Reply to E.G.D. Cohen, L. Rondoni, Physica A 306 (2002) 117
Extracted source text ยท arXiv nlin/0210059
\documentclass[aps,twoside]{revtex4}
%\usepackage{a4}
\textwidth 165mm
\def\baselinestretch{1.2}
\def\ie{\emph{i.e.,} }
\begin{document}
\thispagestyle{empty}
\parbox{\textwidth}{\hfill to appear in \emph{Physica A}}
\\
\begin{center}
{\large
Reply to the paper \\[2mm]
E.G.D. Cohen, L. Rondoni: \emph{Particles, Maps and Irreversible Thermodynamics I.}}\\[2mm]
\emph{Physica} \textbf{A 306} (2002) 117
%by\\
%
\end{center}
In \cite{CohRon02} Cohen and Rondoni severely question the physical
relevance of studies of transport by means of multibaker (MB) maps.
The aim of MB studies (as well as of other work on low-dimensional
chaotic models for transport) is to improve the understanding of
irreversible macroscopic behavior in transport processes along one
spatial dimension. These studies provide an attempt to establish a
macroscopic description consistent with irreversible thermodynamics
(IT) in systems with an \emph{explicit} microscopic dynamics. In this
class of analytically accessible model systems the properties of the
microscopic and macroscopic motion are fully understood such that
their mutual relation can be studied in detail. We believe that it is
admissible to abandon the use of certain concepts of traditional
approaches in order to highlight the general features of irreversible
processes.
The fundamental interest in MB studies lies in identifying minimal
assumptions on the dynamics and on the form of entropy functionals
that lead to a description of the evolution of macroscopic fields
according to the laws of IT --- in particular, to the transport
equations and to the local entropy balance. From this perspective the
arguments of Cohen and Rondoni concerning the assumptions adopted in
the setup of multibaker models are contradictory and part of them do
not hold:
\begin{itemize}
\item %
Rondoni and Cohen \cite[Sect IV]{CohRon02} proclaim that %
'{\sffamily it is the lack of real particles that makes local
thermodynamic equilibrium (LTE) impossible}'.
%
We disagree with this statement since LTE only ascertains that the
thermodynamic description of an equilibrium system applies
\emph{locally} in sufficiently small volumes of a non-equilibrium
system. Since thermodynamics is a phenomenological theory, however,
it does not involve the concept of particles. By definition of its
basic objects the notion of particles can not be needed for its
foundation. Hamiltonian equations of motion or kinetic theory can
be used to underpin it in certain (dilute gas) limits, but IT is
more general and can even describe the evolution of systems with a
different underlying dynamics.
Similar to the Lorentz gas the time evolution of an ensemble of
trajectories in a multibaker map can be interpreted as an assembly
of non-interacting particles. By adopting a Boltzmann-Grad type
limit one can then concentrate on the evolution of fields
characterizing the state of the system. The characterization of
these fields forms the prime objective of studies on multibaker
maps.
% Instead of postulating the existence of particles, the dynamics of
% our version of the MB map involves the evolution of fields which
In the absence of coarse-graining the fields typically develop
structures on ever refining scales as times evolves. Since their
evolution is reversible and they never stop evolving, the
fields are viewed as characteristics of a microscopic motion.
% rather than as thermodynamic fields.
%\item
On the other hand, the averages of the fields over MB cells of
linear extension $a$ in the transport direction (called
coarse-grained quantities in what follows) are thermodynamically
well behaved. Appropriate averages can be interpreted as density
$\varrho$, temperature $T$, and hydrodynamic velocity $v$. They are
considered as local thermodynamic variables, since in a well-defined
continuum limit their time evolution \emph{exactly} follows the
transport equations (the advection-diffusion equation, the heat
conduction equation, and the Navier-Stokes equation of strongly
viscous fluids) of IT. Consequently, the cell size $a$ is
considered as the mesoscopic scale addressed by LTE, below which
macroscopic concepts do not apply.
%\item %
Information-theoretic arguments are evoked to work out the entropy
balance for a coarse-grained entropy $S$ taking the form
\begin{equation}
S = - k_B \varrho \ln\left( \varrho\; T^{-\gamma} \right),
\label{eq:S}
\end{equation}
where $k_B$ is the Boltzmann constant, and $\gamma$ is a positive
constant playing the role of a dimensionless specific heat. We have
shown for a series of transport problems (electric \cite{EarlyWork}
and heat conduction, cross effects \cite{Matyas}, and viscous flows
\cite{MTV:PRE01}) that the local entropy balance of this $S$ is
exactly of the form known from IT. In view of this, (\ref{eq:S}) can
be interpreted as an equation of state, which holds in \emph{every
cell} in spite of the spatial and temporal evolution of $\varrho$
and $T$. This is in our interpretation the analog of LTE. It has
nothing to do with the concept of particles and their collisions.
\item %
We agree with Cohen and Rondoni that the expression (\ref{eq:S})
adopted for the coarse-grained entropy is not obvious, and that it
is also not clear a priori how one can identify \emph{heat} for a
model lacking the concept of work and particles. In our early
papers \cite{EarlyWork,Matyas} we therefore avoided the use of the term
heat, and spoke mainly of entropy and entropy production.
%
% For studies on purely diffusive transport one can not hope to do
% better since (in contrast to an assertion in Cohen and Rondoni's
% paper) there is no immediate relation in that case between entropy
% production and heat. Indeed one easily checks from
% \cite[CH.XI,\,\,\S7]{dGM} that for such systems the time derivative of the
% temperature can be written as a divergence, \ie it does not involve
% a source term that can be related to dissipative heating.
%
In the visco-thermal MB, on the other hand, a clear relation emerged
\cite{MTV:PRE01}. After coarse graining, the energy balance
condition leads to the temperature equation
\[
\nu \varrho \partial_x T = \lambda \partial^2_x T + Q
\]
with
\[
Q= \nu \varrho (\partial_x v)^2,
\]
where $\nu$ is the kinematic viscosity (a unique expression of the
MB parameters) which coincides in the model with the heat diffusion
coefficient, and $\lambda = \nu \varrho \gamma$ represents the heat
conductivity. The source term $Q=\nu \varrho (\partial_x v)^2$
turns out to be the same as the contribution to the entropy
production due to viscosity multiplied by the temperature. The full
entropy production takes the form
\[
\sigma^{(irr)} = \lambda \left( \frac{\partial_x T}{T} \right)^2
+ \frac{\nu \varrho}{T} (\partial_x v)^2.
\]
This strong analogy with IT encouraged us to call $Q$ the viscous
heating term, since it indeed raises the temperature due to the
local hydrodynamic momentum exchange of neighboring fluid layers,
modelled by neighboring MB cells.
\item %
Concerning the purely diffusive dynamics of the Lorentz gas or an
isothermal MB,
%\cite{EarlyWork,DorfGasp},
which can both be considered as particle models of transport,
Rondoni and Cohen state that their entropy production only
\emph{formally} agrees (in an appropriate limit) with the
thermodynamic expression
\[
\sigma^{(irr)} = D \frac{(\nabla \varrho)^2}{\varrho}
\]
since the entropy production cannot be related to any heating
process in the absence of interacting particles
%
(in their words '{\sffamily ...\ in IT the entropy production
$\sigma_s^{IT}>0$ is due to heat, which involves randomization,
...}').
%
We mention that IT does not predict $\sigma^{(irr)}$ to be a source
term of heat in the temperature equation. Rather [see \cite{dGM}
chapter XI, eq. (235)], there is no heat source at all in this
equation, and local changes of the temperature are due to the Dufour
effect. The assumtion of constant temperature in the diffusive
multibaker only implies a vanishing Dufour coefficient. The entropy
production $\sigma^{(irr)}$ is in this case fully due to mixing
entropy induced by the gradient of the macroscopic density
(phase-space density in the particle picture), which is also present
in microscopic models of non-interacting particles.
\item %
The studies of MB models provide a self-consistent framework for
the description of transport based on a reversible dynamics. In our
eyes they clearly show that the coarse graining
carried out over the underlying chaotic and mixing MB dynamics
unavoidably leads to information loss, which appears to be the basic
ingredient of irreversible entropy production \emph{and} viscous
heating. In contrast to the main assertion of \cite{CohRon02} they
therefore provide a valuable alternative approach to study the
foundations of IT.
% , which sheds light on the prejudices involved in
% perceptions closely relying on traditional approaches like lattice
% gases and kinetic theory.
\end{itemize}
Finally, we would like to point out that our interpretation of LTE is
fully consistent with the definition given by the authors of
\cite{CohRon02}, who say:
\\[2mm]
\hspace*{0.05\textwidth} \parbox{0.9\textwidth}{{\sffamily The
validity of LTE can only be justified \emph{a posteriori}, on the
grounds of the conclusions derived from it.}}
\\[2mm]
%
We consider the recovery of the relations of IT from such a simple
model as the MB map as a clear justification for considering
%the thermodynamically consistent evolution of
the cell-averaged, coarse-grained fields as hallmarks of LTE,
and therefore reject the criticism by Cohen and Rondoni.
