Solutions Thermo Coefficients Sample
Solutions Thermo Coefficients Sample
Solutions Thermo Coefficients Sample
Instructor:
Engr. Michael Allan G. Ramos
Chemical Engineering Department
Technological Institute of the Philippines
1
Modified Raoult’s Law
𝑦! 𝑃 = 𝑥! 𝑃!"#$
2
Henry’s Law & Lewis/Randall Rule
𝑓'!
𝛾! = !"
𝑓'!
3
Henry’s Law & Lewis/Randall Rule (cont.)
Henry’s Law is related to the Lewis/Randall rule through the Gibbs/Duhem equation.
𝜕𝑀 𝜕𝑀
Gibbs/Duhem Equation: 𝑑𝑃 + .% = 0
𝑑𝑇 − , 𝑥% 𝑑 𝑀
𝜕𝑃 !,# 𝜕𝑇 $,#
%
𝜕𝐺 𝜕𝐺
Taking M as Gibbs energy: 𝑑𝑃 + 𝑑𝑇 − , 𝑥% 𝑑 𝐺̅% = 0
𝜕𝑃 !,# 𝜕𝑇 $,#
%
4
Henry’s Law & Lewis/Randall Rule (cont.)
Henry’s Law is related to the Lewis/Randall rule through the Gibbs/Duhem equation.
For binary liquid solution: 𝑥& 𝑑𝜇&' + 𝑥( 𝑑𝜇(' = 0 Recall: 𝜇% ≡ 𝐺̅% = Γ% 𝑇 + 𝑅𝑇 ln 𝑓:%
𝑑 𝑓:&' 𝑑 𝑓:('
𝑑 ln 𝑓:&' 𝑑 ln 𝑓:(' 𝑑𝑥& 𝑑𝑥(
Since dx1 = -dx2: 𝑥& = 𝑥( = '
𝑑𝑥& 𝑑𝑥( :
𝑓& '
𝑓:(
𝑥& 𝑥(
5
Henry’s Law & Lewis/Randall Rule (cont.)
𝑑 𝑓:&' 𝑑 𝑓:('
𝑑𝑥& 𝑑𝑥(
= '
:
𝑓& '
𝑓:(
𝑥& 𝑥(
𝑑 𝑓:&' 𝑑 𝑓:('
𝑑𝑥& 𝑑𝑥(
In the limit as 𝑥& → 1 and 𝑥( → 0: lim = lim
#! →& 𝑓:&' #" *+ 𝑓:('
𝑥& 𝑥(
1
𝑑 𝑓:(' ≡ 𝐻𝑒𝑛𝑟𝑦 ! 𝑠 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡
1 𝑑 𝑓:&' 𝑑𝑥(
#" *+
But 𝑓:&' = 𝑓&' when 𝑥& = 1: =
𝑓&' 𝑑𝑥& 𝑓:('
#! *& lim ≡ 𝐻𝑒𝑛𝑟𝑦 ! 𝑠 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡
#" *+ 𝑥(
6
Henry’s Law & Lewis/Randall Rule (cont.)
1 𝑑 𝑓:&'
=1
𝑓&' 𝑑𝑥& #! *&
7
Relating Activity Coefficients to Excess Gibbs Energy
Taking M as 𝐺 , : 𝐺 , = , 𝑥% 𝐺̅%,
𝑛𝐺 ,
% 𝜕
𝑅𝑇
ln 𝛾% =
𝐺 , = 𝑥& 𝐺&̅ , + 𝑥( 𝐺̅(, 𝜕𝑛% $,!, .#
8
Estimating Activity Coefficients from Experimental VLE
𝑛𝐺 #
𝜕 𝑅𝑇
ln 𝛾" =
𝜕𝑛" $,&, '!
