Refine
Year of publication
Document Type
- Doctoral Thesis (14)
- Article (1)
Has Fulltext
- yes (15)
Is part of the Bibliography
- no (15)
Keywords
- Simulation (15) (remove)
Institute
Publisher
We present classical and hybrid modeling approaches for genetic regulatory networks focusing on promoter analysis for negatively and positively autoregulated networks. The main aim of this thesis is to introduce an alternative mathematical approach to model gene regulatory networks based on piecewise deterministic Markov processes (PDMP). During somitogenesis, a process describing the early segmentation in vertebrates, molecular oscillators play a crucial role as part of a segmentation clock. In mice, these oscillators are called Hes1 and Hes7 and are commonly modeled by a system of two delay differential equations including a Hill function, which describes gene repression by their own gene products. The Hill coefficient, which is a measure of nonlinearity of the binding processes in the promoter, is assumed to be equal to two, based on the fact that Hes1 and Hes7 form dimers.However, by standard arguments applied to binding analysis, we show that a higher Hill coefficient is reasonable. This leads to results different from those in literature which requires a more sophisticated model. For the Hes7 oscillator we present a system of ordinary differential equations including a Michaelis-Menten term describing a nonlinear degradation of the proteins by the ubiquitinpathway. As demonstrated by the Hes1 and Hes7 oscillator, promoter behavior can have strong influence on the dynamical behavior of genetic networks. Since purely deterministic systems cannot reveal phenomenons caused by the inherent random fluctuations, we propose a novel approach based on PDMPs. Such models allow to model binding processes of transcription factors to binding sites in a promoter as random processes, where all other processes like synthesis, degradation or dimerization of the gene products are modeled in deterministic manner. We present and discuss a simulation algorithm for PDMPs and apply it to three types of genetic networks: an unregulated gene, a toggle switch, and a positively autoregulated network. The different regulation characteristics are analyzed and compared by numerical means. Furthermore, we determine analytical solutions of the stationary distributions of one negatively, and three positively autoregulated networks. Based on these results, we analyze attenuation of noise in a negative feedback loop, and the question of graded or binary response in autocatalytic networks.
Background
Despite the current undersupply of cochlear implants (CIs) with simultaneously increasing indication, CI implantation numbers in Germany still are at a relatively low level.
Methods
As there are hardly any solid forecasts available in the literature, we develop a System Dynamics model that forecasts the number and costs of CI implantations in adults for 40 years from a social health insurance (SHI) perspective.
Results
CI demand will grow marginally by demographic changes causing average annual costs of about 538 million €. Medical-technical progress with following relaxed indication criteria and patients’ increasing willingness for implantation will increase implantation numbers significantly with average annual costs of 765 million €.
Conclusion
CI demand by adults will increase in the future, thus will the costs for CI supply. Continuous research and development in CI technology and supply is crucial to ensure long-term financing of the growing CI demand through cost-reducing innovations.
Hintergrund: Trotz der geringen Versorgungsrate mit Cochlea-Implantaten (CIs) bei gleichzeitig steigender Indikation liegen die CI-Implantationszahlen bei Erwachsenen in Deutschland noch immer auf einem relativ niedrigen Niveau.
Methoden: Da in der Literatur kaum fundierte Prognosen vorliegen, wird ein System Dynamics-Modell entwickelt, das die Anzahl und Kosten von CI-Implantationen der erwachsenen Bevölkerung über 40 Jahre aus Sicht der Gesetzlichen Krankenversicherung (GKV) prognostiziert.
Ergebnisse: Die demografische Alterung wird zu einer geringfügigen Steigerung der CI-Nachfrage führen, die durchschnittliche jährliche Kosten von ca. 538 Mio. € verursacht. Der medizinisch-technische Fortschritt mit nachfolgender Ausweitung der CI-Indikationskriterien und die zunehmende Bereitschaft von CI-Kandidaten zur Implantation werden die Implantationszahlen deutlich erhöhen, sodass mit durchschnittlichen jährlichen Kosten von 2,3 Mrd. € zu rechnen ist.
