Refine
Has Fulltext
- yes (7)
Is part of the Bibliography
- yes (7)
Document Type
- Doctoral Thesis (5)
- Journal article (2)
Keywords
- exciton (7) (remove)
Institute
Sonstige beteiligte Institutionen
- Arizona State University, Tempe, Arizona, USA (1)
- Fraunhofer-Institute for Applied Optics and Precision Engineering IOF Jena, Germany (1)
- Friedrich Schiller University Jena, Germany (1)
- Max Planck School of Photonics Jena, Germany (1)
- National Institute for Materials Science, Tsukuba, Japan (1)
- University of Oldenburg, Germany (1)
- University of Science and Technology of China, Hefei, China (1)
Die vorliegende Arbeit befasste sich mit dem Spin- und dem damit eng verbundenen Polarisationszustand von Ladungsträgern in CdSe/ZnSe Quantenpunkten. II-VI Materialsysteme können in geeigneter Weise mit dem Nebengruppenelement Mangan gemischt werden. Diese semimagnetischen Nanostrukturen weisen eine Vielzahl von charakteristischen optischen und elektrischen Besonderheiten auf. Verantwortlich dafür ist eine Austauschwechselwirkung zwischen dem Spin optisch erzeugter Ladungsträger und den 3d Elektronen der Mn Ionen. Im Rahmen dieser Arbeit erfolgte die Adressierung gezielter Spinzustände durch optische Anregung der Ladungsträger. Die Besetzung unterschiedlicher Spinzustände konnte durch Detektion des Polarisationsgrades der emittierten Photolumineszenz (PL) bestimmt werden. Dabei kamen verschiedene optische Methoden wie zeitaufgelöste und zeitintegrierte PL-Spektroskopie sowie Untersuchungen in Magnetfeldern zum Einsatz.
In dieser Dissertation wurde die Exziton- und Ladungsträgerdynamik in halbleitenden und metallischen einwandigen Kohlenstoffnanoröhren (SWNTs) mittels zeitkorreliertem Einzelphotonenzählen (TCSPC) und transienter Absorptionsspektroskopie untersucht. Die Experimente wurden an Tensid- oder DNA-stabilisierten SWNT-Proben in Suspension durchgeführt, in denen durch Dichtegradientenultrazentrifugation (DGU) halbleitende (6,5)-Röhren oder metallische (9,9)-Röhren angereichert wurden. Für die Herstellung der metallischen SWNT-Proben wurde das DGU-Verfahren optimiert. Metallische SWNT-Proben wiesen eine Verunreinigung von etwa 3% halbleitenden SWNTs auf. Von den angereicherten metallischen SWNTs war die (9,9)-Röhre mit einem relativen Anteil von 40% die vorherrschende Chiralität. Für transiente Absorptionsmessungen wurden die metallischen SWNT-Proben zudem durch Filtration aufkonzentriert. Halbleitende (6,5)-Proben wurden mit einem standardmäßig verwendeten Rezept hergestellt. Mit TCSPC-Messungen an (6,5)-Proben wurde erstmals gezeigt, dass halbleitende SWNTs neben der kurzlebigen Fluoreszenz des S1-Exzitons, die auf der ps-Zeitskala abläuft, auch eine langlebig Fluoreszenzkomponente aufweisen. Diese klingt mit t^−1 ab und stammt ebenfalls aus dem S1-Exzitonzustand. Das relative Gewicht der langlebigen Komponente an der Quantenausbeute beträgt (7 ± 2)%. Bei der langlebige Fluoreszenzkomponente handelt es sich um verzögerte Fluoreszenz. Diese entsteht durch die Wiederbesetzung des S1-Zustands aus einem tiefergelegenen Triplettzustand. Der vorherrschende Zerfall des Tripletts skaliert mit t^-0,5 und ist auf das nicht-Fick’sche Diffusionsverhalten der Tripletts zurückzuführen, die an Störstellen gefangen werden und abreagieren. Wird vor dem Übergang in den Grundzustand ein weiteres Triplett eingefangen, so kommt es zu einer Triplett-Triplett-Annihilation, die eine Wiederbesetzung des S1-Zustandes bewirkt. Für die transienten Absorptionsexperimente wurde ein Messaufbau verwirklicht, der Anregung und Abfrage im VIS und NIR Spektralbereich mit einer Zeitauflösung von bis zu 50 fs ermöglicht. Die Detektion des Abfragelichts erfolgt spektral aufgelöst mit einer CCD-Kamera. Der Aufbau ermöglicht Nachweisempfindlichkeiten von bis zu 0,2 mOD bei einer Integrationszeit von einer Sekunde. Durch unterschiedliche Modulation von Anregungs- und Abfragestrahl ist eine Detektion auf der Differenzfrequenz der Modulationen möglich, wodurch Einflüsse des Anregungslichts im Abfragespektrum effizient unterdrückt werden. In transienten Absorptionsexperimenten wurde die Exziton- und Ladungsträgerdynamik der (9,9)-Röhre untersucht. Die transienten Absorptionsdaten wurden mit einer globalen Fitroutine angepasst, der ein Vierniveausystem zugrunde lag. Aus dem globalen Fit sind die Photoanregungsspektren (PAS) - die Beiträge der drei angeregten Niveaus zu den transienten Absorptionsspektren - sowie die Zerfallszeiten zugänglich. Die PAS sind durch die Exzitonresonanz gekennzeichnet. Breite PB-Banden aufgrund der Besetzungsänderung der linearen E00-Bänder sind im Gegensatz zu transienten Absorptionsmessungen an Graphen oder Graphit nicht erkennbar. Die PAS des schnellen und mittleren Zerfalls sind ähnlich und weisen eine starkes PB-Signal bei der Energie des M1-Exzitons der (9,9)-Röhre auf, das von PA-Banden bei höheren undtieferen Energien