Prediction of VLE
9
Activity Coefficient Models
𝐺6
Generally: = 𝑔(𝑇, 𝑃, 𝑥7, 𝑥8, … , 𝑥9 )
𝑅𝑇
𝑛𝐺 ,
𝜕
ln 𝛾& =
𝑅𝑇 𝐺&̅ , = 𝑅𝑇 ln 𝛾& = 𝐴𝑥((
𝜕𝑛&
𝑛𝐺 , $,!, ."
ln 𝛾& = 𝐴𝑥((
𝜕
𝑅𝑇
ln 𝛾% =
𝜕𝑛% $,!, .# 𝑛𝐺 ,
𝜕 𝑅𝑇
ln 𝛾( = 𝐺̅(, = 𝑅𝑇 ln 𝛾( = 𝐴𝑥&(
𝜕𝑛( $,!, .!
ln 𝛾( = 𝐴𝑥&(
11
The Three-suffix Margules Equation (A≠0, B ≠ 0, C=…=0)
𝐺6 𝐺6
= 𝐴 + 𝐵(𝑥7 − 𝑥8) = 𝐴87𝑥7 + 𝐴78𝑥8 𝑥7𝑥8
𝑥7𝑥8𝑅𝑇 𝑅𝑇
Where: 𝐴78 = 𝐴 − 𝐵 (single molecule of 1 surrounded by 2)
𝐴87 = 𝐴 + 𝐵 (single molecule of 2 surrounded by 1)
𝑛𝐺 ,
𝜕
𝑅𝑇 ln 𝛾& = 𝑥(( 𝐴&( + 2 𝐴(& − 𝐴&( 𝑥&
ln 𝛾& =
𝜕𝑛&
𝑛𝐺 , $,!, ."
𝜕
𝑅𝑇
ln 𝛾% =
𝜕𝑛% $,!, .# 𝑛𝐺 ,
𝜕
𝑅𝑇 ln 𝛾( = 𝑥&( 𝐴(& + 2 𝐴&( − 𝐴(& 𝑥(
ln 𝛾( =
𝜕𝑛( $,!, .!
12
The Three-suffix Margules Equation (A≠0, B ≠ 0, C=…=0)
13
The Three-suffix Margules Equation (A≠0, B ≠ 0, C=…=0)
14
The Three-suffix Margules Equation (A≠0, B ≠ 0, C=…=0)
15
Determination of 𝜸𝒊 from Experimental VLE
𝒚𝒊 𝑷 = 𝜸𝒊 𝒙𝒊 𝑷𝒔𝒂𝒕
𝒊
D! E
𝛾! = F E"#$
! !
16
Determination of 𝜸𝒊 from Experimental VLE (cont.)
17
Determination of 𝜸𝒊 from Experimental VLE (cont.)
𝐺,
= 𝑥& ln 𝛾& + 𝑥( ln 𝛾(
𝑅𝑇
18
Determination of 𝜸𝒊 from Experimental VLE (cont.)
19
Determination of 𝜸𝒊 from Experimental VLE (cont.)
𝐺6 𝐺6 Provides a
= 𝐴87𝑥7 + 𝐴78𝑥8 𝑥7𝑥8 = 𝐴87𝑥7 + 𝐴78𝑥8 linear relation
𝑅𝑇 𝑥7𝑥8𝑅𝑇
𝐺6
= 𝐴87𝑥7 + 𝐴78(1 − 𝑥7)
𝑥7𝑥8𝑅𝑇
𝐺6
= 𝐴87 − 𝐴78 𝑥7 + 𝐴78 𝑦 − 𝑖𝑛𝑡𝑒𝑟𝑐𝑒𝑝𝑡 = 𝐴78
𝑥7𝑥8𝑅𝑇
𝑠𝑙𝑜𝑝𝑒 = 𝐴87 − 𝐴78
𝒚 = 𝒎 𝒙 + 𝒃
20
Determination of 𝜸𝒊 from Experimental VLE (cont.)
