Refine
Document Type
- Doctoral thesis (19)
Has Fulltext
- yes (19)
Is part of the Bibliography
- no (19)
Year of publication
Keywords
- Silicium (6)
- Silicon (4)
- Silizium (3)
- Gasphase (2)
- Laserablation (2)
- Oberflächenstruktur (2)
- Ripples (2)
- Vapour growth (2)
- Versetzung <Kristallographie> (2)
- Versetzungen (2)
Institute
- FG Experimentalphysik und funktionale Materialien (19) (remove)
In der Promotionsarbeit wurde ein Züchtungsverfahren zur Herstellung strukturell hochqualitativer AlN-Volumenkristalle mittels PVT-Methode entwickelt. Wesentliche Grundvoraussetzungen dafür sind ein thermisch und chemisch stabiles Tiegelmaterial, ein AlN-Quellmaterial mit Sauerstoffverunreinigungen <300 ppm und AlN-Keime mit hoher kristalliner Perfektion. Unter den getesteten potenziellen Tiegelmaterialien (BN, TaC, TaN, NbC, NbN, TaB2, W) zeigten sich TaC und mit Abstrichen W unter AlN-Züchtungsbedingungen ausreichend stabil und wurden für die Wachstumsversuche verwendet. Zur effektiven Reduzierung der Sauerstoffverunreinigungen im AlN-Quellmaterial wurde ein karbothermischer Reduktionsprozess entwickelt, welcher eine Restsauerstoffkonzentration im Quellmaterial von <300 ppm gewährleistet. AlN-Keime für die Homoepitaxie von AlN-Volumenkristallen wurden durch heteroepitaktisches Wachstum auf SiC-Substraten und über spontane Nukleation freistehender AlN-Kristalle hergestellt. Beim heteroepitaktischen Wachstum auf SiC zeigte sich eine starke Abhängigkeit von der Substratpolarität. Wachstum auf C-polarem SiC ist mit geringeren Ätzgrubendichten von 5*10^4 - 10^6 cm^(-2) gegenüber Si-polarem Wachstum mit Ätzgrubensdichten von 5*10^6 - 10^7 cm^(-2) gekennzeichnet. Für beide Substratpolaritäten wurden Modelle des Anwachsstadiums entwickelt. AlN-Kristalle mit bis zu 35 mm im Durchmesser wurden gezüchtet. Die hohe Rissneigung aufgrund von Abkühlspannungen und Si-Konzentrationen von mehreren Prozent im gewachsenen AlN-Kristall vermindern aber die Kristallqualität erheblich. Eine sehr hohe kristalline Perfektion kann hingegen durch spontane Nukleation freistehender AlN-Kristalle auf einer Zwischenebene in der Tiegelmitte gewährleistet werden. Bei Nukleationstemperaturen von 2200 °C wurden isometrische Kristalle mit 12*12*14 mm^3 gezüchtet. Die Kristalle weisen eine zonare Struktur auf, welche durch einen in [000-1]-Richtung gewachsenen Kernbereich mit Versetzungsdichten <10^2 cm^(-2) und einem senkrecht um den Kernbereich gewachsenen Randbereich mit Versetzungsdichten von 10^2 - 10^4 cm^(-2) gekennzeichnet ist. Strukturell hochqualitative (000-1)-Keime wurden aus spontan nukleierten isometrisch gewachsenen AlN-Kristallen präpariert und für die homoepitaktische Volumenkristallzüchtung verwendet. Unter Zuhilfenahme numerischer Temperaturfeldsimulationen wurde ein angepasster Keimhalter entwickelt, welcher ein leicht konvexes Temperaturfeld am Keim gewährleistet und Parasitärwachstum unterdrückt. Somit konnten AlN-Volumenkristalle homoepitaktisch gezüchtet werden, welche eine Durchmesservergrößerung unter Beibehaltung der hohen strukturellen Qualität der Keimkristalle zeigen. Dieses Verfahren bietet die Grundlage, durch die Züchtung mehrerer Kristallgenerationen eine Durchmesseraufweitung auf industriell relevante Größen von 1-2" zu erreichen.
This thesis addresses fundamental physical processes which take place at the surface region of a target during and after the interaction with ultra-short laser pulses. The general goal is to bring together different phenomena and discuss the non-equilibrium nature of the interaction of femtosecond laser pulses (tp < 100 fs) with various materials, in particular dielectrics and semiconductors. Different experiments, using various techniques, are designed to explore the basic mechanisms of laser ionization, defect creation, electron-lattice energetic transfer, charged particles desorption, optical breakdown, phase transformations and surface morphological changes. Such processes are shown to depend strongly on the laser intensity. Thus, they are analyzed for intensities over four orders of magnitude (10^11-10^14 W/cm2), around the surface optical breakdown (damage) threshold intensity. First, experimental studies using time-of-flight mass spectrometry indicate that non-resonant intense ultra-short laser pulses can efficiently ionize a dielectric (semiconducting) material leading to emission of electrons as well as charged particles, i.e. atomic ions and large clusters, and neutral particles. Under these irradiation conditions, the ionization processes can be at best described by multiphoton ionization and ionization at defects sites. The structural defects provide the means for an increased positive ion desorption rate. A multiple pulse incubation effect in the ion yield can be well related with the reduction of the multi-pulse damage threshold with increasing intensity. Following the initial electron excitation and emission, positive ions are released from the surface in a substantial amount with high ion velocities indicative of a localized microscopic electrostatic expulsion. With increasing intensity, the amount of ions gets larger and larger and their velocity distribution exhibits a bimodal structure. Also, in these conditions, negative ions are detected. The ion desorption can arise from a combination of a localized electrostatic repulsion (macroscopic Coulomb explosion) and a thermal ‘explosive’ mechanism. The later becomes more important with increasing intensity. The very fast energy input and particle emission result in a transient perturbation and deformation of the target lattice. Using pump-probe experiments the temporal evolution of lattice dynamics can be analyzed upon single-pulse excitation for many different target materials. This deformation is indicated to be a material characteristic. It is associated with the generation of transient defects in dielectrics or fast phase transitions in semiconductors and metals. Therefore, it could well give estimates of lifetime of transient defect states or electron-phonon relaxation times.At last the surface morphology after ablation is analyzed, with emphasis on the laser-induced surface periodic patterns (ripples). The patterns observed appear to be very different from the ‘classical’ ripples formed after long pulse ablation. They can have periods much smaller than the incident wavelength and are rather insensitive to the variation of the laser wavelength and angle of incidence. We show that control factors are laser beam polarization and the irradiation dose. Additionally, the patterns exhibit features pointing toward a chaotic origin. Their possible formation mechanism is likely linked with the non-equilibrium nature of the interaction.
This thesis addresses the electro-optical properties of silicon, containing dislocations. The interest in those properties is driven mainly by two practical reasons. One is the optical characterisation of multicrystalline silicon for solar cells, and the other is the design of light emitting diodes based on silicon by enhancement of silicon radiative properties via introduction of dislocations. The work demonstrates that dislocation specific radiation may provide a means for optical diagnostics of solar cell grade silicon. It provides insight into the mechanisms governing the dislocation recombination activity, their radiation, and how are they influenced by other defects present in silicon. We demonstrate that photoluminescence mapping is useful for monitoring the recombination activity in solar cell grade silicon and can be applied for identification of contaminants, based on their photoluminescence signatures. It is shown that the recombination at dislocations is strongly influenced by the presence of metals at the dislocation sites. The dislocation radiation activity correlates with their electrical activity. Thus, photoluminescence mapping at room temperature may provide a means for revealing and characterising of dislocation-rich regions in multicrystalline silicon. It is shown that the dislocation and band-to-band luminescence are essentially anti-correlated. The band-to-band intensity being related to the diffusion length of minority carriers can be used for measurements of diffusion length, as long as the surface recombination rate is controlled. Moreover, photoluminescence mapping can be used for the detection of optically active defects in solar grade materials. Thus, betaFeSi2 precipitates, with a luminescence at 0.8 eV, were detected within the grains of block cast materials. They exhibit a characteristic feature of quantum dots, namely blinking. The second aspect of the thesis concerns the topic of silicon based light emitters for on-chip optical interconnects. The goal is an enhancement of sub-band-gap or band-to-band radiation by controlled formation of dislocation-rich areas in microelectronics-grade silicon as well as understanding of the processes governing such enhancement. For light emitters based on band-to-band emission it is shown, that internal quantum efficiency of nearly 2 % can be achieved, but the emission is essentially generated in the bulk of the wafer. On the other hand, light emitters utilizing the emission from dislocation-rich areas of a well localized wafer depth were explored. Three different methods for reproducible formation of a dislocation-rich region beneath the wafer surface were investigated and evaluated in view of their room temperature sub-band-gap radiation: (1) silicon implantation and annealing, (2) epitaxially grown SiGe buffer, and (3) direct wafer bonding. The most promising dislocation-based emitter appears the utilization of a dislocation network produced by wafer bonding. It is shown, that monochromatic D1 radiation (wavelength 1.5 µm) can be generated in a well localised depth of the wafer. The radiation is not absorbed in silicon and such localized emitter can, potentially, be coupled with silicon waveguides and Ge-based detectors for optical interconnects.
