Work is one of the cornerstones of classical thermodynamics. However, a direct transfer of this concept to quantum systems has proved problematic, especially for non-equilibrium processes. Unlike in the classical case, quantum work cannot be defined unambiguously. Depending on the specific setting and the imposed assumptions, different definitions are well motivated. In particular, in quantum thermodynamics, a clear distinction must be made between the measurement, storage, and use of work, since these three facets of the concept are not necessarily compatible with each other. The present thesis is mainly concerned with the measurement aspect. With the help of illustrative scenarios several approaches to quantum work measurements, their advantages and drawbacks are discussed. The focus will be on the question to what extent quantumness plays a decisive role in such scenarios, both in a qualitative and quantitative sense. Based on the gedankenexperiment of a Szilárd machine a criterion is proposed which can be used to verify genuine quantum correlations between the work medium in a heat engine and its thermal environment. In a Szilárd scenario a Maxwell's demon determines the state of the work medium and uses this information to extract work. We split this model into a bipartite setting. The demon only has access to the environment and, thus, can only indirectly measure the state of the work medium. By sharing the acquired information with another agent, the latter can extract work. The question of the quantumness of the experiment can then be reduced to the question of the maximum attainable work in the context of a suitable quantum steering scenario. For the constructed setting a bound for the work output achievable for classical correlations between the engine and the environment is derived. Work extraction beyond this classical limit thus proves the quantum nature of the machine. The verification of non-classical correlations by means of quantum steering is motivated by the fact that such a scenario reflects the typical asymmetry of a thermodynamic setup. While the machine itself is considered to be controllable and characterized in detail, no requirements are imposed on the correlated environment and the measurements performed on it. Consequently, this verification of a truly quantum heat engine is semi-device-independent. In a second scenario, the compatibility of average work and work fluctuations in a driven system is discussed. Fluctuation theorems play an important role in classical non-equilibrium thermodynamics. The best-known example is the Jarzynski equality. This equation establishes a connection between the free energy difference of two equilibrium states and the fluctuating work measured in a non-equilibrium process. A transfer of the Jarzynski equality to quantum systems succeeds most simply if the work definition is based on a so-called two-point measurement scheme. This approach determines the work as the difference of two projective energy measurements. The disadvantage of this definition is the unavoidable disturbance of the quantum state by the measurement, which makes a determination of the correct average work impossible. By means of a generalized two-point measurement scheme, it is shown how this contradiction between fluctuating and average work can be overcome. The approach is based on the concept of joint measurability. Unsharp measurements with a smaller disturbance of the quantum state can be measured jointly and allow for the determination of the correct average work. Nevertheless, the connection between measured fluctuations and the change of free energy can be preserved by means of a modified Jarzynski equality, as elucidated in this thesis. Even though the two-point measurement scheme - both in its projective form and in the generalized variant presented in this thesis - satisfies a Jarzynski equality, the operationality and the associated experimental significance are to be assessed differently than in the classical case. In classical thermodynamics, the Jarzynski relation can be used practically to determine, for example, the change of free energy in RNA molecules. However, it is crucial for such an experiment that the non-equilibrium work can actually be measured without requiring detailed knowledge of the system under consideration. In contrast, the two-point measurement scheme defines work as the energy difference of the system between the beginning and the end of the process. Crucially, for the measurement of these energies the Hamiltonians have to be known and the free energy difference could therefore be calculated directly from this knowledge without reference to the Jarzynski equality. Thus, the operationality of the quantum Jarzynski relation differs fundamentally from its classical counterpart. In this thesis we develop a measurement scheme which, in principle, allows us to employ a quantum version of the Jarzynski equation without knowledge of the Hamiltonians. The crucial point is to include the apparatus that drives the system out of equilibrium in the quantum picture and to define the work measurement on that very apparatus. Such a work measurement can only be meaningfully defined as a quantum expectation value and work fluctuations cannot directly be measured, in contrast to the classical case. The work along a classical microstate trajectory can be determined in a single run. The trajectory itself does not need to be known for this purpose; its existence is sufficient. Quantum trajectories do not exist unless they are objectified by a measurement. It is shown how measurements on the environment of the system can provide information about the trajectories. A conditioning of the measured work on these trajectories then allows for the determination of work fluctuations in the quantum system. For these fluctuations an inequality is conjectured whose limit is given by the classical Jarzynski equation. Numerical results support the conjecture. A proof is still missing. By means of the presented framework, the free energy difference of a quantum system can, in principle, be determined without knowledge of the underlying Hamiltonian. However, as is shown, this requires an optimization over several external parameters, since the inequality in general provides only an upper bound. Thus, the operationality of the model enforces a quantum disadvantage. The methods presented in this thesis can be applied to various scenarios in quantum thermodynamics. Especially the framework for work measurements on an external apparatus offers an alternative to common approaches when the system under investigation and especially its Hamiltonian is not known in advance. The focus on operationality will help to better understand to what extend the work quantities defined and measured in quantum thermodynamic systems differ from the classical concept of work.

