NATIONAL UNIVERSITY OF COLOMBIA - BOGOTA CAMPUS

THERMODYNAMIC AND TRANSPORT PROPERTIES LABORATORY

DETERMINATION OF BUBBLE TEMPERATURE

1. OBJECTIVES

GENERAL OBJECTIVE

Determine the bubble temperature for binary mixtures.

SPECIFIC OBJECTIVES

Calibrate the refractometer.

Relate the concentrations of the mixtures to the refractive indices.
Build a phase diagram, boiling temperature vs composition of the solution.
Determine the azeotropic point of the mixture.

2. THEORETICAL FOUNDATION

Equilibrium is recognized as a static condition where, over time, no change occurs.

some in the macroscopic properties of a system, which implies a balance of all the
potentials that can cause a change. An isolated system consisting of the phases in
close contact liquid and vapor, over time it reaches a final state where it does not exist
tendency for a change to occur within it. The temperature, the pressure and the
Phase compositions achieve the final values that remain fixed thereafter.

Behavior of the EVL

Vapor/liquid equilibrium (VLE) is considered the state of coexistence of the vapor and liquid phases.
liquid. When N=2, the phase rule will be F=4-π. Since there must be at least one phase (π=1),
the maximum number of variables according to the phase rule that must be specified to fix the
The intensive state of the system is three: in particular P, T, and a molar or mass fraction. Due to
So, it is possible that all the equilibrium states of the system are represented in space.
three-dimensional P-T-composition. In this space, the pairs of phase states that coexist in
equilibrium (F=4-2=2) defines surfaces. In figure 1, a schematic diagram is displayed.
three-dimensional that explains these surfaces for the EVL. This figure indicates in a way
schematic P-T-composition surfaces, which contain the equilibrium states of
vapor and saturated liquid for species 1 and 2 of a binary system. The lower surface
includes the states of saturated vapor; which is the P-T-y1 surface. The upper surface includes the
saturated liquid states; this is the P-T-x1 surface. Such surfaces intersect
transversely along the RKAC1 and UBHC2 lines, which represent the pressure curves of
vapor as a function of T for pure species 1 and 2. Additionally, the lower and upper surfaces form
a rounded surface continuing across the top of the diagram between C1 and C2, the
critical points of pure species 1 and 2; meanwhile, the critical points of the numerous mixtures of
the two species are found along a line on the rounded edge of the surface
between C1 and C2. This critical line is defined by the points at which the vapor and liquid phases are
equilibrium are identical. Later, the critical region will be analyzed in depth.

The subcooled liquid region is located above the upper surface in Figure 1;
while the superheated vapor region is located below the lower surface. The space
the region between both surfaces is the coexistence region of the vapor and liquid phases. If one
starts with a liquid at F and reduces the pressure at constant temperature and composition to
Along the vertical line FG, the first steam bubble appears at point L, which is located on
the upper surface. In this way, L is a bubble point, and the upper surface is the
bubble point surface. The state of the vapor bubble in equilibrium with the liquid at L is
it must be represented by a point on the lower surface at the temperature and pressure of L. Said
point is indicated by V. The line LV is an example of a fold line, which connects the points that
they represent phases in equilibrium.

Figure 1. PTxy diagram for vapor/liquid equilibrium.

Models for Vapor/Liquid Equilibrium

When thermodynamics is applied to vapor/liquid equilibrium, the goal is to find, through the
calculation, of the temperatures, pressures, and the compositions of the phases in equilibrium. Furthermore, it is the
the means by which the predictions of various theories of molecular physics and statistical mechanics is
possible to apply for practical purposes; none is carried out without behavioral models
of the vapor/liquid equilibrium system. The two simplest are Raoult's law and Henry's law.
Raoult's Law

The two main assumptions required to reduce the EVL calculations for the law of
Raoult son:

The vapor phase is an ideal gas.

The liquid phase is an ideal solution.

The first assumption means that Raoult's law applies only to low pressures.
moderate. The second implies that it has approximate validity as long as the species
that constitute the system are chemically similar. Just like the ideal gas, it is useful as
a model with which the behavior of real gas can be compared, the ideal liquid solution
it represents a model with which it is possible to compare the behavior of a real solution.
Often, the behavior of the ideal solution is approximated by liquid phases where the
Molecular species are not very different in size, but they are of the same chemical nature.
A quantitative expression of Raoult's law is:

= ( = 1,2,… , )

Where it is a mole fraction of liquid phase, it is a mole fraction of the vapor phase, and it is the
vapor pressure of the pure species i at the system temperature.