The International Linear Collider offers a lot of different interesting challenges concerning the physics of elementary particles as well as the development of accelerator and detector technologies. In this thesis, we investigate two rather separate topics - the precision measurement of the Higgs boson mass and of its coupling to the neutral gauge boson Z and the research and development of sensors for BeamCal, which is a sub-detector system of the ILC detector. After the Higgs boson has been found, it is important to determine its properties with high precision. We employ the Higgs-strahlung process for this purpose. A virtual Z boson is created in the electron-positron collisions, which emits a Higgs-boson while becoming on-shell. Using the so-called recoil technique, we determine the Higgs boson mass by reconstructing the Z boson momentum and using the center-of-mass energy of the colliding leptons. This technique allows to measure the Higgs boson mass without considering the Higgs boson decay, i.e. it can be applied even to a Higgs boson invisibly decaying. Monte-Carlo studies including a full detector simulation and a full event reconstruction were performed to simulate the impact of a realistic detector model on the precision of the Higgs boson mass and production cross-section measurement. Also, an analytical estimate of the influence of a given detector performance on the Higgs boson mass measurement uncertainty is given. We included a complete sample of background events predicted by the Standard Model, which may have a detector response similar to the signal events. A probabilistic method is used for the signal-background separation. Several other probabilistic methods were used to investigate and improve the measurement of the Higgs-strahlung cross-section and the Higgs boson mass from the recoil mass spectrum obtained after the signal-background separation. For a Higgs boson mass of 120 GeV, a center-of-mass energy of 250 GeV and an integrated luminosity of 50/fb, a relative uncertainty of 10% is obtained for the cross-section measurement, and a precision of 118 MeV for the Higgs boson mass. The original motivation to use the recoil technique for a Higgs boson mass measurement independent on its decay modes could not be completely confirmed. For a Higgs boson mass of 180 GeV and 350 GeV, a statistics corresponding to 50/fb is not sufficient to achieve the necessary significance of the recoil mass peak above the background. The BeamCal is a calorimeter in the very forward region, about 3 m away from the nominal interaction point and surrounding the beam pipe. Due to its location, a lot of beamstrahlung pair particles will hit this calorimeter, representing a challenge for the operational reliability of the sensors under such harsh radiation conditions. We investigated single-crystal and polycrystalline CVD diamond, gallium arsenide and radiation-hard silicon as sensor candidates for their radiation hardness and found that diamond and gallium arsenide are promising. We used a 10 MeV electron beam of few nA to irradiate the samples under investigation up to doses of 5 MGy for diamond, up to about 1.5 MGy for gallium arsenide and up to about 90 kGy for silicon. We measured in regular periods the CCD to characterize the impact of the absorbed dose on the size of the signal, which is generated by electrons of a Sr-90 source crossing the sensor. Additional measurements such as the dark current and the CCD as functions of the voltage completed the characterization of the sensor candidates. For the single-crystal CVD diamond, also the thermally stimulated current was measured to determine amongst others the defect density created by irradiation. In the diamond samples, evidence for strong polarization effects inside the material was found and investigated in more detail. A phenomenological model based on semi-conductor physics was developed to describe the sensor properties as a function of the applied electric field, the dose and the dose rate. Its predictions were compared with the results of the measurements. Several parameters such as time scales and cross-sections were determined using this model, which led to ongoing investigations.