Schlussfolgerung: Die CI-Nachfrage durch Erwachsene wird künftig steigen und damit auch die Kosten für die CI-Versorgung. Kontinuierliche Forschung und Entwicklung in CI-Technologie und -Versorgung sind von entscheidender Bedeutung, um die Finanzierung der wachsenden CI-Nachfrage durch kostensenkende Innovationen langfristig sicherzustellen.
The investigation of complex molecular systems by molecular dynamics simulations has been successfully established and proven as a standard method during the last decades. The use of highly optimized algorithms and steadily increasing, generally available computing resources enables even larger and longer simulations. However, the dynamics of the system itself is not accelerated, and it can be trapped in low energy minima that can only be overcome slowly. A number of methods have therefore been developed to address this problem.
Within the context of this dissertation, a novel algorithm based on replica exchange was developed to solve problems with existing methods, which can now be used for large molecular systems with a low resource consumption. Parameter dependence was systematically evaluated and optimized to define guidelines for correct application. This algorithm was successfully applied to various pharmaceutical and biochemical problems, such as protein folding or protein-protein interactions.
Telemedicine at the Emergency Site – Evaluated by emergency team members in simulated scenarios
(2015)
The hypothesis of this study states that emergency medicine can benefit from telemedicine, whenever paramedics at a remote emergency site request consultation or mentoring by a distant emergency doctor. The hypothesis was semi-qualitatively evaluated in accordance with the protocol of the EU project in the setting of a medical simulation centre. Paramedics encountered simulated standardized emergency case scenarios, connected for teleconsultation and telementoring with emergency doctors by video and audio link through a newly developed real-time HD-video system called LiveCity camera. Paramedics and emergency doctors regarded the simulated scenarios as realistic and relevant and took the simulation seriously. Thus,the following conclusions can be drawn: 1.) Emergency team members encounter situations at the emergency site, in which they would like to get help by a more experienced colleague, especially help with diagnostics and treatment. 2.) The telemedical contact to an emergency doctor makes paramedics feel confirmed in their work, more secure, even in legal aspects. Paramedics do not feel controlled by telemedicine or like a puppet on a string. Their relationship to the patient is not mainly deranged or interfered by the doctor and their course of action is not mainly disrupted. The tele-emergency doctors do not feel like puppet masters and continue feeling as doctors and do not perceive themselves as interferer within the emergency team. 3.) Emergency team members call for a telemedical system providing transmission of vital signs as well as audio- and video-connection. 4.) The LiveCity camera is an effective telemedical tool. The audio quality is good and the orientation on the screen is easy. Paramedics state, that filming the emergency site is easy, does not restrict the field of vision and paramedics can communicate the emergency doctors everything they want to show and tell. Thus the emergency doctors get additional information. While the LiveCity camera is mostly perceived as not too heavy, the LiveCity camera is not easy to operate, very failure-prone and can derange the communication among team members at the emergency site. Nevertheless, the LiveCity camera is not perceived as an additional burden. 5.) Telemedicine is predominantly and largely appreciated by the members of the emergency team. Connecting the tele-emergency doctor to the remote paramedics leads to a perceived faster start of the therapy and is considered as helpful, improving the situation and the quality of patient care. The adherence to medical guidelines and therefore the quality increased, when the paramedics were connected to an emergency doctor through the telemedicine connection. In general, the quality of diagnostics, the correctness of diagnosis and the quality of therapy were rated higher. The majority of paramedics would call a tele-emergency doctor in cases, they wouldn´t normally activate medical support. The emergency team members largely agree in perceiving the tele-emergency doctor system as useful, and they can imagine, working in a tele-emergency system. As a conclusion, the general hypothesis of this study is mainly and in many items supported: Emergency medicine benefits from telemedical support via video- and audio link as studied here with a newly developed real-time HD-video system called LiveCity camera, whenever paramedics at a remote emergency site request consultation or mentoring by a distant emergency doctor.