𝐺6 𝐺6 Provides a
= 𝐴87𝑥7 + 𝐴78𝑥8 𝑥7𝑥8 = 𝐴87𝑥7 + 𝐴78𝑥8 linear relation
𝑅𝑇 𝑥7𝑥8𝑅𝑇
𝐺6
• When 𝑥! = 1, 𝑥" = 0: = 𝐴87 1 + 𝐴78 0 = 0.198
𝑥7𝑥8𝑅𝑇
∴ 𝑨𝟐𝟏 = 𝟎. 𝟏𝟗𝟖
𝐺6
• When 𝑥! = 0, 𝑥" = 1: = 𝐴87 0 + 𝐴78 1 = 0.372
𝑥7𝑥8𝑅𝑇
∴ 𝑨𝟏𝟐 = 𝟎. 𝟑𝟕𝟐
21
Determination of 𝜸𝒊 from Experimental VLE (cont.)
22
Determination of 𝜸𝒊 from Experimental VLE (cont.)
𝐺6 𝐺6
= 𝐴87𝑥7 + 𝐴78𝑥8 = 0.198𝑥7 + 0.372𝑥8 𝑥7𝑥8
𝑥7𝑥8𝑅𝑇 𝑅𝑇
23
Determination of 𝜸𝒊 from Experimental VLE (cont.)
24
Determination of 𝜸𝒊 from Experimental VLE (cont.)
7. Plot again the liquid properties (calculated from correlation) vs 𝑥! .
25
Determination of 𝜸𝒊 from Experimental VLE (cont.)
26
Determination of 𝜸𝒊 from Experimental VLE (cont.)
27
Determination of 𝜸𝒊 from Experimental VLE (cont.)
8. Predict the VLE using Modified Raoult’s Law and the calculated activity coefficients from
the correlation .
𝑃 = 𝑥7𝛾7𝑃7=>? + 𝑥8𝛾8𝑃8=>?
@! A! B!"#$ @! A! B!"#$
𝑦7 = B
= @! A! B!"#$ C@% A% B%"#$
28
Determination of 𝜸𝒊 from Experimental VLE (cont.)
29
Determination of 𝜸𝒊 from Experimental VLE (cont.)
30
Thermodynamic Consistency Test
If the experimental data are inconsistent with the Gibbs/Duhem equation,
then the data are incorrect due to systematic error in the data.
∗
𝐺,
= 𝑥& ln 𝛾&∗ + 𝑥( ln 𝛾(∗ Where: (*) denotes experimental values
𝑅𝑇
∗
𝐺,
𝑑 𝑑 ln 𝛾&∗ 𝑑 ln 𝛾 ∗
𝑅𝑇 (
= 𝑥& + ln 𝛾&∗ + 𝑥( − ln 𝛾(∗
𝑑𝑥& 𝑑𝑥& 𝑑𝑥&
∗
𝐺, 𝐺,
𝑑 𝛾&∗ 𝑑 ln 𝛾&∗ 𝑑 ln 𝛾(∗ 𝑑
𝑅𝑇 𝑅𝑇 𝛾& 𝑑 ln 𝛾& 𝑑 ln 𝛾(
= ln ∗ + 𝑥& + 𝑥( = ln + 𝑥& + 𝑥(
𝑑𝑥& 𝛾( 𝑑𝑥& 𝑑𝑥& 𝑑𝑥& 𝛾( 𝑑𝑥& 𝑑𝑥&
(Based on experimental property values) (Based on derived property values from correlations)
31
Thermodynamic Consistency Test (cont.)
𝐺,
𝑑 𝛾& 𝑑 ln 𝛾& 𝑑 ln 𝛾(
𝑅𝑇
= ln + 𝑥& + 𝑥(
𝑑𝑥& 𝛾( 𝑑𝑥& 𝑑𝑥&
𝐺,
𝑑 𝛾&
𝑅𝑇 (Based on derived property values from correlations)
= ln
𝑑𝑥& 𝛾(
32
Thermodynamic Consistency Test (cont.)
