WEBVTT

00:00:00.000 --> 00:00:02.952 align:middle line:90%
[SQUEAKING]

00:00:02.952 --> 00:00:04.428 align:middle line:90%
[RUSTLING]

00:00:04.428 --> 00:00:06.888 align:middle line:90%
[CLICKING]

00:00:06.888 --> 00:00:11.130 align:middle line:90%


00:00:11.130 --> 00:00:13.860 align:middle line:90%
GIAN PAOLO BERETTA: OK.

00:00:13.860 --> 00:00:15.790 align:middle line:90%
So here we are.

00:00:15.790 --> 00:00:24.070 align:middle line:90%


00:00:24.070 --> 00:00:24.980 align:middle line:90%
We're just starting.

00:00:24.980 --> 00:00:26.878 align:middle line:90%
Yeah, we're a bit late.

00:00:26.878 --> 00:00:27.378 align:middle line:90%
OK.

00:00:27.378 --> 00:00:31.282 align:middle line:90%


00:00:31.282 --> 00:00:38.130 align:middle line:84%
So today, we talk about
liquid-vapor equilibrium

00:00:38.130 --> 00:00:39.080 align:middle line:90%
in mixtures.

00:00:39.080 --> 00:00:42.710 align:middle line:90%


00:00:42.710 --> 00:00:45.280 align:middle line:84%
We have seen
something already when

00:00:45.280 --> 00:00:49.840 align:middle line:90%
we did the Gibbs phase rule.

00:00:49.840 --> 00:00:58.600 align:middle line:84%
And you remember
that if we have--

00:00:58.600 --> 00:01:01.440 align:middle line:90%


00:01:01.440 --> 00:01:09.040 align:middle line:84%
if we consider the list of these
properties like temperature,

00:01:09.040 --> 00:01:16.930 align:middle line:84%
pressure, the compositions
of, say, two phases--

00:01:16.930 --> 00:01:26.380 align:middle line:84%
for example, a liquid
and a gas phase,

00:01:26.380 --> 00:01:32.870 align:middle line:84%
depending on how many
components you have--

00:01:32.870 --> 00:01:35.720 align:middle line:90%
r is the number of components--

00:01:35.720 --> 00:01:44.140 align:middle line:84%
plus 2 minus the
number of phases, this

00:01:44.140 --> 00:01:51.190 align:middle line:84%
was called the variance, namely
the number of these variables

00:01:51.190 --> 00:01:56.660 align:middle line:84%
that are independently
changeable so

00:01:56.660 --> 00:02:01.010 align:middle line:84%
that you remain in the
same kind of equilibrium.

00:02:01.010 --> 00:02:03.630 align:middle line:84%
Like in this case, we are
talking about liquid-vapor.

00:02:03.630 --> 00:02:07.520 align:middle line:84%
But we also talk a little
bit about solid-liquid.

00:02:07.520 --> 00:02:10.169 align:middle line:84%
And you could also do it
for solid-vapor and so on.

00:02:10.169 --> 00:02:14.850 align:middle line:90%


00:02:14.850 --> 00:02:20.990 align:middle line:84%
Of course, the chemical
potentials, as usual,

00:02:20.990 --> 00:02:26.900 align:middle line:84%
will play a major role,
as they did in order

00:02:26.900 --> 00:02:31.420 align:middle line:90%
to derive the Gibbs phase rule.

00:02:31.420 --> 00:02:43.200 align:middle line:84%
So the first result that we want
to derive is the Raoult's law.

00:02:43.200 --> 00:02:46.500 align:middle line:90%
Raoult's law has to do with--

00:02:46.500 --> 00:02:49.150 align:middle line:84%
and we'll do it for
the simplest situation.

00:02:49.150 --> 00:02:54.060 align:middle line:84%
So the simplest situation
will be one like--

00:02:54.060 --> 00:02:56.940 align:middle line:90%
one of these two cases.

00:02:56.940 --> 00:02:58.780 align:middle line:90%
This one is a liquid.

