Chapter 11 - Section B - Non-Numerical Solutions

11.6 Apply Eq. (11.7):

  
  
∂(nT ) ∂n ∂(n P) ∂n
T̄i ≡ =T =T P̄i ≡ =P =P
∂n i P,T,n j ∂n i T,P,n j ∂n i P,T,n j ∂n i T,P,n j

 
∂m
11.7 (a) Let m be the mass of the solution, and define the partial molar mass by: m̄ i ≡
∂n i T,P,n j

Let Mk be the molar mass of species k. Then

m=  nk Mk = ni Mi +  n j M j ( j = i)
k j
 

∂m ∂(n i Mi )
and = = Mi Whence, m̄ i = Mi
∂n i T,P,n j ∂n i T,P,n j
     
∂M t ∂M t ∂n i
(b) Define a partial specific property as: M̃i ≡ =
∂m i T,P,m j ∂n i T,P,m j ∂m i T,P,m j
 
mi ∂n i 1
If Mi is the molar mass of species i, n i = and =
Mi ∂m i T,P,m j Mi

M̄i
Because constant m j implies constant n j , the initial equation may be written: M̃i =
Mi

dV dV
11.8 By Eqs. (10.15) and (10.16), V̄1 = V + x2 and V̄2 = V − x1
d x1 d x1

dV −1 dρ
Because V = ρ −1 then = 2 whence
d x1 ρ d x1
   
1 x2 dρ 1 x2 dρ 1 dρ
V̄1 = − 2 = 1− = ρ − x 2
ρ ρ d x1 ρ ρ d x1 ρ2 d x1

   
1 x1 dρ 1 x1 dρ 1 dρ
V̄2 = + 2 = 1+ = ρ + x 1
ρ ρ d x1 ρ ρ d x1 ρ2 d x1
dρ
With ρ = a0 + a1 x1 + a2 x12 and = a1 + 2a2 x1 these become:
d x1

1 1
V̄1 = [a0 − a1 + 2(a1 − a2 )x1 + 3a2 x12 ] and V̄2 = (a0 + 2a1 x1 + 3a2 x12 )
ρ2 ρ2

679
 11.9 For application of Eq. (11.7) all mole fractions must be eliminated from the given equation by the
relation xi = n i /n:
n1n2n3
n M = n 1 M1 + n 2 M2 + n 3 M3 + C
n2

    
∂(n M) 1 2n 1 ∂n
For M̄1 , = M1 + n 2 n 3 C − 3
∂n 1 T,P,n 2 ,n 3 n2 n ∂n 1 T,P,n 2 ,n 3
 
∂n
Because n = n 1 + n 2 + n 3 , =1
∂n 1 T,P,n 2 ,n 3

n2n3  n1 
Whence, M̄1 = M1 + 1 − 2 C and M̄1 = M1 + x2 x3 [1 − 2x1 ]C
n2 n

Similarly, M̄2 = M2 + x1 x3 [1 − 2x2 ]C and M̄3 = M3 + x1 x2 [1 − 2x3 ]C

One can readily show that application of Eq. (11.11) regenerates the original equation for M. The
infinite dilution values are given by:

M̄i∞ = Mi + x j xk C ( j, k = i)

Here x j and xk are mole fractions on an i-free basis.

11.10 With the given equation and the Dalton’s-law requirement that P = i pi , then:
RT
P=
V  yi Z i
i

For the mixture, P = Z RT /V . These two equations combine to give Z = i yi Z i .

