Third Law of Thermodynamics

Download as pdf or txt
Download as pdf or txt
You are on page 1of 4

4

Third Law of Thermodynamics


The third law of thermodynamics is the latest of the three laws of thermodynamics to
be developed. It insures that the entropy remains well-defined at the absolute zero of
temperature and allows one to define a zero of entropy that is consistent with statistical
mechanics. This avoids having to deal with entropy differences; instead, we can deal with
entropies as absolute quantities, analogous to absolute temperature but unlike energy.

4.1 Statement of the Third Law


The entropy S of a thermodynamic system in internal equilibrium approaches a universal
constant S0 , independent of phase, as the absolute temperature T tends to zero. Alter-
natively, one could say that S → S0 in a state for which the quantity (∂U/∂S){ext} → 0,
where {ext} stands for the remaining members of a complete set of extensive variables.
By convention, and in agreement with statistical mechanics, the value of this universal
constant S → S0 is taken to be zero. Since entropy is a monotonically increasing function
of temperature, this convention results in the entropy being a positive quantity.

4.1.1 Discussion of the Third Law


According to statistical mechanics, as motivated by Eq. (3.77), the entropy of an isolated
system is given by
S = kB ln , (4.1)

where kB is Boltzmann’s constant and  is the number of microstates that correspond to a


given macrostate. If at absolute zero only a unique ground state of the system is occupied,
then  = 1 and S = 0. Possibly the ground state could be degenerate, in which case
 = 1 even at T = 0. But this degeneracy would have to be massive to make a significant
difference in the entropy of a macroscopic system at T = 0. Indeed, to get a contribution
S = 10−10 R = 10−10 kB NA for one mole at absolute zero would require the ground state
degeneracy 0 to satisfy 10−10 NA = ln 0 , where NA is Avogadro’s number. This yields
13 13
0 ∼ e6×10 ∼ 102.6×10 . But such a huge degeneracy is contrary to experience. As the
ground state of a quantum system is approached (as T → 0), the number of accessible
quantum states decreases quite rapidly and is no longer of exponential order, even though
there could still be a ground state of much smaller degeneracy. An illuminating discussion
of this point has been presented by Benjamin Widom [17, chapter 5].
The third “law” is an extension by Max Planck [15, p. 273] of the so-called Nernst
postulate [2, p. 277] that was made in an attempt to justify an empirical rule of Thomsen
and Berthelot for chemical reactions that take place at constant temperature and pressure.
Thermal Physics. http://dx.doi.org/10.1016/B978-0-12-803304-3.00004-1 49
Copyright © 2015 Elsevier Inc. All rights reserved.
50 THERMAL PHYSICS

Nernst conjectured that their empirical rule for equilibrium, which is equivalent to min-
imizing the enthalpy change H of the reaction, would be in agreement with the proper
thermodynamic criterion obtained by minimizing an appropriate change in free energy1
of the reaction, provided that the entropy change S tends to zero as T → 0. This can be
interpreted to mean that the entropy S itself tends to some constant, independent of the
extent of the reaction, as T → 0. For convenience, Planck set this entropy constant to zero,
which agrees with the convention used to define entropy in statistical mechanics. Callen
[2, p. 30] states the third law as an independent postulate, namely that S = 0 in a state
for which ∂U/∂S = 0 (which is true at absolute zero by definition of the thermodynamic
temperature). From the point of view of classical thermodynamics, one could deal with
entropy differences and it would not be necessary to adopt a state of zero entropy;
however, doing so leads to simplicity and builds a strong bridge to statistical mechanics.