∗
𝐺, 𝐺,
𝑑 𝑑 𝛾& 𝛾&∗ 𝑑 ln 𝛾&∗ 𝑑 ln 𝛾(∗
𝑅𝑇 𝑅𝑇
− = ln − ln ∗ + 𝑥& + 𝑥(
𝑑𝑥& 𝑑𝑥& 𝛾( 𝛾( 𝑑𝑥& 𝑑𝑥&
0
𝐺,
𝑑 𝛿 𝑅𝑇 𝛾& 𝑑 ln 𝛾&∗ 𝑑 ln 𝛾(∗
= 𝛿 ln − 𝑥& + 𝑥( Where: (𝛿) denotes residual (derived - actual).
𝑑𝑥& 𝛾( 𝑑𝑥& 𝑑𝑥&
33
Thermodynamic Consistency Test (cont.)
34
Thermodynamic Consistency Test (cont.)
35
Thermodynamic Consistency Test (cont.)
36
Barker’s Method
• With the aid of a computer, values for the Margules parameters A12 and A21 are
searched to yield pressure values as close as possible to the measured values.
Step 1: Get the difference between 𝑷𝒄𝒂𝒍𝒄𝒖𝒍𝒂𝒕𝒆𝒅 & 𝑷𝒆𝒙𝒑𝒆𝒓𝒊𝒎𝒆𝒏𝒕𝒂𝒍. 𝑃R>SR − 𝑃T@U?
"
Step 2: Calculate the sum of squared errors (SSE). Σ 𝑃%&'% − 𝑃()*+
Step 3: Using Excel Solver, set the objective to minimize the SSE by changing variable
cells A12 and A21.
37
The van Laar Equation
0
𝐺, 𝐴&( 𝐴0(&
= 0
𝑥& 𝑥( 𝑅𝑇 𝐴&( 𝑥& + 𝐴0(& 𝑥(
0 1( 1(
0 𝐴&( 𝑥& 𝐴0(& 𝑥(
ln 𝛾& = 𝐴&( 1+ 0 ln 𝛾( = 𝐴0(& 1+ 0
𝐴(& 𝑥( 𝐴&( 𝑥&
ln 𝛾&2 = 𝐴&(
0 (when 𝑥( = 0) ln 𝛾(2 = 𝐴0(& (when 𝑥) = 0)
38
The van Laar Equation (cont.)
0
𝑥& 𝑥( 𝑅𝑇 𝑥& 𝑥( 𝐴&( 𝑥& + 𝐴0(& 𝑥(
= 0 + 0 = 0
𝐺 , 𝐴(& 𝐴&( 𝐴&( 𝐴0(&
𝑥& 𝑥( 𝑅𝑇 1 1 1
= − 𝑥& +
𝐺, 𝐴0(& 𝐴&(
0 0
𝐴&(
𝒚 = 𝒎 𝒙 + 𝒃
0 1 1
𝐴&( = 𝐴0(& =
𝑏 𝑚+𝑏
39
The van Laar Equation (cont.)
40
Features of Margules and van Laar Equations
• No theoretical foundations
41
Local Composition Models
• Theoretical development in the molecular thermodynamics of liquid solution behavior are often
based on the concept of local composition (concentration around a molecule is different from the
bulk concentration).
