatischen Restes, der durch die Stapelwechselwirkung mit dem Guanidinium-Kation die Bindungswärme erhöht und durch den damit verbundenen engeren Komplex den Wert für die Entropie senkt. Für die Evaluierung des Transportvermögens wurden U-Rohr-Versuche verschiedener Art durchgeführt. Es wurde ein Gradient von pH 6 in der Ausgangsphase auf pH 8 in der Zielphase eingesetzt, wodurch der Rezeptor an der Grenzfläche zur Zielphase deprotoniert wurde, was zu gerichtetetem Transport führte. Es ergaben sich recht starke Unterschiede für die Fluxwerte der einzelnen Substraten, die der Reihenfolge Val > Phe > Ala > Trp > Tyr folgten. Dabei wurde das Valinderivat um den Faktor 17 schneller als das Tyrosinderivat befördert, mit dem recht hohen Flux von 1.11*10E-6 mol/m2*s, was nahe an den höchsten literaturbekannten Wert für acetylierte Aminosäuren heranreicht. Durch Verwendung gleicher Substratkonzentrationen in Start- und Zielphase konnte aktiver Transport nachgewiesen werden, d.h. Transport gegen das Konzentrationsgefälle. Die Triebkraft des Transportes war der Gradient von pH 6 auf pH 8 zwischen Ausgangs- und Zielphase, der durch den Symport von Substrat und einem Proton ausgeglichen wurde. Bei einem kompetitiven Versuch mit einer Mischung der verschiedenen Substrate in der Ausgangsphase wurden veränderte Fluxwerte und Selektivitäten festgestellt. Die neue Reihenfolge der Transportgeschwindigkeit war nun Trp > Phe > Val > Tyr > Ala, wobei die Fluxwerte fast durchgehend niedriger waren als im Einzelversuch. Die Veränderung der Werte erschließt sich bei Vergleich mit den thermodynamischen Daten aus den Extraktionsexperimenten. Bei direkter Konkurrenz um den Rezeptor wurden diejenigen Substrate mit den höchsten Bindungskonstanten bevorzugt, unabhängig von ihrer Transportgeschwindigkeit. Die schwächer bindenden Substrate wurden aus dem Komplex verdrängt und wiesen deswegen niedrigere Transportwerte auf. Der kompetitive Versuch ist somit eine stärkere Abbildung der Bindungsstärke und entspricht eher der Situation in einer realen Zelle. / Within the scope of this work a new membrane carrier for acetylated amino acids was designed and synthesized. For the binding site of the carboxylate the guanidinio-carbonylpyrrole motif by Schmuck was selected. In the pyrrole’s side chain an L-valinamide residue was introduced, to allow for additional hydrogen bonding and potentially achieve substrate- and enantioselectivity. For solubility in nonpolar media such as the inner part of the cell membrane a lipophilic group had to be introduced. Tris-(dodecyloxy)-phenylmethylene, which bears three long, nonpolar alkyl chains, was selected to procure the desired solubility. All in all this yielded a receptor for oxo-anions and especially for amino acid carboxylates with increased solubility in organic media. This design resulted in the ability for membrane transport. In force field calculations the probable structure of the receptor-substrate-complex was obtained. It showed a combination of a salt bridge, hydrogen bonds and pi-stacking between the guanidinium cation, the benzyl group and, if applicable, the amino acid’s aromatic residue. After the successful synthesis, extraction experiments were carried out to test the receptor’s ability to transfer amino acid carboxylates from an aqueous into an organic phase. The best extractability was attained for Ac-Trp-OH, followed by Ac-Phe-OH and Ac-Tyr-OH. A new equation was established to calculate the binding constants of the receptor-substrate-complexes with the known pKS-values of the substrates and the extraction data with and without receptor. The values of the binding constants followed the order Trp > Tyr > Phe ~ Val with the highest values for the tryptophane derivative with 1.5*10E4 1/M. To confirm the binding constants, ITC experiments were conducted. Measurements of the receptor in chloroform with the tert-butylammonium salts of the acetylated amino acids phenylalanine, tyrosine and valine were conducted.For the enthalpy and entropy consistent values could be determined. These were 3.7*10E3 cal/mol for the tyrosine derivative, 2.8*10E3 cal/mol for the phenylalanine derivative and 1.3*10E3 cal/mol for the valine derivative. This incrementation complies with the influence of the aromatic residue, which increases the binding heat by the pi-stacking and decreases