the Cu–Zn distribution at the beginning of the experiment (frozen
in at 573 K) should be the same as that at the end of the
experiment (573 K). It confirms the suggestion made that the heat
capacity differences seen in figure 2 are solely due to ordering/disordering
processes. Using the enthalpy of ordering (
DH
) values
of table 5, equilibrium
DC
ord
P
values were first calculated by fitting
the
DH
ord
T values to a second order polynomial in T and differentiating
this with respect to temperature. From the so calculated
DC
ord
P
, the entropy of ordering (DS
ord
) was then derived by integration
of
DC
ord
P
/T over the temperature intervals listed in table 5.
The most ordered and most disordered samples were studied by
low-temperature calorimetry, too. No heat capacity differences
could be observed between T = 5 K and 300 K. The entropy of
ordering/disordering derived from the DSC measurements can thus
be associated solely to the change in the configurational entropy
and does not contain any vibrational parts. To discuss these circumstances
in more detail, let us first consider a sample, which
was quenched from 573 K (marked with subscript 1). Its entropy
at 573 K (S
S
573
1
=R ¼ S
0
1
573
1
) is given by the following equation: