🚀 go-pugleaf

On Jan 20, 10:21 am, Pentcho Valev <pva...@yahoo.com> wrote:
> On Jan 18, 2:58 pm, blackboab <blackb...@gmail.com> wrote in
> sci.physics:
>
> > The second law of thermodynics says that the direction of time is
> > dictated by the increase of entropy,.
>
> Does it? Why don't you read Jos Uffink's paper:
>
> http://philsci-archive.pitt.edu/archive/00000313/
> Jos Uffink: "This summary leads to the question whether it is fruitful
> to see irreversibility or time-asymmetry as the essence of the second
> law. Is it not more straightforward, in view of the unargued
> statements of Kelvin, the bold claims of Clausius and the strained
> attempts of Planck, to give up this idea? I believe that Ehrenfest-
> Afanassjewa was right in her verdict that the discussion about the
> arrow of time as expressed in the second law of the thermodynamics is
> actually a RED HERRING."
>
> Jos Uffink:http://www.phys.uu.nl/igg/jos/
>
> Pentcho Valev
> pva...@yahoo.com

i had looked that article. it is interesting, but any consideration of
thermodynamics that not relate this to statistical mechanics seems to
me out of target. and it get lost in mathematical considerations about
things as for example
when a process is irreversible or not. it get lost, because if you not
look at the microscopic level, then you cannot look for a basis from
which assume good sense. take a look to the article and u will see
that.

I think that, the main point is that in practice does not exist
adiabatic systems at all, you can aproach a lot, but u can´t construct
a truly adiabatic system. so any system in practice is not
adiabatically isolated, then u always are in the canonical ensemble
(or grand canonical ensemble), so always the systems become perturbed
from the outside, and that leads to irreversible systems.

the only in principle possible adiabatically isolated system is the
entire universe, but we can do until now very basics measurements as
for example the background radiation temperature. in fact, the system
(the entire universe) is not in thermodynamic equilibrium (at least if
it is expanding), so all the stuff there, about classical
thermodynamics must be taken with care.

a truly adiabatically isolated system, at least for me, is one in
which the microcanonical ensemble holds, because otherway if the
system interchange heat with some reservoir  (although it stay in
equilibrium with the reservoir)  by means of fluctuations, then once
again the mechanical dynamics of the system is perturbed by the noise
introduced by the fluctuations induced by the heat reservoir, as the
detailed dynamical mechanics of the reservoir is unknown (in
principle), then those perturbations lead to a loose of the knowledge
of the initial condition of the mechanical dynamics of the system, in
other words, suppose u know completely the state of the system at the
microlevel at some instant, then we let the system interact with the
heat reservoir (but in such a way that the system initially assumed in
equilibrium has the same temperature that the heat reservoir, so both
are in thermodynamical equilibrium when they both are put in contact)
after a while we will lost the knowledge of the state of the system at
the microlevel, because the system trayectory is perturbed by the heat
reservoir. in this way, we start with zero ignorance (zero entrophy)
an finalize with non zero entrophy
after a long time, we reach the entrophy value of the system in
equilibrium with the heat reservoir, the value stated by the classical
formulation of thermodynamics.

now suppose that u reverses time, then one in principle may think that
entrophy will lower back to zero, but it will not happen, for it to
happen u must reverses time of the heat reservoir too, so all the
system plus the heat reservoir becomes a deterministic mechanical
system that returns to the initial point (system plus heat reservoir
becomes considered in a microcanonical ensemble point of view). but a
good formulation of a heat reservoir does not assume knowledge of its
particular dynamics at the microlevel. it is just a system with an
infinite heat capacity with some definite temperature.

so, in practice, all system suffer from irreversibility, because they
are not properly adiabatically isolated, although they may stay in
thermodynamic equilibrium with the medium. then, classical
formulations of thermodynamics leads to the right predictions (in the
thermodynamic limit, N --> infinity where fluctuations becomes zero
the quasi static process lead to reversibility but assuming that the
only knowledge of the starting stat of the system is the thermodynamic
one, so a infinitely negligible amount of loose of information is done
under the process, that is the area inside the pattern on the
graphical description of the process has zero area, for example in a
carnot cycle in which one goes and return on the same line, but if N
is not infinite then this area cannot be zero at all, zero area equals
negligible perturbations, in other words in the canonical ensemble
reversible process from the point of view of thermodynamical variables
are possible if N equals infinity).

so, in practice, there are two limitations to reversible process, one
is that an adiabatically isolated system is not possible, the other,
is that N never can be infinity although big.

how all this relates to a preferential time arrow direction?? well the
wanted assertion is that times follow the direction that increase
entrophy. but in the canonical ensemble (or grand canonical) any
direction of time will increase entrophy, because the noise
perturbations. so althoug (in the N equals infinity and quasi static
process) reversible process are possible, this will not force an arrow
time direction. a system out of equilibrium, will goes to equilibrium
in any direction of the arrow time, that is, in both directions
entrophy will rise. (note, that a system out of equilibrium will go to
it, but will not follow a quasi static process, so it does not matter
if it is in the N equal infinity limit or not, and if u lead the
system to the equilibrium point by means of a quasi static process,
then it never will reach that equilibrium in a finite time, in fact a
quasi static process is only possible mathematically. but just for
thinking about that. what happens if we do that at least
mathematically if we does a rime reversion in order to reverse the
process?? well, it is hard to say exactly because we will have an
ifinite nonnumerable but concatenated amount of infinite time process,
but the system will low his entrophy after all, and this lead my to
believe that the quasi static process plus N equals infinity leads to
negligible perturbations between the heat reservoir and the system, so
we in fact are not in the canonical ensemble but instead we are in the
microcanonical at all)

only in the microcanonical ensemble a time reversion will lead (though
maybe temporarily) to an entrophy decrease.
but the fact that universe entrophy seems to raise with time is not
the causes of the present time arrow direction, instead this arrow
time direction and entrophy rise are both simply connected facts. in
other words, universe present dynamics seems to rise entrophy (the
causes being the expansion among others) and it coincides with the
present arrow time direction. if a time reversal is done, then
universe will go in the oposite way (contracting until reach the big
bang point, so reversing all it process)  and his entrophy will
decrease.

in conclusion, arrow time direction and entrophy ¨direction¨  (rise or
decrease) are conected by the particular universe dynamics (that is
expansion contraction among others). pick another system (a toy one)
and this could not happen.

finally I want to say, what entrophy measures, essentially the amount
of ignorance to known the microlevel state description of the system
from some known thermodynamic variables values. in other words, given
some thermodynamics variables values, how much we can say about the
microlevel state description.

best regards
rayohauno