11.11 The general principle is simple enough:
Given equations that represent partial properties M̄i , M̄iR , or M̄iE as functions of com-
position, one may combine them by the summability relation to yield a mixture property.
Application of the defining (or equivalent) equations for partial properties then regenerates
the given equations if and only if the given equations obey the Gibbs/Duhen equation.
11.12 (a) Multiply Eq. (A) of Ex. 11.4 by n (= n 1 + n 2 ) and eliminate x1 by x1 = n 1 /(n 1 + n 2 ):
n 31
n H = 600(n 1 + n 2 ) − 180 n 1 − 20
(n 1 + n 2 )2
Form the partial derivative of n H with respect to n 1 at constant n 2 :


3n 21 2n 31 n 21 n 31
H̄1 = 600 − 180 − 20 − = 420 − 60 + 40
(n 1 + n 2 )2 (n 1 + n 2 )3 (n 1 + n 2 )2 (n 1 + n 2 )3

Whence, H̄1 = 420 − 60 x12 + 40 x13

Form the partial derivative of n H with respect to n 2 at constant n 1 :

2 n 31
H̄2 = 600 + 20 or H̄2 = 600 + 40 x13
(n 1 + n 2 )3

680
 (b) In accord with Eq. (11.11),
H = x1 (420 − 60 x12 + 40 x13 ) + (1 − x2 )(600 + 40 x13 )

Whence, H = 600 − 180 x1 − 20 x13

d H̄1 d H̄2
(c) Write Eq. (11.14) for a binary system and divide by d x1 : x1 + x2 =0
d x1 d x1
Differentiate the the boxed equations of part (a):
d H̄1 d H̄2
= −120 x1 + 120 x12 = −120 x1 x2 and = 120 x12
d x1 d x1
Multiply each derivative by the appropriate mole fraction and add:

−120 x12 x2 + 120x12 x2 = 0

(d) Substitute x1 = 1 and x2 = 0 in the first derivative expression of part (c) and substitute x1 = 0
in the second derivative expression of part (c). The results are:
   
d H̄1 d H̄2
= =0
d x1 x1 =1 d x1 x1 =0

(e)

11.13 (a) Substitute x2 = 1 − x1 in the given equation for V and reduce:

V = 70 + 58 x1 − x12 − 7 x13
Apply Eqs. (11.15) and (11.16) to find expressions for V̄1 and V̄2 . First,
dV
= 58 − 2 x1 − 21 x12
d x1

Then, V̄1 = 128 − 2 x1 − 20 x12 + 14 x13 and V̄2 = 70 + x12 + 14 x13

681
 (b) In accord with Eq. (11.11),

V = x1 (128 − 2 x1 − 20 x12 + 14 x13 ) + (1 − x1 )(70 + x12 + 14 x13 )

Whence, V = 70 + 58 x1 − x12 − 7 x13

which is the first equation developed in part (a).

d V̄1 d V̄2
(c) Write Eq. (11.14) for a binary system and divide by d x1 : x1 + x2 =0
d x1 d x1
Differentiate the the boxed equations of part (a):

d V̄1 d V̄2
= −2 − 40 x1 + 42 x12 and = 2 x1 + 42 x12
d x1 d x1
Multiply each derivative by the appropriate mole fraction and add:

x1 (−2 − 40 x1 + 42 x12 ) + (1 − x1 )(2 x1 + 42 x12 ) = 0

The validity of this equation is readily confirmed.

(d) Substitute x1 = 1 in the first derivative expression of part (c) and substitute x1 = 0 in the second
derivative expression of part (c). The results are:

   
d V̄1 d V̄2
= =0
d x1 x1 =1 d x1 x1 =0

(e)

11.14 By Eqs. (11.15) and (11.16):

dH dH
H̄1 = H + x2 and H̄2 = H − x1
d x1 d x1

682
 Given that: H = x1 (a1 + b1 x1 ) + x2 (a2 + b2 x2 )

dH
Then, after simplification, = a1 + 2b1 x1 − (a2 + 2b2 x2 )
d x1

Combining these equations gives after reduction:

H̄1 = a1 + b1 x1 + x2 (x1 b1 − x2 b2 ) and H̄2 = a2 + b2 x2 − x1 (x1 b1 − x2 b2 )

These clearly are not the same as the suggested expressions, which are therefore not correct. Note
that application of the summability equation to the derived partial-property expressions reproduces
the original equation for H . Note further that differentiation of these same expressions yields results
that satisfy the Gibbs/Duhem equation, Eq. (11.14), written:

d H̄1 d H̄2
x1 + x2 =0
d x1 d x1
The suggested expresions do not obey this equation, further evidence that they cannot be valid.