%\bibliographystyle{apsrev}
%\bibliography{Transport,StatPhys,Chaos,Literat}
%\end{document}
\begin{thebibliography}{10}
\expandafter\ifx\csname bibnamefont\endcsname\relax
\def\bibnamefont#1{#1}\fi
\expandafter\ifx\csname bibfnamefont\endcsname\relax
\def\bibfnamefont#1{#1}\fi
\expandafter\ifx\csname url\endcsname\relax
\def\url#1{\texttt{#1}}\fi
\expandafter\ifx\csname urlprefix\endcsname\relax\def\urlprefix{URL }\fi
\expandafter\ifx\csname bibinfo\endcsname\relax \def\bibinfo#1#2{#2}\fi
\expandafter\ifx\csname eprint\endcsname\relax \def\eprint#1{#1}\fi
\bibitem{CohRon02}
\bibinfo{author}{\bibfnamefont{E.~G.~D.} \bibnamefont{Cohen}} \bibnamefont{and}
\bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Rondoni}},
\bibinfo{journal}{Physica A} \textbf{\bibinfo{volume}{306}},
\bibinfo{pages}{117} (\bibinfo{year}{2002}).
\bibitem{EarlyWork}
\bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Vollmer}},
\bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{T\'{e}l}}, \bibnamefont{and}
\bibinfo{author}{\bibfnamefont{W.}~\bibnamefont{Breymann}},
\bibinfo{journal}{Phys. Rev. Lett.}
\textbf{\bibinfo{volume}{79}}(\bibinfo{number}{15}), \bibinfo{pages}{2759}
(\bibinfo{year}{1997});
%
%\bibitem{VTB98}
%\bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Vollmer}},
% \bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{T\'{e}l}}, \bibnamefont{and}
% \bibinfo{author}{\bibfnamefont{W.}~\bibnamefont{Breymann}},
\bibinfo{journal}{Phys. Rev. E}
\textbf{\bibinfo{volume}{58}}(\bibinfo{number}{2}), \bibinfo{pages}{1672}
(\bibinfo{year}{1998});
%
%\bibitem{BTV98}
\bibinfo{author}{\bibfnamefont{W.}~\bibnamefont{Breymann}},
\bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{T\'{e}l}}, \bibnamefont{and}
\bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Vollmer}},
\bibinfo{journal}{\uppercase{Chaos}}
\textbf{\bibinfo{volume}{8}}(\bibinfo{number}{2}), \bibinfo{pages}{396}
(\bibinfo{year}{1998}).
\bibitem{Matyas}
%\bibitem{MTV:PRE00a}
\bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{M\'{a}ty\'{a}s}},
\bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{T\'{e}l}}, \bibnamefont{and}
\bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Vollmer}},
\bibinfo{journal}{Phys. Rev. E}
\textbf{\bibinfo{volume}{61}}(\bibinfo{number}{4}), \bibinfo{pages}{R3295}
(\bibinfo{year}{2000}),
\bibinfo{note}{{\tt arXiv:chao-dyn/9912028}};
%
%\bibitem{MTV:PRE00b}
%\bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{M\'{a}ty\'{a}s}},
% \bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{T\'{e}l}}, \bibnamefont{and}
% \bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Vollmer}},
\bibinfo{journal}{Phys. Rev. E}
\textbf{\bibinfo{volume}{62}}(\bibinfo{number}{1}), \bibinfo{pages}{349}
(\bibinfo{year}{2000}),
\bibinfo{note}{{\tt arXiv:chao-dyn/9912034}};
%
%\bibitem{VTM:JSP00}
\bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Vollmer}},
\bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{T\'{e}l}}, \bibnamefont{and}
\bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{M\'{a}ty\'{a}s}},
\bibinfo{journal}{J. Stat. Phys.}
\textbf{\bibinfo{volume}{101}}(\bibinfo{number}{1/2}), \bibinfo{pages}{79}
(\bibinfo{year}{2000}),
\bibinfo{note}{{\tt arXiv:nlin.CD/0001016}}.
%\bibitem{VTB97}
%\bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Vollmer}},
% \bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{T\'{e}l}}, \bibnamefont{and}
% \bibinfo{author}{\bibfnamefont{W.}~\bibnamefont{Breymann}},
% \bibinfo{journal}{Phys. Rev. Lett.}
% \textbf{\bibinfo{volume}{79}}(\bibinfo{number}{15}), \bibinfo{pages}{2759}
% (\bibinfo{year}{1997}).
%\bibitem{VTB98}
%\bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Vollmer}},
% \bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{T\'{e}l}}, \bibnamefont{and}
% \bibinfo{author}{\bibfnamefont{W.}~\bibnamefont{Breymann}},
% \bibinfo{journal}{Phys. Rev. E}
% \textbf{\bibinfo{volume}{58}}(\bibinfo{number}{2}), \bibinfo{pages}{1672}
% (\bibinfo{year}{1998}).