Modern space missions depend more and more on electric propulsion devices for in-space
flights. The superior efficiency by ionizing the feedgas and propelling them using electric
fields with regard to conventional chemical thrusters makes them a great alternative. To
find optimized thruster designs is of high importance for industrial applications. Building
new prototypes is very expensive and takes a lot of time. A cheaper alternative is to rely
on computer simulations to get a deeper understanding of the underlying physics. In order
to gain a realistic simulation the whole system has to be taken into account including the
channel and the plume region. Because numerical models have to resolve the smallest time
and spatial scales, simulations take up an unfeasible amount of time. Usually a self-similarity
scaling scheme is used to greatly speed up these simulations. Until now the limits of this
method have not been thoroughly discussed. Therefore, this thesis investigates the limits
and the influence of the self-similarity scheme on simulations of ion thrusters. The aim
is to validate the self-similarity scaling and to look for application oriented tools to use
for thruster design optimization. As a test system the High-Efficiency-Multistage-Plasma
thruster (HEMP-T) is considered.
To simulate the HEMP-T a fully kinetic method is necessary. For low-temperature plasmas,
as found in the HEMP-T, the Particle-in-Cell (PIC) method has proven to be the best
choice. Unfortunately, PIC requires high spatial and temporal resolution and is hence
computationally costly. This limits the size of the devices PIC is able to simulate as well
as limiting the exploration of a wider design space of different thrusters. The whole system
is physically described using the Boltzmann and Maxwell equations. Using these system
of equations invariants can be derived. In the past, these invariants were used to derive a
self-similarity scaling law, maintaining the exact solution for the plasma volume, which is
applicable to ion thrusters and other plasmas. With the aid of the self-similarity scaling
scheme the computation cost can be reduced drastically. The drawback of the geometrical
scaling of the system is, that the plasma density and therefore the Debye length does not
scale. This expands the length at which charge separation occurs in respect to the system
size. In this thesis the limits of this scaling are investigated and the influence of the scaling
at higher scaling factors is studied. The specific HEMP-T design chosen for these studies is
the DP1.
Because the application of scaling laws is limited by the increasing influence of charge separation with increased scaling, PIC simulations still are computationally costly. Another approach to explore a wider design space is given using Multi-Objective-Design-Optimization
(MDO). MDO uses different tools to generate optimized thruster designs in a comparatively
short amount of time. This new approach is validated using the PIC method. During this
validation the drawback of the MDO surfaces. The MDO calculations are not self-consistent
and are based on empirical values of old thruster designs as input parameters, which not
necessarily match the new optimized thruster design. By simulating the optimized thruster
design with PIC and recalculate the former input parameters, a more realistic thruster design is achieved. This process can be repeated iteratively. The combination of self-consistent
PIC simulations with the performance of MDO is a great way to generate optimized thruster
designs in a comparatively short amount of time. The proof of concept of such a combination
is the pinnacle of this thesis.
Manipulating and utilizing plasmas becomes a more and more important task in various research fields of physics and in industrial developments. Especially in nowadays spacerelevant applications there are different ideas to modify plasmas concerning particular tasks.
One major point of interest is the ability to influence plasmas using magnetic fields. To study the underlying physical effects that were achieved by these magnetic fields for both scenarios Particle-in-Cell simulations were done. Two examples are discussed in this thesis.
The first example originates from an experiment performed by the European Space Agency ESA in collaboration with the German Space Agency DLR. To verify the possibility of heat-flux reduction by magnetic fields onto the thermal protection system of a space vehicle a simplified experiment on earth was developed. Most of the heat that is created during re-entry comes from compression of the air ahead of the hypersonic vehicle, as a result of the basic thermodynamic relation between temperature and pressure. The shock front, which builds up in front of the vehicle deflects most of the heat and prohibits the surface of the space vehicle from direct contact with the maximum flux. State of the art spacecrafts use highly developed materials like ceramics to handle the enormous heat. An attractive approach to reduce costs is to use magnetic fields for heat-flux reduction. This would allow the use of cheaper materials and thus reduce costs for the whole space mission. A partially-ionized Argon beam was used to create a certain heat-flux onto a target. The main finding of the experimental campaign was a large mitigation of heat-flux by applying a dipole-like magnetic field. The Particle-in-Cell method was able to reproduce experimental observations like the heat-flux reduction. An additionally implemented optical diagnostics module allowed to confirm the results of the spectroscopy done during the experiment. The underlying effect that is responsible for the heat-flux reduction was identified as a coupling between the modified plasma and the dominating neutral flux component. The plasma, that is guided towards the target, act as a shield in front of the target surface for arriving neutrals. These neutrals are slowed down by charge-exchange collisions. Furthermore the magnetic field induces an increased turbulent transport that is also needed to reach a reduction in heat-ux. The turbulent transport was also obtained by three-dimensional Direct Simulation Monte Carlo simulations. Unfortunately, such source driven turbulence can not be expected in space, so that a heat flux reduction in real space applications is questionable. Nevertheless, other effects like the induced turbulence by the rotating vehicle can compensate the missing source driven effect.