00:02:58.780 --> 00:03:03.870 align:middle line:84%
It's a liquid solution
with components 1 and 2.

00:03:03.870 --> 00:03:07.020 align:middle line:84%
For example, you
can think of water

00:03:07.020 --> 00:03:15.180 align:middle line:84%
as being component 1 and 2 could
be salt. So this is salt water.

00:03:15.180 --> 00:03:22.380 align:middle line:84%
And it is in mutual equilibrium
with the gaseous phase, in which

00:03:22.380 --> 00:03:25.470 align:middle line:90%
there is also water, so 1.

00:03:25.470 --> 00:03:28.440 align:middle line:84%
And there is another
component, 3.

00:03:28.440 --> 00:03:33.350 align:middle line:84%
For example, it could be
air because humid air--

00:03:33.350 --> 00:03:34.930 align:middle line:90%
air has some humidity.

00:03:34.930 --> 00:03:37.480 align:middle line:90%
And humidity is water vapor.

00:03:37.480 --> 00:03:45.780 align:middle line:84%
So this is, for example,
salt water and humid air,

00:03:45.780 --> 00:03:51.090 align:middle line:90%
moist air in mutual equilibrium.

00:03:51.090 --> 00:03:53.880 align:middle line:84%
We are considering
the case that--

00:03:53.880 --> 00:03:58.350 align:middle line:84%
we will do this as
a major example--

00:03:58.350 --> 00:04:05.840 align:middle line:90%
in which the other component--

00:04:05.840 --> 00:04:09.980 align:middle line:84%
in this case, air-- so the
molecules of oxygen and nitrogen

00:04:09.980 --> 00:04:13.050 align:middle line:90%
in air are non-condensable.

00:04:13.050 --> 00:04:17.930 align:middle line:84%
So you do not find them
in the liquid phase.

00:04:17.930 --> 00:04:21.529 align:middle line:84%
And conversely,
the other component

00:04:21.529 --> 00:04:24.960 align:middle line:84%
in the liquid phase,
salt, is non-volatile.

00:04:24.960 --> 00:04:27.445 align:middle line:84%
So you don't find it
in the vapor phase.

00:04:27.445 --> 00:04:30.800 align:middle line:90%


00:04:30.800 --> 00:04:35.300 align:middle line:84%
That means that, for
mutual equilibrium,

00:04:35.300 --> 00:04:39.290 align:middle line:84%
it is required only that
the chemical potential

00:04:39.290 --> 00:04:43.040 align:middle line:84%
of the component that can be
exchanged between the two phases

00:04:43.040 --> 00:04:44.350 align:middle line:90%
be equal.

00:04:44.350 --> 00:04:49.760 align:middle line:84%
So it means that the chemical
potential of water in the liquid

00:04:49.760 --> 00:04:54.230 align:middle line:84%
must be equal to the chemical
potential of water vapor

00:04:54.230 --> 00:04:58.360 align:middle line:90%
in the gaseous phase.

00:04:58.360 --> 00:05:00.970 align:middle line:90%
And this is one example.

00:05:00.970 --> 00:05:09.290 align:middle line:84%
The other is liquid
over a solid.

00:05:09.290 --> 00:05:17.265 align:middle line:84%
So again, think of water
and salt. And this is ice.

00:05:17.265 --> 00:05:20.240 align:middle line:90%


00:05:20.240 --> 00:05:29.170 align:middle line:84%
So again, salt, we
assume, is not freezing.

00:05:29.170 --> 00:05:35.360 align:middle line:84%
So it's a solute of the liquid
solution, but it doesn't freeze.

00:05:35.360 --> 00:05:42.490 align:middle line:84%
And therefore, we
need only equality

00:05:42.490 --> 00:05:46.270 align:middle line:84%
of chemical potential of
component 1 between this

00:05:46.270 --> 00:05:46.990 align:middle line:90%
and this.

00:05:46.990 --> 00:05:48.860 align:middle line:84%
And this actually
is not a mixture.

00:05:48.860 --> 00:05:50.360 align:middle line:90%
It's just a pure component.

