The resource theory approach is a recently developed framework used in the study of thermodynamics for finite sized quantum systems. A simple way to describe how a quantum system evolves under energy-preserving unitaries while interacting with thermal reservoirs, has been formulated in the framework of thermal operations. Despite the conceptual simplicity of these operations, much is still unknown about their mathematical structures. In this work, we studied these mathematical structures, in an attempt to better understand how these operations, and some variants, describe quantum thermodynamics. One of the interesting phenomena that occurs in quantum thermodynamics, is something that we call super-activation. In this phenomenon, we make multiple forbidden transitions possible by combining them together. This phenomenon could, for example, be used to extract more work from two systems, while it is impossible to extract any work from the individual systems. We found conditions for when this phenomenon can occur. As a result of this, we found that qubits with trivial Hamiltonians cannot super-activate each other. On the other hand, we found a way to construct infinitely many examples of super-activation. We also investigated what happens if we combine multiples of the same forbidden transitions together. We found necessary conditions which each have to satisfy in order for this joint transition to be possible. In particular, these conditions show that this special case of super-activation can only occur in one direction. Another topic that we studied in this work is smoothed Rényi divergences. In an attempt to give an operational meaning to these quantities, we studied two special states that allow us to clarify the relation between the smoothed Rényi divergences with the possibility of a transition. These special states are the steepest state and the attest state. For a given state p, the steepest state is a state that is e-close to p in terms of trace distance and can be transformed, in the presence of a thermal bath and while conserving the total energy, to any other state that is also e-close to p. The attest state is a state that is also e-close to p in terms of trace distance, but all e-close states of p can be transformed to it. We found a way to construct this steepest state for limited values of e, and the attest state for any e.
The resource theory approach is a recently developed framework used in the study of thermodynamics for finite sized quantum systems. A simple way to describe how a quantum system evolves under energy-preserving unitaries while interacting with thermal reservoirs, has been formulated in the framework of thermal operations. Despite the conceptual simplicity of these operations, much is still unknown about their mathematical structures. In this work, we studied these mathematical structures, in an attempt to better understand how these operations, and some variants, describe quantum thermodynamics. One of the interesting phenomena that occurs in quantum thermodynamics, is something that we call super-activation. In this phenomenon, we make multiple forbidden transitions possible by combining them together. This phenomenon could, for example, be used to extract more work from two systems, while it is impossible to extract any work from the individual systems. We found conditions for when this phenomenon can occur. As a result of this, we found that qubits with trivial Hamiltonians cannot super-activate each other. On the other hand, we found a way to construct infinitely many examples of super-activation. We also investigated what happens if we combine multiples of the same forbidden transitions together. We found necessary conditions which each have to satisfy in order for this joint transition to be possible. In particular, these conditions show that this special case of super-activation can only occur in one direction. Another topic that we studied in this work is smoothed Rényi divergences. In an attempt to give an operational meaning to these quantities, we studied two special states that allow us to clarify the relation between the smoothed Rényi divergences with the possibility of a transition. These special states are the steepest state and the attest state. For a given state p, the steepest state is a state that is e-close to p in terms of trace distance and can be transformed, in the presence of a thermal bath and while conserving the total energy, to any other state that is also e-close to p. The attest state is a state that is also e-close to p in terms of trace distance, but all e-close states of p can be transformed to it. We found a way to construct this steepest state for limited values of e, and the attest state for any e.
Resource theories are a set of tools, coming from the field of quantum information theory, that find applications in the study of several physical scenarios. These theories describe the physical world from the perspective of an agent, who acts over a system to modify its quantum state, while having at disposal a limited set of operations. A noticeable example of a physical theory which has recently been described with these tools is quantum thermodynamics, consisting in the study of thermodynamic phenomena at the nano-scale. In the standard approach to resource theories, it is usually the case that the constraints over the set of available operations single out a unique resource. In this thesis, we extend the resource theoretic framework to include situations where multiple resources can be identified, and we apply our findings to the study of quantum thermodynamics, to gain a better understanding of quantities like work and heat in the microscopic regime. We introduce a mathematical framework to study resource theories with multiple resources, and we explore under which conditions these multi-resource theories are reversible. Furthermore, we investigate the interconversion of resources, i.e., in which situations it is possible to exchange between two different kinds of resources. We then apply this formalism to quantum thermodynamics, where the two resources under consideration are energy and entropy. This multi-resource theory allows us to explore thermodynamics when the system under examination is closed or coupled with a thermal environment with a finite size. In addition to our work on multi-resource theories, we study the states of equilibrium of closed systems, known as passive states, and analyse under which circumstances it is possible to extract energy from these states. We show that passivity is energetically unstable, and that, even for closed systems, the only stable states are those with a well-defined temperature.
Quantum thermodynamics is a blossoming research program that uses tools and ideas from quantum information theory to extend the laws of thermodynamics to the domain of systems to which the laws of quantum mechanics apply and the thermodynamic limit does not necessarily apply. The contributions of this program are both foundational, in providing an approach to constructively derive the laws of phenomenological thermodynamics from the postulates of unitary quantum mechanics in a rigorous and bottom-up fashion, but also prac- tical, in providing the theory for an increasing number of experiments that attempt to build thermodynamic machines at scales that were previously inaccessible. One key mathematical tool used in quantum thermodynamics is the framework of resource theories, specifically the resource theory of thermal operations, in which the thermodynamic interaction of a system with a heat bath and various additional components such as work batteries, clocks or catalysts are modeled. In this cumulative thesis, various extensions and modifications of this framework are studied in order to derive novel results and insights about the possible thermodynamic evolution of quantum systems. In particular, following a system- atic exposition of the resource theory of thermal operations from first principles, I develop answers to the following questions: i) What is the smallest heat bath required to provide the necessary randomness for a system to undergo a given stochastic or thermodynamical evolution? ii) Does this size differ depending on whether the interaction between bath and system is quantum or classical? iii) How can quantum systems thermodynamically evolve in the presence of catalytic bystander systems and how can this be put to use? iv) What are the thermodynamic state transitions that an agent can operationally affect when she only has access to partial information about the underlying states of system and bath? v) How can we understand the emergence of the canonical ensemble in statistical thermodynamics from quantum mechanics? These questions cover a wide ground, but it will become clear that they can in fact all be discussed using the same formal tools. As such, the answers to the above questions provided in this thesis are both interesting in their own right — showing for example that there exists a gap in how efficiently randomness can be exploited quantumly as compared to classically or that catalytic bystander systems can be used to extract finite amounts of work per particle from macroscopic systems with non-vanishing probability — as well as illustrate the power of quantum thermodynamics as a set of tools to connect quantum mechanics and the theory of thermodynamics more generally.
Thermodynamics is one of the main pillars of theoretical physics, and it has a special appeal of having wide applicability to a large variety of different physical systems. However, many assumptions in thermodynamics apply only to systems which are bulk material, i.e. consisting a large number of microscopic classical particles. Due to the advancement of designing nanoscale engines, especially in the light of devices that are used today in the processing of quantum information, is thermodynamics still applicable? Can we refine the core principles of thermodynamics to suit such nanoscale quantum systems as well? The central aim of this thesis is to construct a theory of thermodynamics that holds for nanoscale quantum systems, even those as small and simple as a single qubit. We do this by starting out from the core basics of quantum theory: unitary dynamics on closed quantum systems. We adapt a resource theoretic approach inspired by quantum information theory, which defines the quantum states and operations allowed to be used in a thermodynamic evolution. With this framework that naturally adopts the first law as an energy preserving condition, we show the refinement of both the zeroeth and second law of thermodynamics. The zeroeth law explains the physical significance of the Gibbs thermal state. On the other hand, we show that the second law sees refinement in the quantum nanoregime: instead of having the free energy as the sole quantity dictating the possibility of a thermodynamic state transition, we derive a family of generalized free energies that also constitute necessary conditions for a transition to occur. Moreover, these conditions become sufficient for states which are block-diagonal in the energy eigenbasis. In this thesis, we also brought our approach of thermodynamics to the next step: we apply our findings on the second laws, in order to analyze the maximum achievable efficiency for quantum heat engines. In classical thermodynamics, the Carnot efficiency has been long known as the theoretical maximum which does not depend on the specific structure of the thermal baths used, but only on its temperature. With the additional free energies we discover, we show that although quantum heat engines may achieve the Carnot efficiency, such an achievability is no longer independent of the Hamiltonians of the thermal baths. In other words, we find additional restrictions that surface in the study of quantum nanoscale heat engines, which are a direct consequence of the generalized second laws. This has provided us with a deeper understanding into the fundamental limitations of how efficient devices can be made in the realm of microscopic quantum systems.