Aufgrund des weltweiten Trends zur Miniaturisierung, u. a. von elektronischen Bauteilen, von Sensoren, von Speichermedien, oder bei der gezielten Funktionalisierung von Nanopartikeln als Kontrastmittel in bildgebenden medizinischen Verfahren, nimmt die Erforschung von Nanosystemen eine interdisziplinäre Schlüsselrolle ein. Ein grundlegendes physikalisches, chemisches und biologisches Verständnis von Nanosystemen auf Grundlage von experimentellen und theoretischen Untersuchungen steht dabei ebenso im Fokus wie die konzeptionelle Entwicklung geeigneter Nanotechnologien zur kontrollierten Herstellung von Nanostrukturen über „bottom-up“ und „top-down“ Strategien. Getriebene Nanosysteme befinden sich fern vom thermischen Gleichgewicht. Zur ihrer Beschreibung gibt es bisher keine allgemein ausgearbeitete Theorie. Dies hat zur Konsequenz, dass getriebene Nanosysteme problemspezifisch modelliert und untersucht werden müssen. Die vorliegende Schrift ist in drei Themengebiete unterteilt: (i) konzeptionelle Beschreibung stochastischer Fluktuationen der Arbeit und der Wärme im Rahmen der stochastischen Thermodynamik, (ii) konzeptionelle Beschreibung von Vielteilchen-Transportproblemen mit repulsiven Nächste-Nachbarwechselwirkungen auf Grundlage der klassischen zeitabhängigen Dichtefunktionaltheorie und (iii) selbstorganisiertes Wachstum von metallischen und organischen Nanostrukturen.

Informationsthermodynamik ist der aktuelle Trend in der statistischen Physik. Es ist die theoretische Konstruktion eines einheitlichen Rahmens für die Beschreibung der Nichtgleichgewichtsmerkmale stochastischer dynamischer Systeme, wie die Dissipation der Arbeit und die Irreversibilität von Trajektorien, unter Verwendung der Sprache der Fluktuationstheoreme und der Informationstheorie. Die modellunabhängige Natur von Information und Irreversibilität ermöglicht eine breite Anwendbarkeit der Theorie auf allgemeinere (nichtphysikalische) Modelle aus der Systembiologie und der quantitativen Finanzmathematik, in denen asymmetrische Wechselwirkungen und Nichtlinearitäten allgegenwärtig sind. Insbesondere interessieren wir uns für Zeitreihe, die aus Messungen gewonnen werden oder aus einer Zeitdiskretisierung kontinuierlicher Modelle resultieren. In dieser Arbeit untersuchen wir die Irreversibilität von Zeitreihen unter Berücksichtigung der statistischen Eigenschaften ihrer Zeitumkehrung, und leiten daraus ein Fluktuationstheorem ab, das für Signal-Antwort-Modelle gilt, und das Irreversibilität sowie bedingte Informationen mit der Vergangenheit verknüpft. Interagierende Systeme tauschen kontinuierlich Informationen aus und beeinflussen sich gegenseitig. Intuitiv ist der kausale Einfluss der Effekt dieser Wechselwirkungen, der im Hinblick auf den Informationsfluss über die Zeit beobachtet werden kann, aber seine quantitative Definition wird in der Fachgemeinschaft immer noch diskutiert. Wir wenden insbesondere das Schema der partiellen Informationszerlegung (PID) an, das kürzlich definiert wurde, um synergistische und redundante Effekte aus informationstheoretischen Maßen zu entfernen. Hier schlagen wir unsere PID vor und diskutieren die resultierende Definition des kausalen Einflusses für den Sonderfall linearer Signal-Antwort-Modelle. / Information thermodynamics is the current trend in statistical physics. It is the theoretical research of a unified framework for the description of nonequilibrium features of stochastic dynamical systems like work dissipation and the irreversibility of trajectories, using the language of fluctuation theorems and information theory. The model-independent nature of information and irreversibility allows a wide applicability of the theory to more general (nonphysical) models from systems biology and quantitative finance, where asymmetric interactions and nonlinearities are ubiquitous. In particular, we are interested in time series obtained from measurements or resulting from a time discretization of continuous models. In this thesis we study the irreversibility of time series considering the statistical properties of their time-reversal, and we derive a fluctuation theorem that holds for time series of signal-response models, and that links irreversibility and conditional information towards past. Interacting systems continuously share information while influencing each other dynamics. Intuitively, the causal influence is the effect of those interactions observed in terms of information flow over time, but its quantitative definition is still under debate in the community. In particular, we adopt the scheme of partial information decomposition (PID), that was recently defined in the attempt to remove synergistic and redundant effects from information-theoretic measures. Here we propose our PID, and motivate the resulting definition of causal influence for the special case of linear signal-response models. The thermodynamic role of causal influences can only be discussed for time series of linear signal-response models in the continuous limit, and its generalization to general time series remains in our opinion the open problem in information thermodynamics.