Re: Why do we care about heat?

["John S. Denker" <jsd@MONMOUTH.COM>]

At 10:19 AM 10/31/99 -0800, John Mallinckrodt wrote:
>  Macroscopic
> processes in the real world are almost without exception irreversible ...

To my ears that statement is not just unorthodox -- it is bizarre.  I'm
trying to figure out where JM is coming from but I'm just stumped.  Totally.

If I haul a weight from height H=1m to height H=2m that is the epitome of
macroscopic work -- and there is no reason why it should not be reversible
(to an exceedingly good approximation).

> We
> calculate the change in entropy of a system by devising an imaginary
> *reversible* thermodynamic path from the initial to the final state.

What do you mean "we", Kemosabe?  Does that mean if I have gas confined to
half a container and suddenly pull out the divider, "we" cannot calculate
the change in entropy produced by the non-adiabatic, irreversible
expansion?  I bet I can.

The notion that heating has to be reversible is just so foreign to
real-world thermodynamics that it makes my head spin.

> In any event, the heat we are
> talking about here is *always* a quasistatic exchange of energy between
> two systems that occurs specifically as a result of an infinitesimal
> difference in temperature.

That statement is consistent with the previous bizarre statement, and is
consistent with some of JM's other incomprehensible viewpoints, such as
(10:30 AM 10/31/99 -0800):
> Furthermore, I think that the modern perspective of virtually all
> textbooks is that the short and long term result in the Joule experiment
> is increased internal energy, not heat.

The conventional name of that experiment is "the mechanical equivalent of
heat".  The notion that the result of this experiment was "not heat" is
beyond unorthodox.  It's beyond unconventional.  It's bizarre.

> I'd like to reserve the word heat to mean *essentially* this same thing in
> all circumstances.  I say "essentially"  because I am willing to soften my
> definition to include *nonquasistatic* exchanges between systems that
> occur as a result of *finite* temperature differences.  Otherwise, I'd
> just as soon call everything else work to make a clear distinction between
> work--which can be arbitrarily distinguished from heat for use in the
> first law--and heat--which must conform to a far more rigid definition for
> use in the second law.

What about inductive heating?  Suppose I shine a high-power microwave beam
onto a chunk of butter.  I think it heats the butter.  Does anybody really
think this should be described as work not heat?