The thesis presents the results of the investigations of electronic properties and defect states of dislocation networks (DNs) in silicon produced by wafers direct bonding technique. Practical interest for the investigations in this area issued – first of all – from the potential application of such dislocation networks in microelectronics as all-Si light emitter for on-chip interconnection. Besides, dislocation networks may serve as a perfect model object to get new information about the fundamental properties of dislocations and grain boundaries in Si, what is of particular importance for multicrystalline silicon solar cells performance. Despite of a long story of studying of dislocations in silicon, a new insight into the understanding of their very attractive properties was succeeded due to the usage of a new, recently developed silicon wafer direct bonding technique, allowing to create regular dislocation networks with predefined dislocation types and densities. Samples for the investigations were prepared by hydrophilic bonding of p-type Si (100) wafers with same small misorientation tilt angle (~0,5°), but with four different twist misorientation angles Atw (being of <1°, 3°, 6° and 30°, respectively), thus giving rise to the different DN microstructure on every particular sample. The main experimental approach of this work was the measurements of current and capacitance of Schottky diodes prepared on the samples which contained the dislocation network at a depth that allowed one to realize all capabilities of different methods of space charge region spectroscopy (such as CV/IV, DLTS, ITS, etc.). The key tasks for the investigations were specified as the exploration of the DN-related gap states, their variations with gradually increasing twist angle Atw, investigation of the electrical field impact on the carrier emission from the dislocation-related states, as well as the establishing of the correlation between the electrical (DLTS), optical (photoluminescence PL) and structural (TEM) properties of DNs. The most important conclusions drawn from the experimental investigations and theoretical calculations can be formulated as follows: - DLTS measurements have revealed a great difference in the electronic structure of small-angle (SA) and large-angle (LA) bonded interfaces: dominating shallow level and a set of 6-7 deep levels were found in SA-samples with Atw of 1° and 3°, whereas the prevalent deep levels – in LA-samples with Atw of 6° and 30°. The critical twist misorientation angle separating SA- and LA- interfaces was estimated as Atw*≈ 3,5±0,5°, what agrees quiet well with the results of previous PL and TEM investigations. - For the dominating shallow traps in SA-samples (denoted as ST1/ST3 traps) a new phenomenon – that is ‘giant Poole-Frenkel effect’ of enhanced carrier emission due to dislocations elastic strain field was observed for the first time. Performed theoretical calculations have shown that in the investigated samples such an effect should be ascribed to the row of 60° dislocations rather than to the mesh of screw ones. In this respect, shallow traps ST1/ST3 were identified either with shallow 1D bands (directly or as being coupled with them) or with shallow stacking fault states on splitted 60° dislocation. - From the comparison and correlations of measured DLTS spectra with the results of PL and TEM investigations it was established, that shallow ST1/ST3 traps participate in D1 radiative recombination and that the structural elements, responsible for D1 luminescence of small-angle DNs, are the triple knots (intersections with screw dislocations) along the 60° dislocations. However, the optimal density of 60° dislocations as well as of triple knots, in other words – the optimal tilt and twist misorientation angles for maximal D1 intensity – needs further clarification.
Die stetig fortschreitende Miniaturisierung in der Halbleiterindustrie macht es notwendig, Oberflächenparameter mit Auflösung im Nanometerbereich zu messen und auch abzubilden. Von größtem Interesse sind hierbei das Oberflächenpotential und die Kapazität der oberflächennahen Bereiche, da diese Aussagen über die elektronische Struktur erlauben. Hierbei muss großes Augenmerk auf die Möglichkeit der zerstörungsfreien und präparationsarmen Messung gelegt werden, da jede Behandlung der zu untersuchenden Materialien deren Oberflächeneigenschaften ändert. Im Rahmen dieser Arbeit wurden auf der Atomkraftmikroskopie basierende Methoden sowohl experimentell als auch mit Hilfe von Simulationen auf ihre Anwendbarkeit für die Untersuchung von Halbleiteroberflächen evaluiert. Es stellt sich heraus, dass die kontaktfreien Methoden „Scanning Kelvin Probe Microscopy“ und „Scanning Capacitance Microscopy“ sehr gut geeignet sind, um die elektronische Struktur der Probenoberfläche qualitativ zu beurteilen. Allerdings muss für quantitative Aussagen ein recht großer rechentechnischer Aufwand betrieben werden.
