home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!usc!zaphod.mps.ohio-state.edu!uwm.edu!spool.mu.edu!agate!ames!network.ucsd.edu!sdphu1.ucsd.edu!kezhao
- From: kezhao@sdphu1.ucsd.edu (Kezhao Zhang)
- Newsgroups: sci.physics
- Subject: Maxwell's demon: new results
- Date: 17 Nov 1992 06:59:45 GMT
- Organization: Physics Department, UCSD
- Lines: 26
- Distribution: world
- Message-ID: <1ea591INNpt3@network.ucsd.edu>
- NNTP-Posting-Host: sdphu1.ucsd.edu
-
-
- Although there have been some "proofs" of why Maxwell's demon
- cannot work (e.g. Brillouin's arguments), explicit models of Maxwell's
- demon *do* exist even in the realm of classical mechanics.
- As shown in a paper in the current issue of the Physical Review A
- (October 15, vol.46, no.8, p.4598), a purely mechanical model of Maxwell's
- demon can be constructed explicitly under the laws of classical mechanics.
- (The performance of the demon, of course, violates the second law of
- thermodynamics.) It is crucial to realize that the effects of Maxwell's
- demon on the motion of the gas particles are equivalent to a force which
- is velocity-dependent. To construct the mechanical demon, there is
- no need to consider how information about the movement of the particles
- is collected, processed, and used to control a trap-door of some kind.
- Instead, one can consider what kind of velocity-dependent force fields
- (Lorentz force doesn't work) can implement the overall effects of the demon.
- The model is simple and straightforward. Interested readers are referred
- to the above-mentioned article. If your want PostScript version of text
- (or LaTex version, if you prefer) and the PostScript figures, please send
- your request to kezhao@sdphu1.ucsd.edu. We'll send you a copy via e-mail.
-
- Kezhao Zhang
- kezhao@sdphu1.ucsd.edu
- Department of Physics
- University of California, San Diego
- La Jolla, CA 92093
-
-