• Local compositions presumed to account for the short-range order and non-random molecular
orientations that results from differences in molecular size and intermolecular forces
42
The Wilson Equation
𝐺,
= −𝑥& ln 𝑥& + 𝑥( Λ&( − 𝑥( ln 𝑥( + 𝑥& Λ (&
𝑅𝑇
Λ&( Λ (&
ln 𝛾& = − ln 𝑥& + 𝑥( Λ&( + 𝑥( − ln 𝛾&2 = − ln Λ&( + 1 − Λ (&
𝑥& + 𝑥( Λ&( 𝑥( + 𝑥& Λ (&
Λ&( Λ (&
ln 𝛾( = − ln 𝑥( + 𝑥& Λ (& − 𝑥& − ln 𝛾(2 = − ln Λ (& + 1 − Λ&(
𝑥& + 𝑥( Λ&( 𝑥( + 𝑥& Λ (&
𝑉( 𝜆&(
Λ&( = exp −
𝑉& 𝑅𝑇 Wilson parameters must be restricted to positive numbers to
𝑉& 𝜆(& ensure validity of the above equations at infinite dilution.
Λ (& = exp −
𝑉( 𝑅𝑇
𝜆%3 is an energetic parameter (decribes how i-i interaction varies from i-j interaction)
𝑉3 /𝑉% is ratio of molar volumes (accounts for the relative sizes of molecules of i and j)
43
The Wilson Equation (cont.)
𝐺,
= −𝑥& ln 𝑥& + 𝑥( Λ&( − 𝑥( ln 𝑥( + 𝑥& Λ (&
𝑅𝑇
Λ&( Λ (&
ln 𝛾& = − ln 𝑥& + 𝑥( Λ&( + 𝑥( − ln 𝛾&2 = − ln Λ&( + 1 − Λ (&
𝑥& + 𝑥( Λ&( 𝑥( + 𝑥& Λ (&
Λ&( Λ (&
ln 𝛾( = − ln 𝑥( + 𝑥& Λ (& − 𝑥& − ln 𝛾(2 = − ln Λ (& + 1 − Λ&(
𝑥& + 𝑥( Λ&( 𝑥( + 𝑥& Λ (&
Applicability: • Mixtures of polar and non polar species (i.e. alcohols & alkanes)
• Hydrocarbon mixtures
Limitation: • Systems exhibiting partial miscibility
44
The Wilson Equation (cont.)
45
The NRTL Equation
𝐺, 𝐺(& 𝜏(& 𝐺&( 𝜏&(
= +
𝑥& 𝑥( 𝑅𝑇 𝑥& + 𝑥( 𝐺(& 𝑥( + 𝑥& 𝐺&(
(
𝐺(& 𝐺&( 𝜏&(
ln 𝛾& = 𝑥(( 𝜏(& + ( ln 𝛾&2 = 𝜏(& + 𝜏&( exp −𝛼 𝜏&(
𝑥& + 𝑥( 𝐺(& 𝑥( + 𝑥& 𝐺&(
(
𝐺&( 𝐺(& 𝜏(&
ln 𝛾( = 𝑥&( 𝜏&( + (
ln 𝛾(2 = 𝜏&( + 𝜏(& exp −𝛼 𝜏(&
𝑥( + 𝑥& 𝐺&( 𝑥& + 𝑥( 𝐺(&
47
The UNIQUAC Model
• A more sophisticated excess Gibbs energy model.
• Divides the excess Gibbs energy into two parts: combinatorial part & residual part.
𝐺 6
g6 ≡
𝑅𝑇
g6 = g6RYZ[!9>?Y\!>S + g6\T=!"]>S
49
The UNIQUAC Model (cont.)
50
The UNIFAC Model
• Termed as predictive activity coefficient model: if data are available for all the functional
groups or subgroups that comprise the molecules in a system, activity coefficients can be
calculated without data from the specific molecules in the mixture.
51
The UNIFAC Model (cont.)
52
The UNIFAC Model (cont.)
53
The UNIFAC Model (cont.)
54
END.
References:
• Smith et al. (2018). Introduction to Chemical Engineering Thermodynamics (8th ed.)
• Koretzky, M. (2013). Engineering and Chemical Thermodynamics (2nd ed.)
• Dugos, N. (De La Salle University). Solution Thermodynamics: Applications. PIChE-MMAC UPLIFT X. Seminar-
Workshop on Teaching Solutions Thermodynamics. October 17, 2020.