11.15 Apply the following general equation of differential calculus:

       
∂x ∂x ∂x ∂w
= +
∂y z ∂y w ∂w y ∂ y z



  
∂(n M) ∂(n M) ∂(n M) ∂V
= +
∂n i T,P,n j ∂n i T,V,n j ∂V T,n ∂n i T,P,n j

Whence,
       
∂M ∂V ∂M ∂V
M̄i = M̃i + n or M̃i = M̄i − n
∂V T,n ∂n i T,P,n j ∂V T,n ∂n i T,P,n j

By definition,

    
∂(nV ) ∂V ∂V
V̄i ≡ =n +V or n = V̄i − V
∂n i T,P,n j ∂n i T,P,n j ∂n i T,P,n j

 
∂M
Therefore, M̃i = M̄i + (V − V̄i )
∂V T,x

11.20 Equation (11.59) demonstrates that ln φ̂i is a partial property with respect to G R /RT . Thus ln φ̂i =
Ḡ i /RT . The partial-property analogs of Eqs. (11.57) and (11.58) are:

∂ ln φ̂i V̄i R ∂ ln φ̂i H̄iR

= and =−
∂P RT ∂T RT 2
T,x P,x

The summability and Gibbs/Duhem equations take on the following forms:

GR
RT
=  xi ln φ̂i and i xi d ln φ̂i = 0 (const T, P)
i

683
11.26 For a pressure low enough that Z and ln φ are given approximately by Eqs. (3.38) and (11.36):
BP BP
Z =1+ and ln φ =
RT RT

then: ln φ ≈ Z − 1

11.28 (a) Because Eq. (11.96) shows that ln γi is a partial property with respect to G E/RT , Eqs. (11.15)
and (11.16) may be written for M ≡ G E/RT :

GE d(G E/RT ) GE d(G E/RT )

ln γ1 = + x2 ln γ2 = − x1
RT d x1 RT d x1
Substitute x2 = 1 − x1 in the given equaiton for G E/RT and reduce:

GE d(G E/RT )
= −1.8 x1 + x12 + 0.8 x13 whence = −1.8 + 2 x1 + 2.4 x12
RT d x1

Then, ln γ1 = −1.8 + 2 x1 + 1.4 x12 − 1.6 x13 and ln γ2 = −x12 − 1.6 x13

(b) In accord with Eq. (11.11),

GE
= x1 ln γ1 + x2 ln γ2 = x1 (−1.8 + 2 x1 + 1.4 x12 − 1.6 x13 ) + (1 − x1 )(−x12 − 1.6 x13 )
RT

GE
Whence, = −1.8 x1 + x12 + 0.8 x13
RT

which is the first equation developed in part (a).

(c) Write Eq. (11.14) for a binary system with M̄i = ln γi and divide by d x1 :
d ln γ1 d ln γ2
x1 + x2 =0
d x1 d x1
Differentiate the the boxed equations of part (a):
d ln γ1 d ln γ2
= 2 + 2.8 x1 − 4.8 x12 and = −2 x1 − 4.8 x12
d x1 d x1
Multiply each derivative by the appropriate mole fraction and add:

x1 (2 + 2.8 x1 − 4.8 x12 ) + (1 − x1 )(−2 x1 − 4.8 x12 ) = 0

The validity of this equation is readily confirmed.

(d) Substitute x1 = 1 in the first derivative expression of part (c) and substitute x1 = 0 in the second
derivative expression of part (c). The results are:
   
d ln γ1 d ln γ2
= =0
d x1 x1 =1 d x1 x1 =0

684
 (e)

11.29 Combine definitions of the activity coefficient and the fugacity coefficients:

fˆi /xi P φ̂i

γi ≡ or γi =
f i /P φi

Note: See Eq. (14.54).

