Equilibrium thermodynamics describes the energy exchange of a body with its environment. Here, we describe the global energy exchange of an ideal gas in the Coutte flow in a thermodynamic-like manner. We derive a fundamental relation between internal energy as a function of parameters of state. We analyze a non-equilibrium transition in the system and postulate the extremum principle, which determines stable steady states in the system. The steady-state thermodynamic framework resembles equilibrium thermodynamics.
I. INTRODUCTION
There is mounting evidence that the exchange of energy of a macroscopic steady state system with its environment can be described in an equilibrium thermodynamic-like fashion. It has been considered in the 1990s of the 20th century for systems with chemical reactions that may constantly occur in time.1–3 Such a steady state situation goes beyond equilibrium thermodynamics which excludes any macroscopic flows and currents.4 Around a similar time, the thermodynamic-like approach was introduced for systems with shear-flow5–8 and is still under development.9–12 Another class of macroscopic systems for which the attempt to introduce thermodynamic description has been undertaken is systems with heat flow.13–17
The above approaches to formulate steady state thermodynamics focus on chemical reactions and shear flow, both in homogeneous temperature profiles or systems with heat flow without chemical reactions and macroscopic flows. If steady state thermodynamics is ever formulated on general grounds, it is now at its inception. About two decades ago, Oono and Paniconi18 and Sasa and Tasaki19 postulated a thermodynamic-like description based on a general footing. However, because the descriptions were postulated, they require validation and further investigation on the eventual limitation. In particular, it is not clear whether the nonequilibrium entropy defined by the integral of the local entropy density over the volume of the system14 or rather the excess heat-based entropy18 should be used to construct the principles of steady-state thermodynamics. These two entropies are not equivalent.16 It shows that the steady state thermodynamics is far from complete, and the core fundamental questions still remain open.14,17 It is worth mentioning stochastic thermodynamics, which focuses on thermodynamic notions for the system on the level of individual trajectories.20,21 Here we focus on a reduced, macroscopic description of a steady state system.22
Allowing various macroscopic constant fluxes drives the system from equilibrium to a steady state and opens up phenomena that eludes equilibrium thermodynamics. Take for example a quiescent liquid in a uniform gravitational field in equilibrium. This situation assumes a uniform temperature and no macroscopic flows, which, together with the equations of state, determine the thermodynamic parameters at each point of the system. Allowing a vertical or horizontal heat flow radically changes the situation. The flow of heat combined with the gravitational force can cause regular mass movement, either because the denser liquid is at the top and the thinner is at the bottom or because gravity unevenly acts on areas with different horizontal densities. This unstable situation causes a mass movement in the atmosphere, oceans, planets, and stars.23 The core feature of this phenomenon is the coupling between the heat flow and the mass flow. However, the question arises whether some reduced description for non-equilibrium steady states also emerges from hydrodynamics. Can it take the shape of equilibrium thermodynamics, in which the system’s behavior is described by only a few parameters and the rules of extremum containing only these parameters?
Here we present a thermodynamic-like description of a system with heat and mass flow coupling. We consider ideal gas in shear flow with a dissipative temperature profile shown schematically in Fig. 1. We introduce the first law for this system that describes different ways of exchanging the system’s energy with its environment. We also consider a movable wall as a thermodynamic constraint in the system. We introduce the extremum principle that determines the position of the wall. We show that there is a critical shear in the system above which the equilibrium position of the internal wall becomes unstable, and the system undergoes a nonequilibrium second order phase transition. We give a complete thermodynamic-like description of this steady state system in which the position of the movable wall is a thermodynamic constraint. For the vanishing shear flow, the steady state extremum principle directly reduces to the minimum principle in the corresponding problem in thermodynamics.