The second scenario in which a magnetic field is used to modify the heat flux of a plasma is the operation of the pulsed cathodic arc thruster. The same Particle-in-Cell code was used to simulate a typical pulse of this newly developed thruster of Neumann Space Pty Ltd. The typical behavior of the thruster could be reproduced numerically. The thrust is mainly produced by fast electrons. These electrons are accelerated by electric fields as a result of a plasma-beam instability. This plasma-beam instability was verified by a phase space diagnostics for the electrons. To demonstrate the influence of the magnetic field a simulation of the cathodic arc thruster without magnetic field and one with magnetic field were compared. It was shown that the use of a magnetic field leads to a ten times larger thrust by directing the heat ux. The resulting narrow plume is an additional Advantage of the particle guiding magnetic field. This narrowness of the plume reduces the danger of interaction with other components of the space vehicle.
Both scenarios demonstrate the different capabilities for electromagnetic fields to manipulate plasmas and especially the corresponding heat-flux with respect to certain tasks. The possibilities range from reducing the heat-flux onto a target to maximizing the thrust by directing the heat-ux. This thesis demonstrates that simulations are a great tool to support experiments and to deliver an improved physics understanding. They help to identify the basic physics principles in the different systems, because they can deliver information not accessible to experiments.
In particular, a better understanding of the influence of electromagnetic fields on the heat-flux distribution in space-relevant applications was obtained. This can be the basis for further simulation-guided optimization, e.g. for the design of more effective cathodic arc thrusters. Here, the goal is to minimize costs for prototypes by replacing the hardware by virtual prototypes in the simulations. This allows to test basic design ideas in advance and get more highly-optimized designs at a fraction of time and costs.
Understanding the fundamental mechanisms in the extracellular matrix of cells (ECM) is crucial for the development of drugs and biomaterials. Therefore, an atomistic model of the extracellular matrix is a cost-efficient way to observe influences of drugs, test the effect of mutations or misfolds in proteins or study the properties of fibril or network-forming peptides.
With this thesis, a refined molecular model of an adhesion complex is proposed that contains collagen, fibronectin and the cell receptor integrin. During the building of the model, major new insights are given for each of these proteins and a powerful protein-folding algorithm is
developed.
Ziel dieser Untersuchung sollte eine Validierung der Sehschärfeprüfung mit einem modifizierten Optotypen sein, um Simulation und Aggravation im gutachterlichen Sehtest aufzudecken. Dieses Testverfahren wurde hier nicht an simulierenden, sondern an wahrheitsgemäß antwortenden Probanden erprobt, um deren Verhalten auf besondere Optotypen bei einem Sehtest standardisiert zu evaluieren.