begleitet wird. Der langsame Zerfall ist hingegen durch eine blauverschobene PB-Bande gekennzeichnet, die nur auf der niederenergetischen Seite mit einem PA-Signal einhergeht. Die Zerfallszeiten nehmen mit steigender Anregungsleistung zu und liegen im Bereich von 30 fs bis 120 fs, 500 fs bis 1000 fs und 40 ps. Die schnelle Zerfallskomponente wird mit der Dissoziation der Exzitonen sowie der Thermalisierung der freien Ladungsträgen in den linearen Leitungsbändern zu einer heißen Ladungsträgerverteilung assoziiert. Die mittlere Zerfallskomponente beschreibt die Abkühlung und Rekombination der freien Elektronen und Löcher. Entscheidender Mechanismus ist hierbei die Streuung an hochenergetischen optischen Phononmoden. Die langsame Zerfallskomponente kann durch langlebige, wahrscheinlich an Störstellen gefangene Ladungsträger erklärt werden, deren elektrische Felder durch den Stark-Effekt das ableitungsähnliche transiente Absorptionsspektrum erzeugen. Mittels transienter Absorptionsmessungen an (6,5)-Röhren wurde aus dem anregungsleistungsabhängigen maximalen PB-Signal des S1-Exzitons die Größe des S1-Exzitons zu (7,2 ± 2,5) nm bestimmt. Aus dem Vergleich der leistungsabhängigen maximalen PB-Signale bei Anregung in das S1- und das S2-Exziton ergibt sich, dass die Konversionseffizienz aus dem S2- in den S1-Zustand 1 ± 0,1 beträgt und innerhalb der experimentellen Zeitauflösung von 60 fs vollständig abläuft. Die Exzitongröße in metallischen (9,9)-Röhren wurde bei Exzitonlebensdauern von 15 fs bis 30 fs zu etwa 7 nm bis 12 nm abgeschätzt.
In the context of this dissertation very long ranged exciton diffusion lengths (LD) were simulated for perylene-based materials under ideal conditions. This leads to the conclusion that the short LD values in existing materials result from an extrinsic and intrinsic immobilization. The latter, which is a specific material property, is based on a relaxation of the exciton into self-trapping states. An in-depth understanding of the atomistic processes defining self-trapping is essential to developing materials with long LD in the future, in which intrinsic immobilization is prevented. For the development of such a mechanistic understanding it is crucial that a clear relationship between molecular structure and LD is available. This is given by single crystals of diindeno perylene (DIP) and α-perylene tetracarboxylic anhydride (α-PTCDA). An extraordinary large LD of 90 nm was measured for the first one, while the latter possesses only 22 nm. Part of this thesis was to deliver reasons for this discrepancy. Only self-trapping comes into question to explain the different LD values. One reason for the different self-trapping in DIP and α-PTCDA could lie in the electronic structure. However, it was possible to demonstrate that a wide range of perylene-based materials possess no significant differences in their electronic structures. Consequently, such differences can be neglected for the explanation of immobilization mechanisms for the exciton. A further possible explanation could be polarization effects in the crystal, which influences the electronic structure of perylene based materials differently. Especially their influence on charge transfer (CT) states, which are located above the optically bright Frenkel state, was in question because such states could be stabilized by a polarizable surrounding. A significant influence of polarization effects on all considered states were excluded by using a polarizable continuum model. Hence, the small LD values in α-PTCDA are an evidence for self-trapping, which produces a crystal structure built up by π-stacks, while the one of DIP is of herringbone type. Since polarization effects can be neglected, is the dimer only via steric restrictions influenced by the crystal. Hence, a method describing self-trapping has to consider such effects, so that a mechanical embedding QM/MM approach is sufficient. Now, potential energy surfaces were calculated, on which wave packet dynamics were subsequently performed. In this way, atomistic mechanisms for the immobilization of excitons were described for the first time in organic materials. Self-trapping was studied in crystals of α-PTCDA by potential energy surfaces, which map an intermolecular shift motion of the dimer in the crystal. An immobilization of excitons occurs within 500 fs, which results from an irreversible energy loss together with a local deformation of the crystal lattice. This prevents a further transport of the exciton. In the case of DIP, this immobilization does not proceed due to high barriers. These barriers result from the herringbone type packing motif in the DIP crystal. This discrepancy in the dynamics explains the different LD values in DIP and α-PTCDA. In a further example, an exciton immobilization was found in helical π-aggregates of perylene tetracarboxylic bisimide (PBI) molecules. Self-trapping