True nature of the Curzon-Ahlborn efficiency.

The Curzon-Ahlborn efficiency has long served as the definite upper bound for the thermal efficiency at maximum output power, and has thus shaped the development of finite-time thermodynamics. In this paper, we repeal the ruling consensus according to which it has a genuine universal character that can be derived from linear irreversible thermodynamics. We demonstrate that the Curzon-Ahlborn efficiency should instead properly be associated with a particular case of nonlinear heat engines, and we derive a generalized expression for the efficiency at maximum power beyond the restrictive case of linear models.

Revisiting Feynman's ratchet with thermoelectric transport theory.

We show how the formalism used for thermoelectric transport may be adapted to Smoluchowski's seminal thought experiment, also known as Feynman's ratchet and pawl system. Our analysis rests on the notion of useful flux, which for a thermoelectric system is the electrical current and for Feynman's ratchet is the effective jump frequency. Our approach yields original insight into the derivation and analysis of the system's properties. In particular we define an entropy per tooth in analogy with the entropy per carrier or Seebeck coefficient, and we derive the analog to Kelvin's second relation for Feynman's ratchet. Owing to the formal similarity between the heat fluxes balance equations for a thermoelectric generator (TEG) and those for Feynman's ratchet, we introduce a distribution parameter γ that quantifies the amount of heat that flows through the cold and hot sides of both heat engines. While it is well established that γ = 1/2 for a TEG, it is equal to 1 for Feynman's ratchet. This implies that no heat may be rejected in the cold reservoir for the latter case. Further, the analysis of the efficiency at maximum power shows that the so-called Feynman efficiency corresponds to that of an exoreversible engine, with γ = 1. Then, turning to the nonlinear regime, we generalize the approach based on the convection picture and introduce two different types of resistance to distinguish the dynamical behavior of the considered system from its ability to dissipate energy. We finally put forth the strong similarity between the original Feynman ratchet and a mesoscopic thermoelectric generator with a single conducting channel.

From local force-flux relationships to internal dissipations and their impact on heat engine performance: the illustrative case of a thermoelectric generator.

We present an in-depth analysis of the sometimes understated role of the principle of energy conservation in linear irreversible thermodynamics. Our case study is that of a thermoelectric generator (TEG), which is a heat engine of choice in irreversible thermodynamics, owing to the coupling between the electrical and heat fluxes. We show why Onsager's reciprocal relations must be considered locally and how internal dissipative processes emerge from the extension of these relations to a global scale: The linear behavior of a heat engine at the local scale is associated with a dissipation process that must partake in the global energy balance. We discuss the consequences of internal dissipations on the so-called efficiency at maximum power, in the light of our comparative analyses of exoreversibility and endoreversibility on the one hand and of two classes of heat engines, autonomous and periodically driven, on the other hand. Finally, basing our analysis on energy conservation, we also discuss recent works which claim the possibility to overcome the traditional boundaries on efficiency imposed by finite-time thermodynamics in thermoelectric systems with broken time-reversal symmetry; this we do by introducing a "thermal" thermopower and an "electrical" thermopower which permits an analysis of the thermoelectric response of the TEG considering a possible dissymmetry between the electrical/thermal and the thermal/electrical couplings.

Efficiency at maximum power of thermally coupled heat engines.

We study the efficiency at maximum power of two coupled heat engines, using thermoelectric generators (TEGs) as engines. Assuming that the heat and electric charge fluxes in the TEGs are strongly coupled, we simulate numerically the dependence of the behavior of the global system on the electrical load resistance of each generator in order to obtain the working condition that permits maximization of the output power. It turns out that this condition is not unique. We derive a simple analytic expression giving the relation between the electrical load resistance of each generator permitting output power maximization. We then focus on the efficiency at maximum power (EMP) of the whole system to demonstrate that the Curzon-Ahlborn efficiency may not always be recovered: The EMP varies with the specific working conditions of each generator but remains in the range predicted by irreversible thermodynamics theory. We discuss our results in light of nonideal Carnot engine behavior.

Irreversibilities and efficiency at maximum power of heat engines: the illustrative case of a thermoelectric generator.

Energy conversion efficiency at maximum output power, which embodies the essential characteristics of heat engines, is the main focus of the present work. The so-called Curzon and Ahlborn efficiency η(CA) is commonly believed to be an absolute reference for real heat engines; however, a different but general expression for the case of stochastic heat engines, η(SS), was recently found and then extended to low-dissipation engines. The discrepancy between η(CA) and η(SS) is here analyzed considering different irreversibility sources of heat engines, of both internal and external types. To this end, we choose a thermoelectric generator operating in the strong-coupling regime as a physical system to qualitatively and quantitatively study the impact of the nature of irreversibility on the efficiency at maximum output power. In the limit of pure external dissipation, we obtain η(CA), while η(SS) corresponds to the case of pure internal dissipation. A continuous transition between from one extreme to the other, which may be operated by tuning the different sources of irreversibility, also is evidenced.

Two-dimensional electron gas with universal subbands at the surface of SrTiO(3).

As silicon is the basis of conventional electronics, so strontium titanate (SrTiO(3)) is the foundation of the emerging field of oxide electronics. SrTiO(3) is the preferred template for the creation of exotic, two-dimensional (2D) phases of electron matter at oxide interfaces that have metal-insulator transitions, superconductivity or large negative magnetoresistance. However, the physical nature of the electronic structure underlying these 2D electron gases (2DEGs), which is crucial to understanding their remarkable properties, remains elusive. Here we show, using angle-resolved photoemission spectroscopy, that there is a highly metallic universal 2DEG at the vacuum-cleaved surface of SrTiO(3) (including the non-doped insulating material) independently of bulk carrier densities over more than seven decades. This 2DEG is confined within a region of about five unit cells and has a sheet carrier density of ∼0.33 electrons per square lattice parameter. The electronic structure consists of multiple subbands of heavy and light electrons. The similarity of this 2DEG to those reported in SrTiO(3)-based heterostructures and field-effect transistors suggests that different forms of electron confinement at the surface of SrTiO(3) lead to essentially the same 2DEG. Our discovery provides a model system for the study of the electronic structure of 2DEGs in SrTiO(3)-based devices and a novel means of generating 2DEGs at the surfaces of transition-metal oxides.