00:05:50.360 --> 00:05:51.840 align:middle line:90%
I mean, this is just pure ice.

00:05:51.840 --> 00:05:57.100 align:middle line:90%


00:05:57.100 --> 00:06:06.290 align:middle line:84%
And of course, this
example is very practical--

00:06:06.290 --> 00:06:15.730 align:middle line:84%
for example, when you
put salt on the sidewalks

00:06:15.730 --> 00:06:19.870 align:middle line:84%
for avoiding freezing, for
changing the temperature

00:06:19.870 --> 00:06:22.640 align:middle line:90%
at which freezing occurs.

00:06:22.640 --> 00:06:27.230 align:middle line:84%
And this is also you
put salt in water.

00:06:27.230 --> 00:06:32.860 align:middle line:84%
And that will change the
boiling point of the solution.

00:06:32.860 --> 00:06:43.420 align:middle line:84%
So when you make pasta and
you have salted your water,

00:06:43.420 --> 00:06:47.290 align:middle line:84%
the boiling point is
not 100 degrees Celsius.

00:06:47.290 --> 00:06:49.630 align:middle line:84%
It's just a little
bit more, depending

00:06:49.630 --> 00:06:50.835 align:middle line:90%
on how much salt you put.

00:06:50.835 --> 00:06:57.670 align:middle line:90%


00:06:57.670 --> 00:07:00.700 align:middle line:84%
So how do we derive
Raoult's law?

00:07:00.700 --> 00:07:04.630 align:middle line:84%
As usual, we only need to
apply the equality of chemical

00:07:04.630 --> 00:07:11.300 align:middle line:84%
potentials, and then manipulate
the equations somehow.

00:07:11.300 --> 00:07:15.710 align:middle line:84%
So let's say that we have two
phases that, for a moment,

00:07:15.710 --> 00:07:21.310 align:middle line:84%
we call A and B, just
generic two phases.

00:07:21.310 --> 00:07:26.260 align:middle line:84%
So the chemical potential
of component 1 in phase A--

00:07:26.260 --> 00:07:30.280 align:middle line:84%
you're going to
assume, to derive

00:07:30.280 --> 00:07:33.250 align:middle line:84%
Raoult's law, ideal
mixture or solution

00:07:33.250 --> 00:07:35.186 align:middle line:90%
behavior for all components--

00:07:35.186 --> 00:07:38.397 align:middle line:90%


00:07:38.397 --> 00:07:40.730 align:middle line:84%
or actually, for this component,
not for all components,

00:07:40.730 --> 00:07:42.050 align:middle line:90%
for this component.

00:07:42.050 --> 00:07:47.530 align:middle line:90%
And so this is the pure--

00:07:47.530 --> 00:07:51.730 align:middle line:84%
phase A pure
component 1 plus RT,

00:07:51.730 --> 00:07:58.000 align:middle line:84%
the log of the mole fraction
of component 1 in the phase.

00:07:58.000 --> 00:08:04.560 align:middle line:84%
And then we rewrite the chemical
potential at pressure p.

00:08:04.560 --> 00:08:08.400 align:middle line:84%
We rewrite it by
integrating the derivative

00:08:08.400 --> 00:08:14.215 align:middle line:84%
of the chemical potential with
respect to p from p to p-sat,

00:08:14.215 --> 00:08:16.427 align:middle line:84%
p-sat being the
saturation temperature

00:08:16.427 --> 00:08:18.010 align:middle line:84%
(I meant saturation
pressure) at which

00:08:18.010 --> 00:08:22.136 align:middle line:84%
the pure substance
at that temperature

00:08:22.136 --> 00:08:25.005 align:middle line:84%
changes (phase), or
boils, or freezes.

00:08:25.005 --> 00:08:29.850 align:middle line:90%


00:08:29.850 --> 00:08:35.220 align:middle line:84%
So we rewrite this as chemical
potential at another pressure.

00:08:35.220 --> 00:08:38.419 align:middle line:84%
And we pick the
saturation pressure.

00:08:38.419 --> 00:08:40.590 align:middle line:84%
And then we integrate
the derivative.

00:08:40.590 --> 00:08:45.430 align:middle line:84%
And remember, the derivative
is just the specific volume.

00:08:45.430 --> 00:08:51.230 align:middle line:84%
And then we do exactly the same
also for the other phase B,

00:08:51.230 --> 00:08:55.730 align:middle line:90%
just change the A's in B's.

00:08:55.730 --> 00:08:59.430 align:middle line:84%
Then we recall that when
we have a pure substance--

00:08:59.430 --> 00:09:01.040 align:middle line:84%
this is the reason
why we decided

00:09:01.040 --> 00:09:03.970 align:middle line:90%
to use the saturation pressure.

00:09:03.970 --> 00:09:09.470 align:middle line:84%
When we have a pure substance,
the saturation pressure

00:09:09.470 --> 00:09:14.300 align:middle line:84%
is defined precisely by that
pressure at which, at the given

00:09:14.300 --> 00:09:17.060 align:middle line:84%
temperature, the two chemical
potentials of the two phases

00:09:17.060 --> 00:09:19.280 align:middle line:84%
are equal, so they are
in mutual equilibrium.

00:09:19.280 --> 00:09:25.700 align:middle line:84%
So liquid and vapor, for
example, at the boiling point,

00:09:25.700 --> 00:09:29.480 align:middle line:84%
have equal temperature,
equal pressure,

00:09:29.480 --> 00:09:31.100 align:middle line:90%
and equal chemical potential.

00:09:31.100 --> 00:09:32.730 align:middle line:84%
But the three are
not independent.

00:09:32.730 --> 00:09:36.620 align:middle line:84%
And the equality of
chemical potential

00:09:36.620 --> 00:09:43.390 align:middle line:84%
defines the saturation
pressure at that temperature.

00:09:43.390 --> 00:09:48.730 align:middle line:84%
So that means that this and that
cancel because they are equal.

00:09:48.730 --> 00:09:52.420 align:middle line:84%
The moment we equate the
two chemical potentials.

00:09:52.420 --> 00:09:53.650 align:middle line:90%
These two will cancel.

00:09:53.650 --> 00:09:56.690 align:middle line:84%
And we are left with
just the remaining terms.

00:09:56.690 --> 00:09:58.818 align:middle line:90%
And so these remaining terms.

00:09:58.818 --> 00:10:01.490 align:middle line:84%
The part with the logarithms
that we put together.

00:10:01.490 --> 00:10:02.980 align:middle line:84%
The logarithm
becomes the logarithm

00:10:02.980 --> 00:10:05.380 align:middle line:84%
of the ratio of
the mole fractions.

00:10:05.380 --> 00:10:09.640 align:middle line:84%
And the integral, you
can put it in this form.

00:10:09.640 --> 00:10:11.530 align:middle line:84%
It's the integral
of the difference

00:10:11.530 --> 00:10:17.440 align:middle line:84%
between the specific volumes of
the pure phases, pure substance

00:10:17.440 --> 00:10:20.860 align:middle line:90%
in the two different phases.

00:10:20.860 --> 00:10:28.470 align:middle line:84%
So this is what you would
call vfg, for example, or vjf.

00:10:28.470 --> 00:10:33.640 align:middle line:84%
It's the volume of
vaporization or freezing

00:10:33.640 --> 00:10:38.200 align:middle line:90%
and so on of the pure substance.

00:10:38.200 --> 00:10:46.660 align:middle line:84%
So this connects the
compositions of these mixtures

00:10:46.660 --> 00:10:50.750 align:middle line:84%
to properties of
pure substances.

00:10:50.750 --> 00:10:54.340 align:middle line:90%


00:10:54.340 --> 00:11:01.930 align:middle line:84%
For example, if we
take this case, where

00:11:01.930 --> 00:11:09.580 align:middle line:84%
we have moist air
on salted water,

00:11:09.580 --> 00:11:11.770 align:middle line:90%
then, for air, we can assume--

00:11:11.770 --> 00:11:19.160 align:middle line:84%
for the specific volume, we
can assume the ideal gas law.