%\bibitem{BTV98}
%\bibinfo{author}{\bibfnamefont{W.}~\bibnamefont{Breymann}},
% \bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{T\'{e}l}}, \bibnamefont{and}
% \bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Vollmer}},
% \bibinfo{journal}{\uppercase{Chaos}}
% \textbf{\bibinfo{volume}{8}}(\bibinfo{number}{2}), \bibinfo{pages}{396}
% (\bibinfo{year}{1998}).
%\bibitem{MTV:PRE00a}
%\bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{M\'{a}ty\'{a}s}},
% \bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{T\'{e}l}}, \bibnamefont{and}
% \bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Vollmer}},
% \bibinfo{journal}{Phys. Rev. E}
% \textbf{\bibinfo{volume}{61}}(\bibinfo{number}{4}), \bibinfo{pages}{R3295}
% (\bibinfo{year}{2000}), \bibinfo{note}{also availabel at {\tt
% arXiv:chao-dyn/9912028}}.
%\bibitem{MTV:PRE00b}
%\bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{M\'{a}ty\'{a}s}},
% \bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{T\'{e}l}}, \bibnamefont{and}
% \bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Vollmer}},
% \bibinfo{journal}{Phys. Rev. E}
% \textbf{\bibinfo{volume}{62}}(\bibinfo{number}{1}), \bibinfo{pages}{349}
% (\bibinfo{year}{2000}), \bibinfo{note}{also available at {\tt
% arXiv:chao-dyn/9912034}}.
%\bibitem{VTM:JSP00}
%\bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Vollmer}},
% \bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{T\'{e}l}}, \bibnamefont{and}
% \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{M\'{a}ty\'{a}s}},
% \bibinfo{journal}{J. Stat. Phys.}
% \textbf{\bibinfo{volume}{101}}(\bibinfo{number}{1/2}), \bibinfo{pages}{79}
% (\bibinfo{year}{2000}), \bibinfo{note}{[special issue on occasion of
% G.~Nicolis' $60$th birthday]. Also available at {\tt arXiv:nlin.CD/0001016}}.
%\bibitem{TMV:EPL01}
%\bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{T\'{e}l}},
% \bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Vollmer}}, \bibnamefont{and}
% \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{M\'{a}ty\'{a}s}},
% \bibinfo{journal}{Europhys. Lett.}
% \textbf{\bibinfo{volume}{53}}(\bibinfo{number}{4}), \bibinfo{pages}{458}
% (\bibinfo{year}{2001}), \bibinfo{note}{available at {\tt
% arXiv:nlin.CD/0009013}}.
\bibitem{MTV:PRE01}
%\bibitem{TMV:EPL01}
\bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{T\'{e}l}},
\bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Vollmer}}, \bibnamefont{and}
\bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{M\'{a}ty\'{a}s}},
\bibinfo{journal}{Europhys. Lett.}
\textbf{\bibinfo{volume}{53}}(\bibinfo{number}{4}),
\bibinfo{pages}{458} (\bibinfo{year}{2001}),
\bibinfo{note}{{\tt arXiv:nlin.CD/0009013}};
%
%\bibitem{MTV:PRE01}
\bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{M\'{a}ty\'{a}s}},
\bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{T\'{e}l}}, \bibnamefont{and}
\bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Vollmer}},
% \emph{\bibinfo{title}{A MultiBaker Map for Shear Flow and Viscous Heating}}
\bibinfo{journal}{Phys. Rev. E}
\textbf{\bibinfo{volume}{64}}, \bibinfo{pages}{056106}
(\bibinfo{year}{2001}),
\bibinfo{note}{{\tt arXiv:chao-dyn/0102013}}.
\bibitem{dGM}
\bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{de~Groot}} \bibnamefont{and}
\bibinfo{author}{\bibfnamefont{P.}~\bibnamefont{Mazur}},
\emph{\bibinfo{title}{Nonequilibrium Thermodynamics}}
(\bibinfo{publisher}{Elsevier}, \bibinfo{address}{Amsterdam},
\bibinfo{year}{1962}), \bibinfo{note}{reprinted by Dover, New York, 1984}.
\end{thebibliography}
\ \\[10mm]
\begin{tabular}{lll}
\parbox{0.36\textwidth}{
L\'aszl\'{o} M\'aty\'as\\
Max-Planck-Institute for the\\ Physics of Complex Systems\\
N\"{o}thnitzer Str.\ 38\\
01187 Dresden\\
Germany}
&
\parbox{0.36\textwidth}{
Tam\'as T\'{e}l\\
Institute for Theoretical Physics\\
E\"{o}tv\"{o}s University\\
P. O. Box 32\\
H-1518 Budapest\\
Hungary}
&
\parbox{0.3\textwidth}{
J\"{u}rgen Vollmer\\
Max-Planck-Institute for\\ Polymer Research\\
Ackermannweg 10\\
55128 Mainz\\
Germany}
\end{tabular}
\end{document}