Simulationsbasierte Analyse von Operationsprozessen am Beispiel eines Grund- und Regelversorgers
(2013)
Die ökonomische Analyse von Prozessen sowie der Auslastung einzelner Ressourcen spielen im Krankenhaus zunehmend eine wichtige Rolle. Werkzeuge aus dem Bereich des Operations Research können die Darstellung, die Bewertung und die Gestaltung von Prozessen im Allgemeinen und von Operationsprozessen im Speziellen maßgeblich unterstützen. Insbesondere die diskrete Ereignissimulation gilt als vielversprechendes Verfahren zur Unterstützung wichtiger Analysen im Krankenhaus. Die vorliegende Arbeit untersucht Operationsprozesse eines Grund- und Regelversorgers. Mit Hilfe einer stochastischen diskreten Ereignissimulation werden sowohl bestehende Prozessabläufe modelliert als auch Auswirkungen veränderter Parameter mittels Szenarienrechnungen simuliert und analysiert. Ein besonderer Fokus liegt auf der Untersuchung der personellen und räumlichen Ressourcen sowie wichtiger Prozesskennzahlen, die durch Betrachtungen der entstehenden Kosten ergänzt wird. Die Arbeit zeigt, dass mit Hilfe einer diskreten Ereignissimulation operative Prozesse eines Grund- und Regelversorgers abgebildet und mittels verschiedener Szenarien Auswirkungen von Prozessänderungen betrachtet werden können. In diesen Szenarien lassen sich sowohl die Auslastung der verschiedenen Ressourcen als auch andere wichtige Prozesskennzahlen beeinflussen. Beispielsweise könnte das derzeit im Krankenhaus durchgeführte Leistungsspektrum in einer geringeren Anzahl an OP-Sälen durchgeführt werden, so dass frei werdende Ressourcen für eine alternative Nutzung zur Verfügung stünden. So ließe sich die mittlere Gesamtauslastung der OP-Säle je nach Szenario maßgeblich steigern. Die diskrete Ereignissimulation zeigt sich als hervorragendes Werkzeug für die Analyse wichtiger Fragestellungen im Krankenhaus und dient somit als hilfreiche Unterstützung von Entscheidungsprozessen.
Ziel dieser Arbeit ist es, die Problematik des steigenden Qualitätsanspruchs und des Kostendrucks im Gesundheitssystem aufzugreifen und vor diesem Hintergrund zwei entscheidende Prozesse im Krankenhaus, die Poliklinik als ambulante medizinische Einheit und die Station als stationäre Pflegeeinheit, näher zu untersuchen. Derzeit sind diese beiden Einheiten noch weitgehend unabhängig voneinander bzw. sie sind im Behandlungsprozess hintereinandergeschaltet. Die Effizienz und die Wirksamkeit von Gesundheitsleistungen hängen maßgeblich von der Entwicklung der Arbeitsteilung und vom Zusammenspiel des ambulanten und des stationären Sektors ab. Im Rahmen dieser Arbeit sollen diese organisch gewachsenen Organisationsstrukturen nun aufgebrochen werden. Es findet ein Perspektivenwechsel von traditionellen Strukturen zu einer prozessualen Sichtweise statt. Anhand verschiedener Prozessoptimierungsansätze soll der Prozess der stationären Versorgung auf der Station in den Prozess der Versorgung in der Poliklinik teilweise integriert werden. Ein zentraler Punkt ist dabei die Identifizierung und Bewertung von Synergiepotenzialen durch die Bündelung bzw. durch die Verlagerung der Ressourcen Pflegepersonal und ärztliches Personal. Neben den quantifizierbaren Folgen für den Ressourceneinsatz sollen durch die Neugestaltung der Prozesse auch Verbesserungen in den medizinischen Behandlungsabläufen und in der Qualitätssicherung erreicht werden. Ein Kerngedanke für die Prozessoptimierung ist eine Klassifikation des Prozessobjektes Patient in mobile und immobile stationäre Patienten. Anhand einer Simulation mit dem Programm MedModel werden die Prozessveränderungen im Ist- und Sollzustand ausgewertet und verglichen. Der Erfolg der angestrebten Prozessoptimierung wird an den Dimensionen der Prozessleistungsfähigkeit Kosten, Qualität und Zeit gemessen.