Henry's Law

The application of Raoult's law for species i requires a value for to the temperature
of application, and for this reason it is inappropriate for a species whose temperature c the criticism is less than
the application temperature. Henry's law states that the partial pressure of the species in the
vapor phase is directly proportional to its mole fraction in the liquid phase. For this reason,

= Hi

Where Ԋi is the Henry constant.

Non-ideal liquid-vapor equilibrium (Smith, Van Ness, & Abbot)

In every phase equilibrium, the fugacities of a component i in each phase are

equal. In a multicomponent system, the ELV is given by the following expression:
 ̂ = ̂

Where the fugacity of the vapor phase ( ̂ ,) for ideal behavior at low pressures (less than
1 atm), e.g.:

̂ =

Being the mole fraction of component i in the vapor and P the working pressure.

And the fleeting nature of the liquid phasê ) is given by:

̂ =

Where it is the activity coefficient, which indicates the variation of the fugacity of a
component in liquid phase when it is in mixture compared to when it is pure; the
mole fraction of component i in the liquid and the vapor pressure of component i when it
find pure.

To finally give the expression that governs the ELV:

On the other hand, when discussing a non-ideal solution, it is necessary to introduce the concept of
excess property, which refers to the difference of a molar property (M) in the real solution
regarding the ideal, at the same pressure, temperature, and composition.

= − . .

This concept can be applied to the molar Gibbs Free Energy of a mixture as follows
way
̂ = ̂ − ̂ =. ∑. ̅

In phase equilibrium, the most important property is the partial molar Gibbs free energy.
excess ̅ ), since it is directly related to the activity coefficient, as
is shown below:

̅1 ̂ (1 - x1 ) ̂
= + ( )
1

In which:

̅
̅ = ( ) o = ( )

Where it is the activity coefficient given by the relation of fugacities of real and ideal mixture
in liquid state:
 ̂ ( )
=
( )

3. EXPERIMENTAL PROCEDURE

Calibration curve.

1. Prepare different alcohol solutions with water that cover the working range.
2. Pass each of the samples and measure the refractive index.

Bubble temperature

Add 50 ml of the selected alcohol to the team's flask.

2. Let water pass through the coolant in counterflow.
3. Start the warm-up.
4. When a constant drip is achieved, take a sample of the funds and of the
condensed in such a way that they can be deposited in a vial.
5. Add 3 ml of distilled water to the flask.
6. Repeat steps 3 to 5 15 times.

4. EQUIPMENT DIAGRAM
5. DATA TABLES

The data collected in the practice consisted of the calibration of the refractometer and the indices of
refraction of each of the samples from the evaporation process. Table No. 1 presents
the refractive indices measured for eight samples of a water-ethanol mixture along with a
Sample of pure ethanol and water. Because these indices were taken at a temperature
each one underwent an average calibration process at 20°C using data reported in the
literature.

∆ɳ ɳ (° ) − ɳ 20°C
= =| |
∆ T (°G) - 20°

ɳ 20= ɳ + F(T - 20)

Table 1. Data table for refractometer calibration

Ethanol Fraction IR T (°C) IR (20°C)

1 1,35987 23.72 1,36113108
0.9 1.36139 23.62 1.36261718
0.8 1.36218 23.62 1.36340718
0.7 1,36279 23.71 1.36404769
0.6 1.36304 23.66 1.36428074
0.5 1,36282 23.73 1,36408447
0.4 1.36201 23.68 1.36325752
 0.3 1.36009 23.73 1.36135447
0.2 1.35604 23.72 1.35730108
0.1 1.34698 23.72 1.34824108
0 1.3321 23.72 1.33336108

Figure 2. Calibration curve of the refractometer

Calibration curve of the index of

refraction.
1.37

1.36

1.35

1.34

1.33
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1
Molar fraction Ethanol

Subsequently, as the heating process of the ethanol-water mixture progressed

Samples of both the liquid and the condensed vapor were taken, from which measurements were made.
its respective refractive index at the temperature reported by the equipment, and likewise that
the calibration, the data was corrected to 20°C.

Table 2. Calculation of corrected refractive indices of liquid and vapor samples at 20°C

Liquid IR Sample T (°C) IR 20 °C Liquid IR Vapor T (°C) IR 20 °C Vapor

1 1,36089 27 1,363263 1,36023 24 1,361586
2 1,36165 27 1,364023 1,36082 24 1,362176
3 1.3619 27 1,364273 1.36138 24 1,362736
4 1.36067 27 1,363043 1.36152 24 1.362876
5 1,35658 27 1,358953 1.36175 24 1,363,106
6 1.35641 27 1,358783 1.3616 24 1,362956
7 1.34506 27 1,347433 1.3448 24 1.346156

6. SAMPLE OF CALCULATIONS AND RESULTS

With the temperature-corrected data, we proceeded to find the compositions that
corresponded to the refractive indices of each of the samples taken, for this one
he built a trend line, however the curve of refractive index against composition
it exhibited a semi-parabolic behavior and therefore for the same refractive index in
a certain interval could correspond to two different compositions, due to this it was decided
divide the calibration line of the refractometer to achieve better results:

Figure 3. Calibration curve from 0 to 0.5 mole fraction of ethanol

FRACTION REGRESSION 0 TO 0.5 EtOH

1.37
1.365
1.36
1.355 y = 0.299x3- 0.4008x2+ 0.1872x + 1.3333
R² = 0.9999
1.35
1.345
1.34
1.335
1.33
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55
Molar fraction of ethanol

Figure 4. Calibration curve from 0.6 to 1 molar fraction of ethanol

FRACTIONAL REGRESSION 0.6 TO 1 EtOH

1.3645
1.364
1.3635
1.363
1.3625
1.362 y = -0.019x2+ 0.0226x + 1.3575
1.3615 R² = 0.9973

1.361
1.3605
0.4 0.5 0.6 0.7 0.8 0.9 1 1.1
Having divided the regression of the data, we proceeded to look for the compositions of each
sample, this was done by equating 'y' to the refractive index measured in the laboratory and searching
the corresponding "x" that would yield that result through an iteration process until the
difference between "y" and the experimental value outside zero.

Table 3. Calculated compositions of liquid and vapor samples of the ethanol-water mixture

Liquid Liquid fraction Vapor Vapor fraction

-0.00000768091 0.82012624 0 0.967106658
-3.32908E-06 0.697553848 0 0.922771365
-5.2484E-05 0.594 0 0.874259585
0 0.38645344 7.81076E-06 0.85997909
0 0.233250 0.0000818747 0.399987565
0 0.22983 0.00000100114 0.379161261
0 0.09258 0 0.082280359

Using the appropriate regressions through the estimation of the equilibrium trend Ethanol–
water was used with higher-order fraction regression for the first 3 compositions of the liquid and the
4 firsts of the steam.

Figure 5. Experimental LV equilibrium of ethanol and water at a constant pressure of 560 mmHg

ETHANOL - WATER EQUILIBRIUM

88
86
84
82
80
78 Liquid
76 Vapor
74
72
70
0 0.2 0.4 0.6 0.8 1 1.2
Molar fraction of ethanol (x,y)

The fractions were graphed against the temperature reported on the equipment upon reaching a
stable state.
For the respective comparison of the data, an approach had been proposed.
through the Wilson model to generate the equilibrium of the mixture, this was done using
Aspen and a binary analysis at constant pressure:

Figure 5. LV equilibrium using Wilson's model of ethanol and water at a constant pressure of 560 mmHg

Comparison of experimental data with model

(Wilson)
95

90

85

80

75

70
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Molar fraction of ethanol (x,y)

Liquid (Wilson) Vapor (Wilson) Liquid (Exp) Vapor (Exp)

In addition to the comparison of the experimental data, a second one was carried out.
regression in order to find the parameters corresponding to the work equilibrium of three
thermodynamic models, NRTL, UNIQUAC and Wilson. This procedure was done with the
Aspen binary analysis tool and calculating the parameters that can be obtained
directly from that database. Subsequently, an adjustment was made...

i Water
j Ethanol

NRTL (NOT RANDOM TWO LIQUIDS)

Table 4. Equilibrium data using the NRTL model

NRTL
T (°C) ETHANOL (V)
91.6725061 0 0
78,0245061 0,46317 0.1111111
74,9020061 0,5548111 0,2222222
73,5228061 0,6011412 0,3333333
 72,5948061 0,6406338 0,4444444
71,8413061 0,6845104 0,5555556
71,2316061 0,7380513 0,6666667
70,8043061 0,8052623 0,7777778
70,6198061 0,8905848 0,8888889
70.7494061 1 1

Table 5. Ethanol - water equilibrium parameters (Theoretical and experimental)

NRTL PARAMETERS NRTL EXPERIMENTAL PARAMETERS

aj 3.4578 aj NRTL/1 38.0358
aji -0.8009 aji NRTL/1 5.59408476
by -1054,9456 by NRTL/2 5274,7281
bji 443,123997 at NRTL/2 -2215.62
cij 0.3 cij NRTL/3 0.04196405

UNIQUAC

Table 6. Equilibrium data using the UNIQUAC model.