is caused by a relaxation mechanism, in which the exciton is transferred by asymmetric vibrations of the aggregate from the bright to a dark Frenkel state within 200 fs, whereby the transition is mediated by a CT state. However, the CT state is almost non-populated during the whole mechanism so that its participation could not yet be proven experimentally. This entire procedure is solely possible in helical aggregates, because only for such structures is there a CT state located next to the bright Frenkel state. At the final Frenkel state a torsional motion around the π-stacking axis is possible so that the loss in energy and the local rearrangement of the aggregate structure occurs, which means a self-trapping of the exciton. This mechanism is in perfect agreement with all available experimental data. These insights allow the conclusion that in future materials for organic solar cells an irreversible and ultrafast deformation of aggregates after photo-absorption must be avoided. Only in this way long LD values can be achieved and exciton self-trapping can be prevented. However, small LD values are always predicted in helical aggregates of perylene-based materials, because exciton immobilization occurs already due to small molecular motions. For this reason such aggregates are inappropriate for the use in organic solar cells. Long LD values are expected for aggregate structures with long intermolecular shifts or molecules with bulky substituents.
As organic semiconductors gain more importance for application, research into their properties has become necessary. This work investigated the exciton and charge transport properties of organic semiconducting crystals. Based on a hopping approach, protocols have been developed for the calculation of Charge mobilities and singlet exciton diffusion coefficients. The protocols do not require any input from experimental data except for the x-ray crystal structure, since all needed quantities can be taken from high-level quantum chemical calculations. Hence, they allow to predict the transport properties of yet unknown compounds for given packings, which is important for a rational design of new materials. Different thermally activated hopping models based on time-dependent perturbation theory were studied for the charge and exciton transport; i. e. the spectral overlap approach, the Marcus theory, and the Levich-Jortner theory. Their derivations were presented coherently in order to emphasize the different levels of approximations and their respective prerequisites. A short reference was made to the empirical Miller-Abrahams hopping rate. Rate equation approaches to calculate the stationary charge carrier mobilities and exciton diffusion coefficients have been developed, which are based on the master equation. The rate equation approach is faster and more efficient than the frequently used Monte Carlo method and, therefore, provides the possibility to study the anisotropy of the transport parameters and their three-dimensional representation in the crystal. The Marcus theory, originally derived for outer sphere electron transfer in solvents, had already been well established for charge transport in organic solids. It was shown that this theory fits even better for excitons than for charges compared with the experiment. The Levich-Jortner theory strongly overestimates the charge carrier mobilities and the results deviate even stronger from the experiment than those obtained with the Marcus theory. The latter contains larger approximations by treating all vibrational modes classically. The spectral overlap approach in combination with the developed rate equations leads to even quantitatively very good results for exciton diffusion lengths compared to experiment. This approach and the appendant rate equations have also been adapted to charge transport. The Einstein relation, which relates the diffusion coefficient with the mobility, is important for the rate equations, which have been developed here for transport in organic crystals. It has been argued that this relation does not hold in disordered organic materials. This was analyzed within the Framework of the Gaussian disorder model and the Miller-Abrahams hopping rate.
We present a theoretical study on exciton–exciton annihilation (EEA) in a molecular dimer. This process is monitored using a fifth-order coherent two-dimensional (2D) spectroscopy as was recently proposed by Dostál et al. [Nat. Commun. 9, 2466 (2018)]. Using an electronic three-level system for each monomer, we analyze the different paths which contribute to the 2D spectrum. The spectrum is determined by two entangled relaxation processes, namely, the EEA and the direct relaxation of higher lying excited states. It is shown that the change of the spectrum as a function of a pulse delay can be linked directly to the presence of the EEA process.