Gallium nitride (GaN) is a III-V semiconductor, characterized by direct, wide band gap of 3.4 eV at RT. As a material of particular interest for opto- and power electronics applications, it has been thoroughly studied in recent years. Utilization of GaN homoepitaxy in manufacturing of laser diodes (LDs), light-emitting diodes (LEDs), power devices, etc. would be beneficial in terms of reducing defect density, thus improving their lifetime and performance. Yet cost-effective process for providing native GaN substrates has not been established so far.
The focus of this work is put on development of a new method to grow single crystalline GaN layers from Ga vapour. Our approach exploits microwave (MW) plasma as a source of excited nitrogen species, in contrast to classical physical vapour transport (PVT)-based technique, in which ammonia (NH3) serves as a source of reactive nitrogen. Novelty of MW plasma enhanced growth of GaN from vapour lies in MW nitrogen plasma formation in the vicinity of the seed, at moderate pressure (200 – 800 mbar range), and concurrent physical vapour transport of Ga to the growth zone. Simulations of the growth setup (HEpiGaN software) and of the MW plasma source (CST Microwave software) have followed the extensive investigations of material properties. The growth setup and the MW plasma source, with the resonance cavity being its crucial part, have been constructed and implemented into the existing growth reactor.
The stability of MW plasma in function of temperature and pressure has been studied along with its influence on the seed temperature, and thus on the growth conditions. Furthermore, optical emission spectroscopy (OES) has been utilized for in-situ characterization of the growth atmosphere. Studies on the interaction of Ga vapour with the nitrogen discharge were interpreted on the basis of the level structure of lower excited states of Ga.
Deposition experiments have been conducted, using sapphire seeds, GaN, AlN and AlGaN templates, while GaN single crystalline layers have been grown on sapphire and GaN templates. Characterization of GaN layers have been done by various methods, i.e. structure of layers by scanning electron microscopy (SEM), their composition by energy dispersive X-ray spectroscopy (EDX) and secondary ion mass spectrometry (SIMS), and crystal quality by high resolution X-ray diffraction (HRXRD). Results of the characterization together with outcome of OES measurements revealed importance of carbon for the sub-atmospheric MW plasma enhanced growth of GaN from vapour. In addition, this fact was confirmed by experiments in the setup with reduced carbon content. Possible routes for GaN synthesis have been discussed, with the most probable being CN-assisted GaN formation. While CN was detected in the plasma spectra, there was no evidence for the existence of GaN molecules in vapour phase.
Zur Deckung des stetig steigenden Energiebedarfs und unter Berücksichtigung des Umweltschutzes werden unter anderem Solarzellen genutzt. Defekte in Solarzellen können die Effizienz verringern. In dieser Arbeit wird das Defektverständnis von Silizium basierten Solarzellen erweitert.
Neue experimentelle Erkenntnisse konnten zu folgenden Schwerpunkten gewonnen werden:
• Siliziumnitrid- und Siliziumkarbidausscheidungen in multikristallinem Silizium
• Rekombinationsaktivität in dünnen Siliziumschichten
• Charakterisierung der Rekombinationsaktivität von Germanium als Modellmaterial für die Siliziumkristallisation
• Rekombinationsaktivität und Verspannung an Korngrenzen
Um Ergebnisse experimenteller Defektcharakterisierung auf allgemein gültige Parameter der Rekombination wie z.B. Ladungsträgerdiffusionslänge oder Oberflächenrekombinationsgeschwindigkeit zurückführen zu können, wurden zu folgenden Phänomenen Modelle entwickelt und Simulationen durchgeführt:
• Getterzonen an Korngrenzen
• Ermittlung von Diffusionslänge und Diffusionskoeffizient in Dünnschichtmaterial
• 3D-Raumladungseffekte
Die Nutzung der gewonnenen Erkenntnisse über die Rekombinationseigenschaften der Defekte liegt in einer möglicherweise zukünftigen kontrollierten Defektbeeinflussung und somit einer Effizienzverbesserung von Solarzellen.