11.30 For C PE = const., the following equations are readily developed from those given in the last column
of Table 11.1 (page 415):
 
∂G E T
H = C P T
E E
and S = −
E
= C PE
∂ T P,x T 

Working equations are then:

H1E − G 1E T
S1E = and S2E = S1E + C PE
T1 T 

H2E = H1E + C PE T and G 2E = H2E − T2 S2E

For T1 = 298.15, T2 = 328.15, T  = 313.15 and T = 30, results for all parts of the problem are
given in the following table:

I. II. For C PE = 0

G 1E H1E S1E C PE S2E H2E G 2E S2E H2E G 2E

(a) −622 −1920 −4.354 4.2 −3.951 −1794 −497.4 −4.354 −1920 −491.4
(b) 1095 1595 1.677 3.3 1.993 1694 1039.9 1.677 1595 1044.7
(c) 407 984 1.935 −2.7 1.677 903 352.8 1.935 984 348.9
(d) 632 −208 −2.817 23.0 −0.614 482 683.5 −2.817 −208 716.5
(e) 1445 605 −2.817 11.0 −1.764 935 1513.7 −2.817 605 1529.5
(f) 734 −416 −3.857 11.0 −2.803 −86 833.9 −3.857 −416 849.7
(g) 759 1465 2.368 −8.0 1.602 1225 699.5 2.368 1465 688.0

685
11.31 (a) Multiply the given equation by n (= n 1 + n 2 ), and convert remaining mole fractions to ratios of
mole numbers:
nG E n1n2 n1n3 n2n3
= A12 + A13 + A23
RT n n n
Differentiation with respect to n 1 in accord with Eq. (11.96) yields [(∂n/∂n 1 )n 2 ,n 3 = 1]:
   
1 n1 1 n1 n2n3
ln γ1 = A12 n 2 − 2 + A13 n 3 − 2 − A23 2
n n n n n
= A12 x2 (1 − x1 ) + A13 x3 (1 − x1 ) − A23 x2 x3
Similarly, ln γ2 = A12 x1 (1 − x2 ) − A13 x1 x3 + A23 x3 (1 − x2 )
ln γ3 = −A12 x1 x2 + A13 x1 (1 − x3 ) + A23 x2 (1 − x3 )

(b) Each ln γi is multiplied by xi , and the terms are summed. Consider the first terms on the right of
each expression for ln γi . Multiplying each of these terms by the appropriate xi and adding gives:
A12 (x1 x2 − x12 x2 + x2 x1 − x22 x1 − x1 x2 x3 ) = A12 x1 x2 (1 − x1 + 1 − x2 − x3 )
= A12 x1 x2 [2 − (x1 + x2 + x3 )] = A12 x1 x2
An analogous result is obtained for the second and third terms on the right, and adding them
yields the given equation for G E/RT .

(c) For infinite dilution of species 1, x1 = 0: ln γ1 (x1 = 0) = A12 x2 + A13 x3 − A23 x2 x3

For pure species 1, x1 = 1: ln γ1 (x1 = 1) = 0
For infinite dilution of species 2, x2 = 0: ln γ1 (x2 = 0) = A13 x32
For infinite dilution of species 3, x3 = 0: ln γ1 (x3 = 0) = A12 x22

11.35 By Eq. (11.87), written with M ≡ G and with x replaced by y: GE = GR − i yi G iR

Equations (11.33) and (11.36) together give G iR = Bii P. Then for a binary mixture:
G E = B P − y1 B11 P − y2 B22 P or G E = P(B − y1 B11 − y2 B22 )

Combine this equation with the last equation on Pg. 402: G E = δ12 P y1 y2

 
∂G E
From the last column of Table 11.1 (page 415): S = − E
∂T P,x

dδ12
Because δ12 is a function of T only: SE = − P y1 y2
dT

 
dδ12
By the definition of G , H = G + T S ; whence,
E E E E
H = δ12 − T
E
P y1 y2
dT

 
∂HE
Again from the last column of Table 11.1: C PE =
∂T P,x

d 2 δ12
This equation and the preceding one lead directly to: C PE = −T P y1 y2
dT 2

686
    
∂(G E /RT ) −H E ∂(G E /T ) −H E
11.41 From Eq. (11.95): = or =
∂T P RT 2 ∂T P T2
 
∂(G E /T ) (G E /T ) −H E
To an excellent approximation, write: ≈ ≈ 2
∂T P T Tmean

(G E /T ) 785/323 − 805/298 −0.271

From the given data: = = = −0.01084
T 323 − 298 25
−H E −1060
and 2
= = −0.01082
Tmean 3132
The data are evidently thermodynamically consistent.
d M̄1 d M̄2
11.42 By Eq. (11.14), the Gibbs/Duhem equation, x1 + x2 =0
d x1 d x1
d M̄1 d M̄2
Given that M̄1 = M1 + Ax2 and M̄2 = M2 + Ax1 then = −A and =A
d x1 d x1
d M̄1 d M̄2
Then x1 + x2 = −x1 A + x2 A = A(x2 − x1 ) = 0
d x1 d x1
The given expressions cannot be correct.
11.45 (a) For M E = Ax12 x22 find M̄1E = Ax1 x22 (2 − 3x1 ) and M̄2E = Ax12 x2 (2 − 3x2 )
Note that at both x1 = 0 (x2 = 1) and x1 = 1 (x2 = 0), M̄1E = M̄2E = 0
In particular, ( M̄1E )∞ = ( M̄2E )∞ = 0
Although M E has the same sign over the whole composition range, both M̄1E and M̄2E change
sign, which is unusual behavior. Find also that

d M̄1E d M̄2E
= 2Ax2 (1 − 6x1 x2 ) and = −2Ax1 (1 − 6x1 x2 )
d x1 d x1
The two slopes are thus of opposite sign, as required; they also change sign, which is unusual.
d M̄1E d M̄2E
For x1 = 0 = 2A and =0
d x1 d x1

d M̄1E d M̄2E
For x1 = 1 =0 and = −2A
d x1 d x1
(b) For M E = A sin(π x1 ) find:
M̄1E = A sin(π x1 ) + Aπ x2 cos(π x1 ) and M̄2E = A sin(π x1 ) − Aπ x1 cos(π x1 )
d M̄1E d M̄2E
= −Aπ 2 x2 sin(π x1 ) and = Aπ 2 x1 sin(π x1 )
d x1 d x1
The two slopes are thus of opposite sign, as required. But note the following, which is unusual:
d M̄1E d M̄2E
For x1 = 0 and x1 = 1 =0 and =0
d x1 d x1

PLOTS OF THE FUNCTIONS ARE SHOWN ON THE FOLLOWING PAGE.

687
 Pb. 11.45 (a) A 10 i 0 .. 100 xi .00001 .01 . i

MEi A . xi 2 . 1 xi 2 MEbar1i A . xi . 1 xi 2 2 3 . xi

MEbar2i A . xi . xi . 1 xi . 2 3. 1 xi

1.5
MEi
1
MEbar1
i
0.5
MEbar2i

0.5
0 0.2 0.4 0.6 0.8 1
xi

Pb. 11.45 (b) MEi A . sin p . xi (pi prints as bf p)

MEbar1i A . sin p . xi A.p . 1 xi . cos p . xi

MEbar2i A . sin p . xi A . p . xi . cos p . xi

40

30

MEi
20
MEbar1
i

MEbar2i 10

10
0 0.2 0.4 0.6 0.8 1
xi

687A

  
∂(n M) ∂M
11.46 By Eq. (11.7), M̄i = = M +n
∂n i T,P,n j ∂n i T,P,n j
 
∂M
At constant T and P, dM = d xk
k
∂ xk T,P,x j

Divide by dn i with restriction to constant n j ( j = i):