Simulations of Short Model Peptides and Practically Relevant Modeled Titanium Implant Surfaces
(2014)
One of the aims of this work was to generate a non restrained force field model including carbon contamination to make the adsorption simulations more realistic and comparable with experimental data. Another purpose was to find out how the special recognition of small linker proteins on titanium dioxide is working. During this work a fixed and a non restrained rutile (100) model was used and critical properties were observed which are not only related to the surface. The rigid water layers on top of the oxide are very important for the protein and peptide adsorption. Therefore the first discussing object were the properties of the water layers and how they can be influenced. The charge distribution on the surface was found to have a big effect on them. Depending on the charges of the surface atoms or the functional groups, resulting out of the hydroxylation equilibrium, precisely the first water layer gets more rigid or smother. This has a big effect on biomolecule adsorption. The peptides need to penetrate these water layers to generate direct interaction points. The correct description of the surface in molecular dynamic simulations therefore has a high influence on the results. The better the model is the better the findings are comparable with experimental ones. Additionally carbon contamination was mimicked by using a monolayer of pentanol molecules. This fits very good with experimental data (e.g. contact angle) and make the oxide model more hydrophobic. Interaction of proteins and peptides in experiments or in medical use are often observed under normal air conditions, which means that the scaffold is i) hydroxylated by water and ii) carbon contaminated in a short period of time. Therefore investigations were done to find out how the contamination influences the adsorption of a formally know good or bad binding peptide (TiOBP1; TiOBP2). It was found that the TiOBP1 is able to bind the different surface modifications very well which coincides with observations made in experiments. The way of adsorption (direct or indirect) depends on the water layers properties. The first layer on high charged surface models is that rigid, that the peptide is not able to adsorb in a direct way. On the carbon contaminated oxide model the adsorption is possible by reducing the flexibility of the secondary structure motive. In the case of TiOBP2 adsorption on the clean surface model results in only weak binding or even in no interaction. Whereas on the carbon contaminated dioxide the once know bad binder is able to interact with the Pentanol monolayer. No direct adsorption is observed but the hydrophobic side chains have the possibility to orient themselves according to the hydrophobic layer without changing significantly in the secondary structure motive. An additional test peptide (minTBP) adsorbs without being affected by the contamination. This raises the question if the distribution of hydrophobic to hydrophilic amino acids has influence on the adsorption ability according to clean and contaminated surface. For experimental application it could be of interest to generated peptides (GEPI´s) which bind both surface types without changing the secondary structure motives then as we know functionality is based on these structures. In the case of the PHMB polymer adsorption was observed depending on the hydroxylation ratio and therefore on the charge density of the rutile (100) surface. After analysis of the simulations takeaways from experiments could be substantiated. The PHMB interacts with the negative charged surface via the first water layer as a film. So the new force field model describing the rutile (100) titanium dioxide surface with additional carbon contamination model of one monolayer pentanol fits the experimental data very well. The adsorption studied on this surfaces indicates that the contamination as expected makes the surface more hydrophobic and influences the adsorption behavior of the tested peptides especially the secondary structure of TiOBP1. This indeed enhances experimental investigations. Peptides which e.g. link organic and inorganic parts should be good adsorbing on clean and contaminated surfaces by keeping their functionality. Furthermore experimental data can be substantiated by using atomistic simulations like in the case of PHMB adsorption.
Prozess- und Kostenanalyse der Verpackungsoptionen von Sterilgütern in deutschen Krankenhäusern
(2021)
Die Sterilgutversorgung durch die Wiederaufbereitung von Medizinprodukten liegt als ein patientenferner Prozess häufig außerhalb des Fokus von Krankenhausmanagern, obwohl dieser Bereich sowohl medizinisch als auch ökonomisch von wesentlicher Bedeutung für die Qualität des Behandlungsprozesses ist. Der Einsatz verschiedener Erhebungstechniken, wie die Dokumentenanalyse und Zeitmessung, kann zu einem umfassenden Gesamtüberblick beitragen und damit eine Entscheidung unterstützen.
Die vorliegende Arbeit untersucht die unterschiedlichen Verpackungsoptionen von Sterilgütern in sechs Krankenhäusern in Deutschland. Mit Hilfe einer umfangreichen Zeitmessstudie werden zunächst die zeitlichen Unterschiede sowohl bei den verpackungsbezogenen Prozessen als auch bei den infrastrukturellen Prozessen verdeutlicht. Weiterhin werden alle relevanten Kostenpositionen ermittelt, sodass im Ergebnis Kosten für jede Verpackungsoptionen resultieren. Sowohl für die erhobenen Zeiten als auch für die Kosten wird eine Verteilungsanpassung und anschließend auf dieser Grundlage eine Monte-Carlo-Simulation durchgeführt. Um die gewonnen Ergebnisse auf ihre Stabilität zu prüfen, erfolgt eine ausführliche Szenarioanalyse, in der verschiedenste Inputparameter, wie die Personal- und Materialkosten, aber auch die Umschlagshäufigkeiten verändert werden und somit aufzeigen, unter welchen Umständen die Vorteilhaftigkeit einer einzelnen Verpackungsoption gegeben ist. Ergeben sich beispielsweise durch einen Wechsel der Verpackungsoptionen freie personelle Kapazitäten, können diese für eine Ausweitung des Leistungsspektrums sowohl im eigenen Haus als auch für Dritte genutzt werden.
