346 ENGINEERING THERMODYNAMICS

!# ∂h $&
To find " ∂p % ; let h = f(s, p)
T

!# ∂h $& ! ∂h $
Then, dh = " ∂s % ds + #" ∂p &% dp
p s

!# ∂h $& ! ∂h $ ! ∂s $ ! ∂h $
" ∂p % = #" ∂s &% #" ∂p &% + #" ∂p &%
T p T s

But
!# ∂h $& ! ∂s $ ! ∂v $ ! ∂h $
= T, # & = – # & , # & =v
" ∂s % p
" ∂p % T
" ∂p % " ∂p %
p s

Hence !# ∂h $& !# ∂v $&

From eqn. (7.31), we get

dh = cp dT +
')v − T ! ∂v $ +) dp ...(7.33)
() #" ∂T &% ,)
* p -
7.6. MEASURABLE QUANTITIES
Out of eight thermodynamic properties, as earlier stated, only p, v and T are directly
measurable. Let us now examine the information that can be obtained from measurements of
these primary properties, and then see what other easily measurable quantities can be introduced.
The following will be discussed :
(i) Equation of state
(ii) Co-efficient of expansion and compressibility
(iii) Specific heats
(iv) Joule-Thomson co-efficient.
7.6.1. Equation of State
Let us imagine a series of experiments in which the volume of a substance is measured over
a range of temperatures while the pressure is maintained constant, this being repeated for various
pressures. The results might be represented graphically by a three-dimensional surface, or by a
family of constant pressure lines on a v-T diagram. It is useful if an equation can be found to
express the relation between p, v and T, and this can always be done over a limited range of states.
No single equation will hold for all phases of a substance, and usually more than one equation is
required even in one phase if the accuracy of the equation is to match that of the experimental
results. Equations relating p, v and T are called equations of state or characteristic equations.
Accurate equations of state are usually complicated, a typical form being
B C
pv = A +
+ + ......
v v2
where A, B, C, ...... are functions of temperature which differ for different substances.
An equation of state of a particular substance is an empirical result, and it cannot be
deduced from the laws of thermodynamics. Nevertheless the general form of the equation may be

dharm
\M-therm\Th7-1.pm5
THERMODYNAMIC RELATIONS 347

predicted from hypotheses about the microscopic structure of matter. This type of prediction has
been developed to a high degree of precision for gases, and to a lesser extent for liquids and solids.
The simplest postulates about the molecular structure of gases lead to the concept of the perfect
gas which has the equation of state pv = RT. Experiments have shown that the behaviour of real
gases at low pressure with high temperature agrees well with this equation.
7.6.2. Co-efficient of Expansion and Compressibility
From p-v-T measurements, we find that an equation of state is not the only useful informa-
tion which can be obtained. When the experimental results are plotted as a series of constant
pressure lines on a v-T diagrams, as in Fig. 7.1 (a), the slope of a constant pressure line at any
∂v
!# $&
given state is ∂T" % . If the gradient is divided by the volume at that state, we have a value of a
p
property of the substance called its co-efficient of cubical expansion β. That is,

Fig. 7.1. Determination of co-efficient of expansion from p-v-T data.

1 ∂v
!# $&
β= ...(7.34)
v ∂T" % p
Value of β can be tabulated for a range of pressures and temperatures, or plotted graphically
as in Fig. 7.2 (b). For solids and liquids over the normal working range of pressure and tempera-
ture, the variation of β is small and can often be neglected. In tables of physical properties β is
usually quoted as an average value over a small range of temperature, the pressure being atmos-
pheric. This average co-efficient may be symbolised by β and it is defined by
v2 − v1
β = v (T − T ) ...(7.35)
1 2 1
Fig. 7.2 (a) can be replotted to show the variation of volume with pressure for various
∂v !# $&
constant values of temperature. In this case, the gradient of a curve at any state is ∂p . When
" %
T
this gradient is divided by the volume at that state, we have a property known as the compressibility
K of the substance. Since this gradient is always negative, i.e., the volume of a substance always
decreases with increase of pressure when the temperature is constant, the compressibility is
usually made a positive quantity by defining it as

dharm
\M-therm\Th7-1.pm5
348 ENGINEERING THERMODYNAMICS

Fig. 7.2. Determination of compressibility from p-T data.

1 ∂v
!# $&
K=– ...(7.36)
v ∂p" % T
K can be regarded as a constant for many purposes for solids and liquids. In tables of
properties it is often quoted as an average a value over a small range of pressure at atmospheric
temperature, i.e.,
v2 − v1
K =–
v1 ( p2 − p1 )
When β and K are known, we have
!# ∂p $& !# ∂T $& !# ∂v $& = – 1
" ∂T % " ∂v % " ∂p%
v p T

Since
!# ∂v $& = βv and !# ∂v $& = – Kv,
" ∂T % p " ∂p % T

!# ∂p $& = β ...(7.37)
" ∂T % K
v
When the equation of state is known, the co-efficient of cubical expansion and compressibility
can be found by differentiation. For a perfect gas, for example, we have
!# ∂v $& R !# ∂v $& RT
= and =
" ∂T % p
p " ∂p % T p2

!# $& 1 ∂v R 1
Hence β= = = ,
" % v ∂T p pv T
1 ! ∂v $ RT 1
and K=– # & = = .
v " ∂p % T
p2v p
7.6.3. Specific Heats
Following are the three differential co-efficients which can be relatively easily determined
experimentally.

dharm
\M-therm\Th7-1.pm5