     
∂M ∂M ∂ xk
=
∂n i T,P,n j k
∂ xk T,P,x j ∂n i nj

 n
 k
nk

∂ xk
  − n2
 (k = i)
With xk = =
n ∂n i 
 1 n
nj  − i (k = i)
n n2

     
∂M 1 ∂M 1 ∂M
=− xk + (1 − xi )
∂n i T,P,n j n k=i
∂ xk T,P,x j n ∂ xi T,P,x j
   
1 ∂M 1 ∂M
= − xk
n ∂ xi T,P,x j n k
∂ xk T,P,x j

   
∂M ∂M
M̄i = M + − xk
∂ xi T,P,x j k
∂ xk T,P,x j

For species 1 of a binary mixture (all derivatives at constant T and P):

          
∂M ∂M ∂M ∂M ∂M
M̄1 = M + − x1 − x2 = M + x2 −
∂ x1 x2 ∂ x1 x2 ∂ x2 x1 ∂ x1 x2 ∂ x2 x1

Because x1 + x2 = 1, the partial derivatives in this equation are physically unrealistic; however, they
do have mathematical significance. Because M = M(x1 , x2 ), we can quite properly write:
   
∂M ∂M
dM = d x1 + d x2
∂ x 1 x2 ∂ x 2 x1

Division by d x1 yields:
       
dM ∂M ∂M d x2 ∂M ∂M
= + = −
d x1 ∂ x1 x2 ∂ x2 x1 d x1 ∂ x1 x2 ∂ x2 x1

wherein the physical constraint on the mole fractions is recognized. Therefore

dM
M̄1 = M + x2
d x1

The expression for M̄2 is found similarly.

688


∂(n M E )
11.47 (a) Apply Eq. (11.7) to species 1: M̄1E =
∂n 1 n2

Multiply the given equation by n and eliminate the mole fractions in favor of mole numbers:
 
1 1
n M = An 1 n 2
E
+
n 1 + Bn 2 n 2 + Bn 1
   
1 1 −1 B
M̄1E = An 2 + + n1 −
n 1 + Bn 2 n 2 + Bn 1 (n 1 + Bn 2 )2 (n 2 + Bn 1 )2
Conversion back to mole fractions yields:
   
1 1 1 B
M̄1 = Ax2
E
+ − x1 +
x1 + Bx2 x2 + Bx1 (x1 + Bx2 )2 (x2 + Bx1 )2

The first term in the first parentheses is combined with the first term in the second parentheses
and the second terms are similarly combined:
    
1 x1 1 Bx1
M̄1 = Ax2
E
1− + 1−
x1 + Bx2 x1 + Bx2 x2 + Bx1 x2 + Bx1

Reduction yields:


B 1
M̄1E = Ax22 +
(x1 + Bx2 )2 (x2 + Bx1 )2

Similarly,


1 B
M̄2E = Ax12 +
(x1 + Bx2 )2 (x2 + Bx1 )2

(b) The excess partial properties should obey the Gibbs/Duhem equation, Eq. (11.14), when written
for excess properties in a binary system at constant T and P:

d M̄1E d M̄2E
x1 + x2 =0
d x1 d x1

If the answers to part (a) are mathematically correct, this is inevitable, because they were derived
from a proper expression for M E . Furthermore, for each partial property M̄iE , its value and
derivative with respect to xi become zero at xi = 1.
   
E ∞ 1 E ∞ 1
(c) ( M̄1 ) = A +1 ( M̄2 ) = A 1 +
B B

11.48 By Eqs. (11.15) and (11.16), written for excess properties, find:

d M̄1E d2 M E d M̄2E d2 M E
= x2 = −x1
d x1 d x12 d x1 d x12

At x1 = 1, d M̄1E /d x1 = 0, and by continuity can only increase or decrease for x1 < 1. Therefore the
sign of d M̄1E /d x1 is the same as the sign of d 2 M E /d x12 . Similarly, at x1 = 0, d M̄2E /d x1 = 0, and by
the same argument the sign of d M̄2E /d x1 is of opposite sign as the sign of d 2 M E /d x12 .

689
11.49 The claim is not in general valid.
 
1 ∂V
β≡ V id = xi Vi
V ∂T P i
 
1 ∂ Vi 1
β id
= xi = xi Vi βi
i xi Vi i
∂T P i xi Vi i

The claim is valid only if all the Vi are equal.

690