UNIQUAC
T (°C) ETHANOL (V)
91.6725061 0 0
78.0081061 0.1111111
74.9372061 0.5528002 0.2222222
73.5538061 0.599363 0.3333333
72.6069061 0.6396658 0.4444444
71,8387061 0,6842461 0.5555556
71,2213061 0,7381038 0.6666667
70,7913061 0,8052435 0.7777778
70.6051061 0.8903655 0.8888889
70.7494061 1 1

Table 7. Ethanol-water equilibrium parameters (Theoretical and experimental)

UNIQUAC PARAMETERS
hey 2.0046
aji -2.4936
 at -1312,1468 PARAMETERS EXP UNIQUAC
bji 1362,50585 aj UNIQ/1 9,66424703
aji UNIQ/1 -7.6525086
at UNIQ/2 -3055.323
at UNIQ/2 -1091,4013

Wilson

Table 8. Equilibrium data using the Wilson model

Wilson
T (°C) ETHANOL (V)
91.6725061 0 0
78,2459061 0,4567318 0,1111111
75,0928061 0,5507741 0,2222222
73,5384061 0,6031783 0,3333333
72,4863061 0,6468782 0,4444444
71.6759061
71,0486061 0,7428796 0,6666667
70.6242061 0.8057888 0.7777778
70,4770061 0,8875071 0,8888889
70.7494061 1 1
Table 9. Ethanol-water equilibrium parameters (Theoretical and experimental)

Wilson PARAMETERS EXP Wilson PARAMETERS

aj -2,7362 aj WILSON/1 -2.2922152
spice 0.0898 chili WILSON/1 0.96335312
at 748,162254 at WILSON/2 419,486102
bji -201,43097 at WILSON/2 59.6150938

ANALYSIS OF RESULTS

After graphing the experimental data and overlaying it against the data calculated using
the Wilson thermodynamic model shows marked differences at almost all points of
which different errors can be inferred from practice; on one hand, despite the fact that the
The refractometer has great accuracy, providing temperature readings for each of the
Samples had different variations, reporting temperatures of 27°C which are referred to.
the temperature of the samples indicates that they should have been allowed to cool more each of the
samples for subsequent measurement.

Similarly, it is possible that the largest percentage of error reported occurred due to the
waiting time for the stabilization of the system, since despite waiting
approximately between 5 and 10 minutes for each temperature when the composition was
changed, more time could have been allowed to wait for a stable temperature, as it is observed
that the sample of almost pure water should have reported a much higher temperature than the
pressure during the distillation process of the mixture, such an error generates because of
so many imprecise temperatures that do not correspond to those of the equilibrium relevant to each one of
the compositions. In addition to this, the following and last source of error may have come from the
calculations and methods for the preparation of calibration samples and for distilling, by a
side because the calibration samples were very sensitive to the weight changes of the two
substances and on the other hand because the sampling and adding water involved outputs of
vapor that could generate changes in the composition.

Regarding the azeotrope, the line where the composition of the vapor and the
the liquid is the same for the mixture, but it can still be observed that in the high compositions
From the experimental points, especially in the dew point line, a trend is observed.
more linear and flat, indicating the existence of an azeotrope at those concentrations.
temperature.

Finally, a calibration method to find the

compositions of the samples by measuring other properties such as density,
which does not depend as substantially on temperature as the refractive index does and also
give a longer waiting time for both the samples taken for their subsequent cooling and
also for the stabilization of the system temperature in order to achieve the
maximum stability temperatures for each composition at equilibrium.

On the other hand, regarding the comparisons with the thermodynamic models, the comparisons were made with the
experimental data with three different NRTL models (recommended for liquid equilibrium
liquid), UNIQUAC (group contribution model) and Wilson (particle interaction)
adjacent). Through this process, parameters of each of these models were analyzed and
Note that following the results in the liquid-vapor equilibrium, each of these
variables differ quite a bit but there is one that differs considerably less and it is the model of
Wilson, which leads to saying that having initially compared the experimental data with this.
the model was a good decision and this is because the model unlike the other two...

CONCLUSIONS
The thermodynamic model of Wilson is the one that best fits the experimental data.
Measurements of the ethanol-water system.
The NRTL and UNIQUAC models are good but they are very precise and accurate for systems.
of other characteristics.
A different calibration method is necessary to avoid dividing the calibration curve.
according to the compositions.
Closer measurements to the compositions of pure substances are necessary for
to make more precise temperature measurements, especially that of the point of
azeotrope.
The stabilization of the temperature must occur over a longer period of time.