II. SYSTEM
III. STEADY STATE
IV. SHEAR FLOW WITHOUT THE INTERNAL WALL
V. TRANSITION BETWEEN STEADY STATES AND ENERGY BALANCE
We consider transitions between steady states in the system. The system is in one steady state at time ti, after which we slightly change one or more control parameters that appear in (or are related to) the governing equations. For example, we slowly modify the velocity of the wall, the position of the upper wall (volume V) or the external temperature T0. In particular, we increase the velocity of the wall by for ti < t < tf, which modifies velocity of the wall from Vw to Vw + dVw. This induces a time dependent hydrodynamic field. After time tf the system reaches another steady state close to the previous one due to a small change in the control parameters. We focus on steady, nonequilibrium states with a constant number of particles in the system.
VI. NONEQUILIBRIUM ENTROPY
VII. SYSTEM WITH INTERNAL WALL
Until now, we have discussed the energy balance independently for each subsystem. But it is particularly interesting to consider a perpendicular motion of the internal wall. As follows from the dynamics, the normal component of the pressure tensor on the wall in a steady state comes solely from hydrostatic pressure. The wall may move vertically only due to the pressure difference, p1 ≠ p2. Its natural position is where these pressures are balanced. Therefore the knowledge of the effective entropy S* and volume for each subsystem is sufficient to determine the force acting on the wall. The wall’s horizontal velocity enters the problem indirectly. The shear velocity plays the role of energy dissipation. The speed of the wall appears in the energy balance Eq. (9) in the source term. It thus plays the role of volumetric heating, .
The ideal gas with volumetric heating and an internal wall has already been considered previously.15 Such a system exhibits continuous nonequilibrium phase transition. Similarly, the system with macroscopic kinetic energy shown in Fig. 1 will exhibit the phase transition as well for the following reason. Let’s focus on the position of the internal wall zw and upper wall velocity, , keeping the external walls temperature T0, the total volume of the system, L1 + L2 = L, and other parameters constant. Number of particles in both subsystems is equal, N1 = N2. Because the hydrostatic pressure gives the sole force normal to the walls, the motion of the internal wall can be determined by the fundamental relations (34) for each subsystem. The pressure follows from . Moreover, the internal energy in steady state is determined by Eq. (15). For the same value of volumetric heating, the system with volumetric heating considered in Ref. 15 and the system with shear flow considered in this paper have the same temperature, density and pressure profiles. This reasoning leads to the conclusion that from the perspective of the position of the internal wall, the steady state behavior of both subsystems is the same. The speed of the upper wall for the system with kinetic energy must be related to the volumetric heating λ from Ref. 15 by .
The appearance of a phase transition can be explained as the result of the competition of two effects. The first effect is decrease of pressure due to expansion of the subsystems and is of equilibrium nature. The opposite effect is due to the dissipation of energy, which acts as volumetric heating of the system. In shear flow, the larger the volume of the subsystem, the larger the energy dissipated, the more significant the increase of its temperature profile and the increase of the pressure in the subsystem. As a result, the effect of shear is that the expansion of a subsystem increases its pressure. For sufficiently large shear flows, the second effect dominates.
Equation (43) puts us in the same position as in equilibrium thermodynamics: if has an integrating factor and the corresponding potential, , then the condition of minimal work (40) by means of expression (43) follows the condition of the minimum of the energy for “adiabatically insulated” system defined by surface. This surface is considered in the space of T0, V1, V2, Vw, because we keep the number of particles N1, N2, and other parameters constant. Let’s notice that expressions (40)–(45) for the case of no shear flow (no heating), Vw = 0, reduces to the problem of finding the position of the wall that separates two homogeneous ideal gases. Thus, in equilibrium thermodynamics, , becomes the heat differential. The nonequilibrium situation has been discussed earlier16 and it suggests the minimum principle as follows.
In the above reasoning the minimum principle follows from the fact, that the internal energy balance is determined by dUi = − pidVi with homogeneous pressure that is solely determined by the internal energy and volume, . Both assumptions also hold when transport coefficients depend on parameters of states.17 In particular, if the viscosity or heat conductivity depends on density or temperature. For this reason, the derived minimum energy principle is also valid beyond the regime of Onsager’s linear response theory.24
VIII. SUMMARY
In this article, we investigate whether there is a description similar to equilibrium thermodynamics for out-of-equilibrium steady states. We consider this problem in the context of the Couette flow, where there is a mass current (velocity field), energy dissipation, and a continuous flow of heat. Each feature independently excludes the theory of equilibrium thermodynamics and its tools.