Die Arbeit zeigt, dass nicht allein die Wahl der Verpackungsoption entscheidend ist, sondern ebenso ein optimaler Prozessablauf maßgeblichen Einfluss auf den Zeitaufwand und die Kosten haben kann. Darüber hinaus liefert die Arbeit Handlungs- und Optimierungsansätze und kann damit bei Entscheidungsprozessen unterstützen.
Ion thrusters are Electric Propulsion systems used for satellites and space missions. Within
this work, the High Efficient Multistage Plasma Thruster (HEMP-T), patented by the
THALES group, is investigated. It relies on plasma production by magnetised electrons.
Since the confined plasma in the thruster channel is non-Maxwellian, the near-field plume
plasma is as well. Therefore, the Particle-In-Cell method combined with a Monte-Carlo
Collision model (PIC-MCC) is used to model both regions. In order to increase the sim-
ulated near-field plume region, a non-equidistant grid is utilised, motivated by the lower
plasma density in the plume. To minimise artificial self-forces at grid points bordered by
cells of different size a modified method for the electric field calculation was developed in
this thesis. In order to investigate the outer plume region, where electric field and collisions
are negligible, a ray-tracing Monte-Carlo model is used. With these simulation methods,
two main questions are addressed in this work.
What are the basic mechanisms for plasma confinement, plasma-wall-interaction
and thrust generation?
For the HEMP-T the plasma is confined by magnetic fields in the thruster channel, generated
by cylindrical permanent magnets with opposite polarity. Due to different Hall parameters,
electrons are magnetised, while ions are not. Therefore, the dominating electron transport
is parallel to the magnetic field lines. In the narrow cusp regions, the magnetic mirror effect
reduces the electron flux towards the wall and confines the electrons like in a magnetic
bottle. At the anode, propellant gas streams into the thruster channel, which gets ionised
by the electrons creating the plasma. As a result of the electron oscillation between the two
cusp regions, ionisation of the propellant gas is efficient.
The magnetic field configuration of the HEMP-T also influences the plasma potential inside
the thruster channel. Close to the symmetry axis, the mainly axial magnetic field results in
a flat potential. At the inner wall, the field configuration reduces the plasma wall interaction
to only the narrow cusp regions. Here, the floating potential of the dielectric channel wall
and its plasma sheath result in a rather low radial potential drop compared to the applied
anode potential. As a result, the electric potential is rather flat and impinging ions at the
thruster channel wall have energies below the sputter threshold energy of the wall material.
Therefore, no sputtering appears at the dielectric wall. At the thruster exit the confinement
by the magnetic field is weakened and the potential drops with nearly the full anode voltage.
The resulting electric field accelerates the generated ions into the plume and generate the
thrust, but they are also able to sputter surfaces. During terrestrial testing, sputteringat vacuum vessel walls leads to the production of impurities. The amount of back-flux
towards the channel exit is determined by the sputter yield of the vacuum chamber wall. A
large distance between thruster exit and vessel wall reduces the back-flux and smooths the
pattern of deposition inside the thruster channel. Dependent on their material, the evolving
deposited layers can get conductive, modify by this the potential distribution and reduce
the thrust.
For the HEMP-T, ions are mainly generated at high potential close to the applied anode
potential. Therefore, the accelerated ions producing the thrust gain the maximum energy
as observed in experiment. Ions emitted from the thruster into different angles in the
plume contribute mainly to the ion current at angles between 30 ◦ and 90 ◦ . They mainly
originate from ionisation at the thruster exit. The resulting angular distribution of the
ejected ion current is close to the one of the experiment, slightly shifted by about ten
degrees to higher emission angles. In front of the thruster exit, electrons are trapped by
electrostatics forces. This enhanced density allows ionisation and an additional electron
density structure establishes.
What are possible physics based ideas for optimisation of an ion thruster?