the value of the entropy because of the resulting tighter complex. For the evaluation of the transport capabilities various U-tube experiments were conducted. A gradient from pH 6 in the source phase to pH 8 in the target phase was employed, which led to deprotonation of the receptor near the interface to the target phase, resulting in directed transport. There were quite strong differences for the substrates’ flux values, which followed the order of Val > Phe > Ala > Trp > Tyr. The valine derivative was transported 17 times faster than the tyrosine derivative, with a quite high flux of 1.11*10E-6 mol/m2*s. This is close to the highest literature-known value for acetylated amino acids. By employing analogous substrate concentrations in the source and target phase, active transport, that is transport against the concentration gradient, could be achieved. The driving force of the transport was the gradient from pH 6 to pH 8 between the source and target phase, which was diminished by the symport of substrate and a proton. In a competitive experiment with a mixture of the various substrates in the source phase different values for flux and selectivity were found. The new order of the transport velocities was now Trp > Phe > Val > Tyr > Ala. Nearly all of the flux values were lower than before. The change of the values can be explained by the comparison with the thermodynamic data from the extraction experiments. With direct competition for the receptor, the substrates with higher binding constants were preferred, independent of their transport velocity. The substrates with weaker binding were expulsed from the complex and now showed lower transport values. The competitive transport experiment is therefor a better depiction of the binding strength and comes closer to the situation in a real cell.

<p>Vid analys av fluider med akustisk spektroskopi finns ett behov av att finna multivariata metoder för att utifrån akustiska spektra prediktera storheter såsom viskositet och densitet. Användning av artificiella neurala nätverk och bayesiska nätverk för detta syfte utreds genom teoretiska och praktiska undersökningar. Förbehandling och uppdelning av data samt en handfull linjära och olinjära multivariata analysmetoder beskrivs och implementeras. Prediktionsfelen för de olika metoderna jämförs och PLS (Partial Least Squares) framstår som den starkaste kandidaten för att prediktera de sökta storheterna.</p> / <p>When analyzing fluids using acoustic spectrometry there is a need of finding multivariate methods for predicting properties such as viscosity and density from acoustic spectra. The utilization of artificial neural networks and Bayesian networks for this purpose is analyzed through theoretical and practical investigations. Preprocessing and division of data along with a handful of linear and non-linear multivariate methods of analysis are described and implemented. The errors of prediction for the different methods are compared and PLS (Partial Least Squares) appear to be the strongest candidate for predicting the sought-after properties.</p>

Vid analys av fluider med akustisk spektroskopi finns ett behov av att finna multivariata metoder för att utifrån akustiska spektra prediktera storheter såsom viskositet och densitet. Användning av artificiella neurala nätverk och bayesiska nätverk för detta syfte utreds genom teoretiska och praktiska undersökningar. Förbehandling och uppdelning av data samt en handfull linjära och olinjära multivariata analysmetoder beskrivs och implementeras. Prediktionsfelen för de olika metoderna jämförs och PLS (Partial Least Squares) framstår som den starkaste kandidaten för att prediktera de sökta storheterna. / When analyzing fluids using acoustic spectrometry there is a need of finding multivariate methods for predicting properties such as viscosity and density from acoustic spectra. The utilization of artificial neural networks and Bayesian networks for this purpose is analyzed through theoretical and practical investigations. Preprocessing and division of data along with a handful of linear and non-linear multivariate methods of analysis are described and implemented. The errors of prediction for the different methods are compared and PLS (Partial Least Squares) appear to be the strongest candidate for predicting the sought-after properties.