Excitons in atomically thin transition-metal dichalcogenides (TMDs) have been established as an attractive platform to explore polaritonic physics, owing to their enormous binding energies and giant oscillator strength. Basic spectral features of exciton polaritons in TMD microcavities, thus far, were conventionally explained via two-coupled-oscillator models. This ignores, however, the impact of phonons on the polariton energy structure. Here we establish and quantify the threefold coupling between excitons, cavity photons, and phonons. For this purpose, we employ energy-momentum-resolved photoluminescence and spatially resolved coherent two-dimensional spectroscopy to investigate the spectral properties of a high-quality-factor microcavity with an embedded WSe\(_2\) van-der-Waals heterostructure at room temperature. Our approach reveals a rich multi-branch structure which thus far has not been captured in previous experiments. Simulation of the data reveals hybridized exciton-photon-phonon states, providing new physical insight into the exciton polariton system based on layered TMDs.
A plethora of novel material concepts are currently being investigated in the condensed matter research community. Some of them hold promise to shape our everyday world in a way that silicon-based semiconductor materials and the related development of semiconductor devices have done in the past. In this regard, the last decades have witnessed an explosion of studies concerned with so called ‘’quantum materials’’ with emerging novel functionalities. These could eventually lead to new generations of electronic and/or spintronic devices. One particular material class, the so called topological materials, play a central role. As far as their technological applicability is concerned, however, they are still facing outstanding challenges to date.
Predicted for the first time in 2005 and experimentally verified in 2007, two-dimensional topological insulators (2D TIs) (a.k.a. quantum spin Hall insulators) exhibit the outstanding property of hosting spin-polarized metallic states along the boundaries of the insulating 2D bulk material, which are protected from elastic single-particle backscattering and give rise to the quantum spin Hall effect (QSHE). Owing to these peculiar properties the QSHE holds promise for dissipationless charge and/or spin transport. However, also in today’s best 2D TIs the observation of the QSHE is still limited to cryogenic temperatures of maximum 100 K. Here, the discovery of bismuthene on SiC(0001) has marked a milestone towards a possible realization of the QSHE at or beyond room-temperature owing to the massively increased electronic bulk energy gap on the order of 1 eV. This thesis is devoted to and motivated by the goal of advancing its synthesis and to build a deeper understanding of its one-particle and two-particle electronic properties that goes beyond prior work.
Regarding the aspect of material synthesis, an improved growth procedure for bismuthene is elaborated that increases the domain size of the material considerably (by a factor of ≈ 3.2 - 6.5 compared to prior work). The improved film quality is an important step towards any future device application of bismuthene, but also facilitates all further basic studies of this material.
Moreover, the deposition of magnetic transition metals (Mn and Co) on bismuthene is investigated. Thereby, the formation of ordered magnetic Bi-Mn/Co alloys is realized, their structure is resolved with scanning tunneling microscopy (STM), and their pristine electronic properties are resolved with scanning tunneling spectroscopy (STS) and photoemission spectroscopy (PES). It is proposed that these ordered magnetic Bi-Mn/Co-alloys offer the potential to study the interplay between magnetism and topology in bismuthene in the future.
In this thesis, a wide variety of spectroscopic techniques are employed that aim to build an understanding of the single-particle, as well as two-particle level of description of bismuthene's electronic structure. The techniques involve STS and angle-resolved PES (ARPES) on the one hand, but also optical spectroscopy and time-resolved ARPES (trARPES), on the other hand. Moreover, these experiments are accompanied by advanced numerical modelling in form of GW and Bethe-Salpeter equation calculations provided by our theoretical colleagues. Notably, by merging many experimental and theoretical techniques, this work sets a benchmark for electronic structure investigations of 2D materials in general.
Based on the STS studies, electronic quasi-particle interferences in quasi-1D line defects in bismuthene that are reminiscent of Fabry-Pérot states are discovered. It is shown that they point to a hybridization of two pairs of helical boundary modes across the line defect, which is accompanied by a (partial) lifting of their topological protection against elastic single-particle backscattering.
Optical spectroscopy is used to reveal bismuthene's two-particle elecronic structure. Despite its monolayer thickness, a strong optical (two-particle) response due to enhanced electron-hole Coulomb interactions is observed. The presented combined experimental and theoretical approach (including GW and Bethe-Salpeter equation calculations) allows to conclude that two prominent optical transitions can be associated with excitonic transitions derived from the Rashba-split valence bands of bismuthene. On a broader scope this discovery might promote further experiments to elucidate links of excitonic and topological physics.
Finally, the excited conduction band states of bismuthene are mapped in energy and momentum space employing trARPES on bismuthene for the first time. The direct and indirect band gaps are succesfully extracted and the effect of excited charge carrier induced gap-renormalization is observed. In addition, an exceptionally fast excited charge carrier relaxation is identified which is explained by the presence of a quasi-metallic density of states from coupled topological boundary states of domain boundaries.