Planck found, when attempting to describe the way in which hot bodies glow, that energy at microscopic scales often comes in discrete chunks. Thus began the long and intimate relationship between the field of thermodynamics, which explores our ability to manipulate heat and other energy transfers between macroscopic systems, and quantum mechanics, which explains the dynamics of individual microscopic systems. Even as both our technology and our theoretical investigations have extended to ever-smaller devices, our understanding of quantum effects on thermodynamics has remained almost exclusively limited to the quantized nature of energy. There is much more, however, to quantum theory than energy quantization. In this thesis we focus on the property of quantum coherence, the ability of quantum systems to emulate Schrodinger’s cat and somehow be neither dead nor alive, but something completely different altogether. We provide a simple and elegant formulation for the processing of coherence in thermodynamics, relying on the fact that thermodynamical processes possess an underlying time-translation symmetry. This fact allows us to quantify the ways in which coherence can play an active role, facilitating otherwise impossible thermodynamic transformations. We argue that coherence should be thought of as a distinctly quantum-mechanical thermodynamic resource. By considering an isolated quantum system connected, for some period of time, to a thermal bath, we give fundamental limitations on how coherence can be irreversibly manipulated. These limitations are related to those dictated on energy transfer by the second law of thermodynamics. Others topics that are investigated here is what is the role of correlations at the smallest scales and the conversion of quantum coherence into work. It has long been appreciated that thermodynamics is subtly interlinked with the notion of information. This work provides evidence that, to apply the laws of thermodynamics to the smallest systems around us, we must develop deeper insights into the nature of quantum information.
Quantum information science is at the interface of physics and information theory, with applications including quantum computing and communication. Within this science, a resource theory is a detailed study of a particular class of physical processes, under which the possible state transitions are restricted. Given any initial physical state, the possible final states are also restricted. Initial states with a wealth of possible final states can then be considered more valuable than less versatile initial states, leading to a notion of "resourcefulness" of states. The broad goals of a resource theory include (1) to identify, characterize, and quantify resources; (2) to find an efficient way to determine whether an arbitrary instance of resource conversion is possible; and (3) to relate practical applications to resource conversion. In this thesis, we will take the resource-theoretic approach to understanding thermodynamics. The mathematical methods used in this work will enable us to also develop resource theories for measurement uncertainty and reference frames; in fact, the resource-theoretic treatment of thermodynamics is a combination of these two. Our approach is based on a model called thermal operations, applicable on individual quantum-mechanical systems out of thermal equilibrium. We study this model in detail, resulting in a better understanding of the resource of thermal inequilibrium. In a simplified version of the model where quantum coherence is neglected, we find that the conditions for state transitions are already more complex than the classical second law of thermodynamics. We extend the model to quantum systems controlled by classical circuitry. We also extend it to incorporate quantum coherence in the low-temperature limit. Measurement uncertainty is an important element in quantum mechanics. But its current understanding is through particular measures of uncertainty, such as the variance. We apply the resource theory approach to develop a measure-independent formalism for uncertainty, even in the case where the measurement is on part of a composite system with quantum correlations. Under physical processes restricted by some symmetry (e.g. rotational symmetry), asymmetric states act as resourceful reference frames. In this resource theory, we find the rate at which large quantities of resources can be interconverted through measurements.
Part 1: Why can we not distinguish between pure, non-orthogonal quantum states? Regarding the quantum state as a state of knowledge rather than something physically real, has the potential to answer this question and explain other quantum properties but such interpretations have recently been undermined. This important no-go result, due to Pusey, Barrett and Rudolph, makes use of a specific example of a task we term state exclusion. Here, a system is prepared in a state chosen from a known set and the aim is to determine a preparation that has not taken place. We formulate state exclusion as a semidefinite program, using it to investigate when exclusion is conclusively possible and how it can be achieved. Based on state exclusion, we construct a communication task which exhibits drastic, `infinite', separations between a variety of classical and quantum information and communication complexity measures. This serves to requisition the aforementioned foundational result for use in information theoretic protocols. Part 2: What does thermodynamics look like in the absence of the thermodynamic limit? In recent years there has been a concerted effort to apply techniques from quantum information theory to study the laws of thermodynamics at the nano-scale. This has led to the resource theory of thermal operations for determining when single-copy transformations are possible. However, if a deterministic transition is forbidden, can it occur probabilistically? Here we compute and bound the maximum probability with which nano-scale thermodynamical transformations can occur. Thermal operations assume that one can precisely manipulate all of the degrees of freedom in a very large heat bath. While this enables the derivation of ultimate limits on nano-scale thermodynamics, it does not make them feasible to perform in reality. We show how allowed transitions can be implemented whilst manipulating only a single bath-qubit, making thermal operations more experimentally palatable.