Since, in general, it is still unclear that such a simple, equilibrium thermodynamic-like description of nonequilibrium steady states is possible, we have chosen the most straightforward possible system, which we believe includes the coupling of mass flow and heat current, i.e., an ideal gas in shear flow. We show that nonequilibrium entropy exists, which describes how the system gains energy through excess heat or dissipation. In addition, the nonequilibrium entropy allows us to construct the principle of minimum energy, which leads to the proper state of the system after removing the constraint, which is the force acting on the wall in the system. It proves the existence of a description of a system with an out-of-equilibrium heat and mass flow, which practically takes the form of equilibrium thermodynamics and reduces to the principle of minimum energy in equilibrium thermodynamics when the shear flow disappears.
It is worth mentioning that the thermodynamics of the flowing system have been considered previously with the use of thermostated simulations.9–11 A numerical technique with a thermostat cannot be used directly to confirm relations derived in our paper, e.g., Eq. (22), because at the paper’s core is the energy balance in the system without local heat sources, which are not related to the dissipation due to shear. However, one could repeat calculations in our paper and consider the effects of the thermostat. It would produce an additional term in Eq. (22) and require the redefinition of the excess heat because the thermostat adds a volumetric heating term to the energy balance equation. A similar effect we have considered in our previous paper.16 The approach presented in our paper applies to the global energy flow from a given volume of the system during the transition between steady states. If the “given volume” would be the simulation cell, this approach could be helpful to generalize such simulation techniques as multiparticle collision dynamics26 to control the energy balance on the level of individual cells.
It is also worth noting that the description formulated in our paper uses the non-equilibrium entropy defined as the potential of the excess heat differential. This entropy differs from the non-equilibrium entropy defined as the integral of the equilibrium entropy density over the volume.16 Our result suggests that the non-equilibrium entropy defined as the excess heat differential is the one that should be utilized to construct a thermodynamic-like description for non-equilibrium steady states.
The thermodynamic-like description introduced in this paper leads to further questions inspired by equilibrium thermodynamics. Particularly interesting are the procedures for measuring non-equilibrium state parameters, effective temperature and entropy, and measuring excess heat, dissipation, and work the wall performs. It is worth noting that the obtained result goes along the line of recent experiments of Yamamoto et al., who extended calorimetry for the measurements of the excess heat of sheared fluids.27 Another issue is the possibility of developing the description of interacting systems, where the fundamental relationship for van der Waals gas with heat flow can be introduced. It is also interesting to ask about Maxwell’s identities, which in equilibrium thermodynamics appear very practical. Finally, it is fascinating to check whether such a description exists for other systems with coupled heat flow and mass current in the atmosphere and the chemical industry.
ACKNOWLEDGMENTS
P.J.Z. would like to acknowledge the support of a project that has received funding from the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie Grant Agreement No. 847413 and was a part of an international co-financed project founded from the program of the Minister of Science and Higher Education entitled “PMW” in the years 2020–2024; Agreement No. 5005/H2020-MSCA-COFUND/2019/2. We express our gratitude to Natalia Pacocha for her contribution in creating the figure.
AUTHOR DECLARATIONS
Conflict of Interest
The authors have no conflicts to disclose.
Author Contributions
Karol Makuch: Conceptualization (equal); Methodology (equal); Writing – original draft (equal); Writing – review & editing (equal). Konrad Giżyński: Methodology (supporting); Writing – review & editing (equal). Robert Hołyst: Conceptualization (equal); Methodology (equal); Writing – review & editing (equal). Anna Maciołek: Methodology (supporting); Writing – review & editing (equal). Paweł J. Żuk: Methodology (supporting); Writing – review & editing (supporting).
DATA AVAILABILITY
Data sharing is not applicable to this article as no new data were created or analyzed in this study.