[Fis] Multiple Szilard Engine Models

Dear Shu-Kun and colleagues,

I think we may be referring to different versions of Szilard�s engine.
You wrote that �Mixing of different ideal gases (oxygen and nitrogen)
originally
separated in two parts of a rigid container at the same and unchanged T
and P cannot generate any mechanical work. This is for sure. If
Szilard's Engine works, the 2nd law will be no law at all. Of course
Szilard's discussion contributed tremendously to information theory.�
In his publication (Z. Phys. 53 1929, p. 840) Szilard described several
models for engines. By far, the model that has received the most notice,
is that of the partitioned, single-molecule engine. (I would guess that
scores of authors have tried to explain it�s behavior.) This is the
model I also analyzed, and have found to be of such extraordinary value
in uncovering the work-entropy-information relationship. It�s also the
model used by Bennett and others as prototype of an information bit
register.
As I calculate the entropy cost of partition movement in this version of
the engine, I�m still excited by what it can teach about time-signal
information and quantum measurement. I need to make this calculation to
determine an entropy balance for a complete engine cycle, which, I�m
convinced, won�t indicate any challenge to the Second Law.
This Szilard engine model is thoroughly impractical, but, I don�t
believe there are any conceptual hindrances to it�s operation, and so,
production of work. I�m unaware of anyone who has said this version of
the engine violates the Second Law (unless you are referring to the same
model I am). Instead, it teaches about the entropy cost of information
processing.
I recall a little about another engine model Szilard proposed, sometimes
called the membrane model. Perhaps that�s the one you describe. I read
the most recent analysis proposed by Leff and Rex for that membrane
engine, and didn�t like their conclusion. So, it would please me,
anyway, if it�s this model engine you say violates the Second Law.
Cordially,
Michael Devereux

_______________________________________________
fis mailing list
fis@listas.unizar.es
http://webmail.unizar.es/mailman/listinfo/fis
Received on Thu Apr 22 22:04:00 2004