Über die Dynamik des Charge-Transfer-Exzitons

In the year 1947 the three physicists John Bardeen, William Bradford Shockley and Walter Houser Brattain connected two metal wires to a germanium platelet. Doing so, they were able to change the conductivity of the p-conducting zone by changing the electric voltage. This was not only the birth of the first bipolar transistor, but also marked the beginning of the age of modern microelectronics. Since these days the miniaturisation of semiconductor devices keeps going, but now reaches its limits. Moore's law, i.e. the observation that the number of transistors in a dense integrated circuit doubles every 18 months, contradicts the laws of quantum mechanics and looses its status as a self-fulfilling prophecy increasingly. Reaching scales of 2-3 nm, i.e. approximately 10 atoms in a usual semiconductor crystal, quantum mechanical effects like tunnelling play a key role and aren't negligible. Transistors don't work reliable anymore. One assumes, that these scales are reached in the year 2020. But already since 2004 there is a stagnation in the clock frequency of microprocessors. Decreasing the scale of a processor and simultaneously increasing the clock frequency unavoidably leads to heating problems. That's the reason, why manufacturers restrict the clock frequency and build multicore processors as an alternative. However, this approach needs a parallelisation of the algorithms, that are going to be performed, in order to increase calculation speed. With miniaturisation physical effects at the interface between different semiconductor layers become important. Nobel laureate Herbert Kroemer once said "the interface is the device". This statement is valid for various semiconductor devices, like transistors, laser diodes or photo detectors. An approach in order to examine the interaction between light and semiconductors driven by basic research is therefore necessary. Modern growth techniques like molecular beam epitaxy (MBE) and molecular vapour phase epitaxy (MOVPE) are able to create semiconductor layers with nanometer precision. Such heterostructures form an ideal model system in order to investigate the influence of the interface. These so called quantum wells deliver unique electrical as well as optical properties, which are a field of research since the 1980s. In contrast to bulk semiconductors, which show a square root dependency for the density of states, quantum wells exhibit a discrete steplike density of states due to their two dimensional nature. But this is only one characteristic improving the performance of semiconductor devices. Hence, quantum wells serve as an active material in numerous devices. Laser diodes, building the backbone of todays optical communication networks, are probably the most important application. But also infrared detectors, saturable absorbers, thermoelectrical elements, tandem solar cells and high electron mobility transistors (HEMTs) form further applications, which already have or will influence technological progress. In order to examine the influence of the interface on the performance of a device, type II heterostructures are the right choice. In the band structure of a type I heterostructure the energetically lowest states for the electrons and holes are located in the same quantum well. As opposed to this, in the band structure of a type II heterostructure the lowest states for electrons and holes are located in different quantum wells. If one now excites the direct transition of such a sample, either electrons or holes tunnel into the other quantum well (according to which charge carrier finds a lower level in a neighbouring quantum well). Due to the Coulomb interaction these now spatially separated charge carriers can form new quasi particles, known as excitons. As there was a charge transfer before the formation of these excitons they are called charge transfer excitons. A comprehensive experimental and theoretical understanding of those charge transfer excitons is considered as \grqq the holy grail of interface physics\grqq. Among others, the reason for this is that the dynamics of those particles mirrors the morphology of the interface. This work is devoted to examine the dynamics of the charge transfer excitons in semiconductor heterostructures. First of all, a short overview about relevant theoretical basics is given in chapter 1. In the following, there are three chapters. Each of these chapters starts with a classification of this work into the state of the art. After that, the used experimental techniques are explained as well as the data analysis. Finally, the obtained data are presented and discussed. Thereby, chapter 3 deals with the coherent dynamics of charge transfer excitons. Beginning with a description of the sample structures and their band structures this chapter gives a detailed introduction into the used method of coherent spectroscopy, i.e. four wave mixing. Using coherent spectroscopy reveals quantum beats between charge transfer excitons and direct excitons as well as the homogeneous line width/coherent life time of charge transfer excitons in a type II heterostructure. As a reference a type I heterostructure was investigated, too. The subject of chapter 4 is the behaviour of a population of charge transfer excitons in the incoherent regime. By means of THz spectroscopy a measurement of the intra-excitonic 1s-2p transition becomes possible. After a detailed depiction of the underlying experimental techniques the obtained data are presented and discussed. Especially, the formation and recombination dynamics of a population of charge transfer excitons under different excitation conditions builds the main part of this chapter. Again, the type I heterostructure was used in order to enable a comparison. Chapter 5 then deals with the field ionisation of charge transfer excitons with strong single cycle THz pulses. The interaction of excitons with strong electrical fields exhibits various many body effects like Rabi oscillations and multi THz photoionisation. In the end, a final review summarises the most important results and discusses possible future experiments.