Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: ENERGY WITH Q



At 07:56 AM 10/26/01 -0700, John Mallinckrodt wrote:
I can't think of any specific situations
in which knowing the "correct" absolute value of the entropy is
important. Can anybody suggest one?

What's wrong with the spin-glass example I mentioned the other day?

A box of coins is easy to visualize, and gives an accurate picture of
what's going on.

The corresponding physical system is a solid containing some spin-1/2
nuclei. Get it cold enough so that mechanical vibrations (thermal phonons)
are frozen out, so that all the remaining entropy is in the nuclear spins.
At high temperatures the entropy per mole will be R log 2. Now start
cooling the sample down, and keep track of the heat capacity. If the spins
undergo an ordering transition (i.e. the system does not become a spin
glass), there will be a spike in the heat capacity, as the spins unload
their entropy. The area under the spike will be TdS. If you cool it all
the way down to some super-low temperature _without_ seeing the entropy
come out, you can assume the entropy is still in the sample at T=0, and
you've got a spin glass. You can confirm this by looking at the magnetic
properties.