Semiconductor nanowires, also called nanorods or nanowhiskers, are of particular interest for various applications in nanotechnology. Especially, germanium as a CMOS compatible material with its good electronic properties has gained renewed interest in recent years due to the availability of modern gate dielectrics. The present work deals with the vapor-liquid-solid growth of germanium nanowires and their characterization. The Growth has been carried out by means of molecular beam epitaxy using differently oriented germanium and silicon substrates whereas gold has been used to create metal catalyst droplets with radii of typically 100 nm and below. All stages from the substrate preparation to the final growth have been investigated in the frame of this work to find significant control parameters that influence the growth result. The droplet formation by means of gold evaporation onto the heated substrates has been investigated extensively on different substrates and for different surface preparations to identify parameters that are crucial for the resulting size distribution. Thereby sticking effects of the droplet circumference turned out to influence the radius distribution significantly. Germanium nanowires have been observed to grow preferentially along the <011> crystallographic directions on all utilized substrate orientations leading to defined possible inclinations of the wires with respect to the substrate normal. In contrast to the faceting known from silicon wires, the sidewalls mainly exhibit four flat {111} facets whereas the tip is roof shaped consisting of another two {111} facets. Different models which describe the inclined growth are presented and discussed. Furthermore, the material transport during the growth has been investigated. The nanowire length was found to be up to eight times larger than the nominal layer thickness according to the total amount of deposited germanium which is explained by surface diffusion towards the nanowires. The diffusion dominated growth regime was confirmed by length-radius-plot showing a decrease of the nanowire length at increasing radii. A temperature dependent diffusion model has been utilized to describe the observed nanowire length as a function of the substrate temperature. Beside conventional nanowires, so-called in-plane nanowires which grow along the substrate surface have been studied. Like their vertically growing counterparts, they also tend to grow along <011> in-plane directions which is particularly distinct on Ge(011) substrates. However, the fraction of nanowires which are aligned along <011> is influenced by substrate imperfections which was intentionally affected by means of wet-chemical substrate preparation. In addition to the nanowire growth, techniques for selective catalyst removal as well as for nanowire embedding in an insulating, transparent matrix have been established which can be important prerequisites for further nanowire processing in terms of electric or optoelectronic applications.
The LHC is the largest particle accelerator and storage ring in the world, used to investigate fundamentals of particle physics and to develop at the same time the technology of accelerators and detectors. Four main experiments (ATLAS, ALICE, CMS and LHCb) , located around the LHC ring, provide insight into the nature of particles and search for answers to as yet unexplained phenomena in the universe. Two proton or heavy ion beams circulate in the LHC and are brought into collision in the four experiments.
The physics potential of each experiment is determined by the luminosity, which is a ratio of the number of the events during a certain time period to the cross section of a physics process. A measurement of the luminosity is therefore essential to determine the cross section of interesting physics processes.
In addition, safe and high-quality data-taking requires stable beam conditions with almost no beam losses. Each experiment has its own detectors to measure beam losses, hereafter called machine induced background. One such detector is installed in CMS, BCM1F. Based on diamond sensors it was designed and built to measure both, the luminosity and the machine induced background. BCM1F ran smoothly during the first LHC running period from 2009-2012 and delivered valuable beam loss and luminosity information to the control rooms of CMS and LHC. At the end of 2012 the LHC was shut down to improve the performance by increasing the proton energy to 7TeV and decreasing the proton bunch spacing to 25ns. Due to the success of BCM1F an upgrade of its sensors and readout components was planned in order to fulfil the new requirements.
The upgrade of the sensors comprises a two pad instead of one pad metallization. 24 instead of the previous 8 single crystal diamond sensors were foreseen for the new BCM1F to enhance the robustness and redundancy. To instrument BCM1F, 59 sensors were electrically characterized by measuring the leakage current, signal stability and charge collection efficiency. Quality criteria were defined to select sensors for the final installation. An overview of these measurements including a summary of the results is given in this thesis. In addition, an upgraded amplifier was developed within the collaboration in 130nm CMOS technology. It has a peaking time of 7ns instead of the 22ns of the one previously installed. A BCM1F prototype comprising a two pad sensor and the upgraded amplifier was tested at the DESY-II accelerator in a 5GeV electron beam. Results of these test-beam measurements are presented in this thesis as well as simulations to interpret the measurements.
The installation of the upgraded BCM1F was completed in 2014. In 2015 BCM1F was commissioned and started to measure luminosity and machine induced background. At the end, the thesis will describe both types of measurements with the focus on machine induced background demonstrating the functionality of BCM1F.