An optimised thruster should have a high ionisation rate inside the thruster channel, low
erosion and an ion angular distribution with small contributions at high angles for min-
imised thruster satellite interactions. In experiments, the HEMP-T satisfies already quite
nicely these requests. In the simulations, low erosion inside the thruster channel and angular
ion distributions close to the experimental data are demonstrated. However, the ionisation
efficiency is lower and radial ion losses are larger than in experiment. A possible explanation
of these differences is an underestimated transport perpendicular to the magnetic field lines,
well known for magnetised plasmas.
A successful example for an optimisation using numerical simulations is the reduction of
back-flux of sputtered impurities during terrestrial experiments by an improved set-up of
the vacuum vessel. The implementation of baffles reduces the back-flux towards the thruster
exit and therefore deposition inside the channel. These improvements were successfully im-
plemented in the experiment and showed a reduction of artefacts during long time measure-
ments. This leads to a stable performance, as it is expected in space.
Achieving commercial production of electricity by magnetic confinement fusion requires improvements in energy and particle confinement. In order to better understand and optimise confinement, numerical simulations of plasma phenomena are useful. One particularly challenging regime is that in which long wavelength MHD phenomena interact with kinetic phenomena. In such a regime, global electromagnetic gyrokinetic simulations are necessary. In this regime, computational requirements have been excessive for Eulerian methods, while Particle-in-Cell (PIC) methods have been particularly badly affected by the "cancellation problem", a numerical problem resulting from the structure of the electromagnetic gyrokinetic equations. A number of researchers have been working on mitigating this problem with some significant successes. Another alternative to mitigating the problem is to move to a hybrid system of fluid and gyrokinetic equations. At the expense of reducing the physical content of the numerical model, particularly electron kinetic physics, it is possible in this way to perform global electromagnetic PIC simulations retaining ion gyrokinetic effects but eliminating the cancellation problem. The focus of this work has been the implementation of two such hybrid models into the gyrokinetic code EUTERPE. The two models treat electrons and the entire bulk plasma respectively as a fluid. Both models are additionally capable of considering the self-consistent interaction of an energetic ion species, described gyrokinetically, with the perturbed fields. These two models have been successfully benchmarked in linear growth rate and frequency against other codes for a Toroidal Alfvén Eigenmode (TAE) case. The m=1 internal kink mode, which is particularly challenging in terms of the fully gyrokinetic cancellation problem, has also been successfully benchmarked using the hybrid models with the MHD eigenvalue code CKA. Non-linear simulations in this TAE case have been performed confirming the analytical prediction of a quadratic relationship between the linear growth rate of the TAE and the saturated amplitude of the TAE for a range of moderate values of the linear growth rate. At higher linear growth rate, a slower scaling of saturated amplitude with linear growth rate is observed. This analysis has been extended to include the non-linear wave-wave coupling between multiple TAE modes. It has been shown that wave-wave coupling results in a significant reduction in the saturated amplitude. It has been demonstrated that both plasma elongation and ion kinetic effects can exert a stabilising influence on the internal kink mode. A population of energetic particles can also exert a stabilising influence at low normalised pressure. At high normalised fast particle pressure the stabilised kink mode has been shown to give way to the m=1 EPM, which has been simulated both linearly and non-linearly (the "fishbone" mode). The first self-consistent simulations of global modes in the magnetic geometry of the optimised stellarator Wendelstein 7-X have been performed both linearly and non-linearly. Limitations have been encountered in performing simulations in 3D geometry. A hypothesis for the cause of these problems is outlined and ideas for mitigation are briefly described. In addition to the hybrid model simulations, some of the first utilisations of a new scheme for mitigating the cancellation problem in the fully gyrokinetic regime have been carried out in the framework of this thesis. This scheme, which was developed separately, is concisely described in this work. The new scheme has been benchmarked with existing gyrokinetic and hybrid results. The linear Wendelstein 7-X simulations and linear and single mode non-linear TAE simulations have been repeated with the new model. It is shown that bulk plasma kinetics can suppress the growth rate of global modes in Wendelstein 7-X. The results of fully gyrokinetic TAE simulations, the first to have been performed to our knowledge, are shown to be in close agreement with those results obtained using hybrid models. In the TAE case, the hybrid models are an order of magnitude less computationally demanding than the new gyrokinetic scheme, which is in turn at least an order of magnitude less computationally demanding than the previous gyrokinetic scheme.