4.2 Implications of the Third Law


The third law has certain implications regarding heat capacities and other properties of
materials as T → 0. From Eq. (3.47) with dV = 0, we obtain CV dT = T dS where CV is the
heat capacity at constant volume. The change in entropy at constant volume from one
temperature to another is given by CV /T dT. Thus
 T1 CV (T , V )
S(T1 , V ) = dT . (4.2)
0 T
In order for this integral to converge, it is necessary for CV to depend on T in such a way
that CV → 0 as T → 0. Recall that CV was taken to be a constant for an ideal gas; clearly
such an ideal gas becomes impossible as T → 0. For insulating solids, one finds both
theoretically and experimentally that CV ∝ T 3 as T → 0. For metals, nearly free electrons
contribute to the heat capacity and CV ∝ T as T → 0. Similar considerations apply to the
heat capacity at constant pressure. From Eq. (3.62) with dp = 0, we obtain Cp dT = T dS,
where Cp is the heat capacity at constant pressure. Thus
 T1 Cp (T , p)
S(T1 , p) = dT (4.3)
0 T

and it is necessary2 for Cp → 0 as T → 0.


An interesting experimental verification of the third law has been discussed by Fermi
[1, p. 146]. At temperatures below T0 = 292 K, gray tin (α, diamond cubic) is the stable
form and above this temperature, white tin (β, tetragonal) is stable. These are allotropic
forms of pure tin. It turns out, however, that white tin can exist (in internal equilibrium)
below 292 K, even though it is unstable with respect to transformation to gray tin. It is also

1 Thisis the change G of the Gibbs free energy of the reaction, whose definition and properties we explore
later.
2 In order for the integrals in Eqs. (4.2) and (4.3) to converge at T = 0, it will suffice for C or C to go to zero
V p
very weakly as T → 0, for instance ∝ T  where  > 0.
Chapter 4 • Third Law of Thermodynamics 51

White

Gray

0K 292 K T
FIGURE 4–1 Entropies S of gray and white tin as a function of absolute temperature T. Below T0 = 292 K, gray
tin is stable and above this temperature white tin is stable. The full curves denote stable phases and the dashed
curves denote unstable phases. White tin can be supercooled below T0 so its heat capacity can be measured and its
entropy can be calculated. The jump in entropy at T0 between gray tin and white tin is due to the latent heat of
transformation.

possible to measure the heat capacities of both forms of tin down to very low tempera-
tures. One can therefore evaluate the entropy of white tin at 292 K in two different ways,
the first by integrating its heat capacity from absolute zero and the second by integrating
the heat capacity of gray tin from absolute zero and then adding the entropy associated
with transformation to white tin at 292 K. See Figure 4–1 for a graphic illustration. Thus
(with subscripts g and w for gray and white), we have
 292 K Cw (T )
Sw (292 K ) = dT = 12.30 cal/mol K, (4.4)
0 T
and
 292 K Cg (T )
Sg (292 K ) = dT = 10.53 cal/mol K. (4.5)
0 T
The heat of transformation from gray to white tin is Hg→w = 535 cal/mol so the entropy
of transformation is Sg→w = Hg→w /T0 = 535/292 = 1.83 cal/mol K. Adding this to the
result of Eq. (4.5) gives 12.36 cal/mol K, in reasonable agreement with Eq. (4.4).
The third law can also shed light on the behavior of the coefficient of thermal expan-
sion, α, and the compressibility, κT , as T → 0. Since S → 0 as T → 0 independent of V or
p, one has
   
∂S ∂S
= 0; = 0. (4.6)
∂V T =0 ∂p T =0

Through a Maxwell relation (see Eq. (5.90)), it can be shown that


   
∂S ∂V
=− = −V α, (4.7)
∂p T ∂T p

where α is the coefficient of isobaric thermal expansion. Indeed, it has been verified
experimentally that α → 0 as T → 0. By means of another Maxwell relation (see Eq. (5.86))
52 THERMAL PHYSICS

   
∂S ∂p
= = α/κT , (4.8)
∂V T ∂T V

where κT is the coefficient of isothermal compressibility. Thus, κT must either remain non-
zero as T → 0 or go to zero more slowly than α.
See Lupis [5, pp. 21-23] for further discussion of experimental verification of the third
law as well as a discussion of some of its other consequences, particularly consequences
concerning chemical reactions. See Fermi [1, p. 150] for an excellent discussion of the
entropy of mercury vapor.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy