I present an experimental study on the formation and dissociation characteristics of carbon dioxide (CO2) gas hydrates using Raman spectroscopy. The CO2 hydrates were formed from aqueous salt solutions with salinities ranging from 0-11 wt-%, and the salts used were sodium chloride (NaCl), potassium chloride (KCl), magnesium chloride (MgCl2) and calcium chloride (CaCl2). The experiments were conducted in a high-pressure vessel, in which the aqueous solution was pressurized with liquid CO2 to 6 MPa. First, I investigated how the addition of salts to a CO2-hydrate forming system inhibits the hydrate formation thermodynamically. For this purpose, the molar enthalpy of reaction between strongly and weakly hydrogen bonded water molecules was determined. I observed a decrease in the molar reaction enthalpy of up to 30 % for the highest salt concentration investigated. In addition, the influence of the salts on the solubility of CO2 in water was studied, which was reduced up to 40 %. The results showed that both properties could be well correlated with the effective mole fraction of salt in solution. Furthermore, the decrease in molar reaction enthalpy could be directly correlated with the equilibrium temperature of gas hydrates. This showed that the shift in equilibrium temperature induced by thermodynamic inhibitors was a direct result from the weakened hydrogen bonded network in the water-rich liquid phase before the onset of gas hydrate formation. Additionally, the growth mechanisms of CO2 hydrates were investigated by determining the amount of solid hydrate formed and the respective reaction constant. The reaction constant was not affected by the addition of salts, whereas the maximum amount of solid hydrate formed also showed a good correlation with the effective mole fraction. This finding leads to the assumption that salt does not affect the intrinsic growth mechanisms of hydrate formation, but that the weakened hydrogen bonded network leads to a decrease in the conversion of liquid water to hydrate and more water molecules stay in a liquid in the form of inclusions between the hydrate cages. Lastly, I analyzed the ratio of CO2 and water and the development of hydrogen bonds after the complete dissociation of hydrate. I observed a supersaturation of CO2 in the water-rich phase and found evidence that the excess CO2 exists as dispersed micro- or nanoscale liquid droplets in the liquid water-rich phase. The development of hydrogen bonds in the liquid water-rich phase was the same as before the hydrate formation. These results could be a possible explanation for the memory effect originating from residual nano- and mircodroplets. With this study, I aim to provide a better understanding of the mode of action of thermodynamic inhibitors and to contribute further insights to the controversially debated phenomenon of the memory effect.

Wie vermischen sich Kraftstoff und Luft, wenn ein flüssiger Kraftstoff in einer Umgebung eingespritzt und zerstäubt wird, deren Parameter Druck und Temperatur den kritischen Druck und die kritische Temperatur des Kraftstoffs überschreiten? In dieser Arbeit wurden Experimente basierend auf Raman-spektoskopischen Methoden zur Gemischbildung unter eben solchen Bedingungen durchgeführt. Ziel der Arbeit war der experimentelle Nachweis der Möglichkeit einphasiger Gemischbildung, d.h. des Übergangs von eingespritztem Kraftstoff in das überkritische Regime, und von da Mischung mit der umgebenden initial überkritischen Stickstoffphase ohne Auftreten von Phasengrenzen. Dazu war es nötig, das Zweiphasengebiet der eingesetzten Stoffe exakt zu charakterisieren (die Gas-Flüssig-Gleichgewichte zu messen), und die Temperatur der Flüssigphase zuverlässig während der Gemischbildung zu messen. Mittels eines Mikrokapillar-Aufbaus wurden Daten zu Gas-Flüssig-Gleichgewichten (engl. Vapor-liquid-equilibria, VLE) bei hohen Drücken und Temperaturen erhoben. Dazu wurden unter kontrollierten Bedingungen phasenspezifische Raman-Spektren der Gas- und der Flüssigphase gemessen, aus denen sich in-situ die Gemischzusammensetzung der Phasen ermitteln ließ. Desweiteren wurden Methoden zur Bestimmung der Temperatur der Flüssigphase erarbeitet, sowie eine Methode zur Unterscheidung von Gas- und Flüssiganteil anhand der Raman-Spektren. Die letzten Methoden basieren auf einer Auswertung des Signals der Hydroxyl-Gruppe von Ethanol, welches in der vorliegenden Arbeit als Kraftstoff-Surrogat verwendet wurde. Danach wurden diese Methoden in einer Hochdruck-Hochtemperatur-Einspritzkammer eingesetzt. Hier wurde Kraftstoff unter realistischen Motorbedingungen eingespritzt, und Raman-Spektroskopie zeitlich und örtlich aufgelöst im entstehenden Spray angewandt. Dies erlaubte die Untersuchung der Gemischbildung ohne Beeinträchtigung des Systems, wie etwa durch Zugabe von Marker-Stoffen oder den Einsatz invasiver Messtechniken. Die gewonnenen VLE-Daten stellen eine erhebliche Verbesserung der Datengrundlage in diesem Druck- und Temperaturbereich dar, da Literaturdaten hier rar sind. Der realisierte Mikrokapillar-Aufbau benötigt nur minimale Volumina an Flüssigkeit und Gas, und lässt vielfältige weitere Einsatzmöglichkeiten wie etwa die Messung von VLE-Daten anderer Stoffe oder auch ternärer Gemische, oder die Untersuchung chemischer Reaktionen zu. Gleichgewichte stellen sich aufgrund des hohen Oberflächen-Volumen-Verhältnisses und der insgesamt kurzen Weglängen schnell ein. Die Zuverlässigkeit der gewonnenen Daten konnte durch Vergleich mit den wenigen vorhandenen Literaturdaten gezeigt werden. Bei Vorliegen von Wasserstoffbrückenbindungen konnte die Zuverlässigkeit und Überlegenheit der Raman-Thermometrie basierend auf der „integrated absolute difference spectroscopy“ gezeigt werden, außerdem erlaubt das charakteristische Raman-Signal der Hydroxyl-Gruppe in Wasserstoff-brückenbindung eine Unterscheidung von Gas- und Flüssigphase in überlagerten Spektren. Zum Nachweis der Durchführbarkeit einer solchen Unterscheidung wurde eine Methode entwickelt, um mittels unterschiedlicher Trigger-Signale phasenspezifische Messungen ohne Überlagerung durch eine alternierende Phase durchzuführen. Die gemessenen, örtlich und zeitlich aufgelösten Daten zur Gemischbildung im Spray erlauben die thermodynamische Charakterisierung der Gemischbildung anhand der ermittelten Parameter „globale Gemischzusammensetzung“, „Flüssigphasenanteil“ und „Flüssigphasentemperatur“. Die Ergebnisse zeigten für hohe Umgebungsdrücke und Temperaturen, dass die Flüssigphase Temperaturen jenseits ihrer kritischen Temperatur erreichen kann. Dies lieferte den Nachweis des Auftretens einphasiger Gemischbildung.:I Abbreviations and symbols II Figures III Tables 1. Introduction 2. State of the art 2.1.1. Objective of this thesis 3. Application-oriented fundamentals 3.1. Thermodynamic states 3.1.1. Single-component systems 3.1.2. Multi-compound systems 3.2. Micro-fluidic systems 3.3. Spray break-up 3.4. Raman spectroscopy 3.4.1. Fundamentals 3.4.2. Quantifiability of Raman signals 3.4.3. Liquid fraction determination 3.4.4. Raman thermometry 4. Vapor-Liquid-Equilibra – Experimental setup 4.1. Overview and auxiliary equipment 4.2. Heating system 4.3. Raman probe 4.4. Light guard technique 4.5. Materials and Experiments 5. Vapor-Liquid-Equilibria – Results and discussion 5.1. Data evaluation 5.2. Calibration 5.3. Liquid film correction 5.4. Results ethanol/nitrogen 5.5. Results decane/nitrogen 5.6. Raman thermometry 6. Sprays – Experimental Setup 6.1. Overview and auxiliary equipment 6.2. Calibration setup 6.3. Spray excitation and detection 6.4. Investigated conditions 7. Sprays – Results and discussion 7.1. Data evaluation 7.1.1. Fuel fraction determination 7.1.2. Liquid fraction determination 7.1.3. Liquid temperature determination 7.2. Calibration results 7.3. Spray results 8. Conclusion 9. References / How do fuel and air mix, when liquid fuel is injected and atomized in an environment with parameters pressure and temperature exceeding the respective critical ones of the fuel? In this work, experiments on mixture formation at such conditions based on methods of Raman spectroscopy were performed. Objective of the work was the experimental proof of single-phase mixing, i.e. the transition of injected fuel into the supercritical regime, and therein mixture with the surrounding initially supercritical nitrogen atmosphere without the formation of phase boundaries. To this end, the characterization of the two-phase regime was necessary (i.e. the measurement of the vapor-liquid-equlibria), and the reliable determination of the temperature of the liquid phase during mixture formation. Data on vapor-liquid-equilibria (VLE) were measured in a micro-capillary setup at high temperatures and pressures. To this end, phase-specific Raman spectra of the liquid and the vapor phase were measured at well-controlled conditions, from which the mixture composition of the respective phases was derived in-situ. Furthermore, Methods for the determination of the liquid phase temperature were developed, as well as an approach for the differentiation of the liquid phase signal from the vapor phase signal. The two latter methods exploit the specific signal of the hydroxyl-group of ethanol, which served as a fuel surrogate in this work. In the next step, these methods were applied in a high pressure, high temperature injection chamber. Here, fuel was injected at realistic engine-like conditions, and Raman spectroscopy was applied temporally and spatially resolved across the created spray cone. This approach allowed the Investigation of the mixture formation without affecting the system, compared to e.g. the addition of markers or the use of invasive measurement techniques. The gathered data are a significant addition to the scarce data base available in this pressure and temperature range. The realized micro-capillary setup needs only minimal volume of fluids, and allows various other operational Scenarios like the measurement of VLE data of other components, binary or ternary, or the Investigation of chemical reactions. Equilibria form very fast due to the high surface-to-volume ratio and the short path lenghts. The reliability of the gathered data were shown by comparison with literature. With the presence of hydrogen bonds, the reliability and superiority of the Raman thermometry based on the 'integrated absolute difference spectroscopy' was shown. Furthermore, the characteristic Raman signal of the hydroxyl-group allows for the differentiation of the vapor- and liquid-phase contributions in superimposed spectra from vapor- and liquid-phase. For the proof of feasibility of such a differentiation, a sophisticated method for the phase-specific measurements was developed by exploiting distinctive trigger Signals from the phases, allowing measurements in one phase without cross-talk from the alternating phase. The temporally and spatially resolved data measured during mixture formation in the spray lead to the thermodynamic characterization of the mixture formation with respect to the Parameters 'global mixture composition', 'liquid phase fraction', and 'liquid phase temperature'. The results for high pressures and temperatures inside the chamber show that the liquid or liquid-like phase can reach temperatures exceeding the critical temperature of the fuel. This provides the proof a the existance of single-phase mixing.:I Abbreviations and symbols II Figures III Tables 1. Introduction 2. State of the art 2.1.1. Objective of this thesis 3. Application-oriented fundamentals 3.1. Thermodynamic states 3.1.1. Single-component systems 3.1.2. Multi-compound systems 3.2. Micro-fluidic systems 3.3. Spray break-up 3.4. Raman spectroscopy 3.4.1. Fundamentals 3.4.2. Quantifiability of Raman signals 3.4.3. Liquid fraction determination 3.4.4. Raman thermometry 4. Vapor-Liquid-Equilibra – Experimental setup 4.1. Overview and auxiliary equipment 4.2. Heating system 4.3. Raman probe 4.4. Light guard technique 4.5. Materials and Experiments 5. Vapor-Liquid-Equilibria – Results and discussion 5.1. Data evaluation 5.2. Calibration 5.3. Liquid film correction 5.4. Results ethanol/nitrogen 5.5. Results decane/nitrogen 5.6. Raman thermometry 6. Sprays – Experimental Setup 6.1. Overview and auxiliary equipment 6.2. Calibration setup 6.3. Spray excitation and detection 6.4. Investigated conditions 7. Sprays – Results and discussion 7.1. Data evaluation 7.1.1. Fuel fraction determination 7.1.2. Liquid fraction determination 7.1.3. Liquid temperature determination 7.2. Calibration results 7.3. Spray results 8. Conclusion 9. References

Hydraulic piston accumulators play a major role especially within the field of stationary hydraulics. The calculation of the amount of hydraulic energy which can be stored in such an accumulator is crucial when it comes to a precise system design. The knowledge of the temperature and pressure within the accumulator is required in order to calculate the amount of energy to be stored. The state of the art solution to estimate the state of change of such an accumulator is the implementation of a costly measurement system within the accumulator which tracks the position of the piston. The goal of this paper is to develop and to analyse a time efficient simulation approach for the gaseous phase within a piston accumulator depending on the accumulator’s load cycle. Temperature, pressure, density and velocity profiles inside of the gaseous phase are calculated transiently in order to achieve that goal. The simulation model is derived in one dimensional environment to save computational effort. Having derived a valid model of the gaseous phase it will be possible in future works to replace the expensive position measurement system by pressure and temperature transducers and then use the model to calculate the position of the piston and therefore estimate the state of change.