00:11:19.160 --> 00:11:21.110 align:middle line:90%
For the liquid, we can assume--

00:11:21.110 --> 00:11:24.990 align:middle line:84%
liquid water-- assume
incompressible fluid model

00:11:24.990 --> 00:11:30.440 align:middle line:84%
so that the volume is constant,
specific volume is constant.

00:11:30.440 --> 00:11:35.280 align:middle line:84%
Remember, these are
molar specific volumes.

00:11:35.280 --> 00:11:37.160 align:middle line:84%
And then you do
the integral here.

00:11:37.160 --> 00:11:39.810 align:middle line:90%


00:11:39.810 --> 00:11:43.420 align:middle line:90%
So do the integral for the--

00:11:43.420 --> 00:11:46.090 align:middle line:84%
this is the part
contributed by the fact

00:11:46.090 --> 00:11:48.230 align:middle line:90%
that the volume is constant.

00:11:48.230 --> 00:11:51.140 align:middle line:84%
So it's just the difference
of the limits of integration.

00:11:51.140 --> 00:11:54.700 align:middle line:84%
And the other is you
integrate dp over p

00:11:54.700 --> 00:11:58.190 align:middle line:90%
so you get the logarithm of p.

00:11:58.190 --> 00:12:03.360 align:middle line:90%
So this is the result.

00:12:03.360 --> 00:12:15.290 align:middle line:84%
Now, it turns out that this
term here is much smaller than 1

00:12:15.290 --> 00:12:21.110 align:middle line:84%
and therefore much smaller
than this other value, which

00:12:21.110 --> 00:12:24.420 align:middle line:90%
is the logarithm of--

00:12:24.420 --> 00:12:25.710 align:middle line:90%
RT times the logarithm.

00:12:25.710 --> 00:12:32.140 align:middle line:90%
So if you neglect that--

00:12:32.140 --> 00:12:34.880 align:middle line:84%
and you can compute it and
find that, for example,

00:12:34.880 --> 00:12:41.410 align:middle line:84%
even for very high pressures and
normal temperatures for liquid

00:12:41.410 --> 00:12:46.480 align:middle line:84%
water and typical temperatures,
this is of the order of less

00:12:46.480 --> 00:12:53.900 align:middle line:84%
than 1/100, which is definitely
enough lower than 1 to be

00:12:53.900 --> 00:12:55.260 align:middle line:90%
negligible.

00:12:55.260 --> 00:12:58.310 align:middle line:90%
And so you neglect that.

00:12:58.310 --> 00:12:59.870 align:middle line:84%
If you consider
that, it would be

00:12:59.870 --> 00:13:05.030 align:middle line:84%
called what is called the
Poynting effect, the Poynting

00:13:05.030 --> 00:13:05.790 align:middle line:90%
contribution.

00:13:05.790 --> 00:13:10.370 align:middle line:84%
So what we are doing
currently is normally

00:13:10.370 --> 00:13:13.980 align:middle line:84%
described as neglecting
the Poynting correction.

00:13:13.980 --> 00:13:16.840 align:middle line:90%


00:13:16.840 --> 00:13:19.920 align:middle line:84%
And Poynting is the
name of a person.

00:13:19.920 --> 00:13:29.940 align:middle line:90%


00:13:29.940 --> 00:13:33.360 align:middle line:90%
So the end result--

00:13:33.360 --> 00:13:39.070 align:middle line:84%
this is a square
bracket too many.

00:13:39.070 --> 00:13:43.810 align:middle line:84%
We find that log
and log can go away.

00:13:43.810 --> 00:13:46.690 align:middle line:84%
And you get that the
ratio of the mole fraction

00:13:46.690 --> 00:13:50.410 align:middle line:90%
of water in the air--

00:13:50.410 --> 00:13:52.660 align:middle line:90%
so that's the moisture content--

00:13:52.660 --> 00:14:01.040 align:middle line:84%
over the mole fraction
of water in the solution.

00:14:01.040 --> 00:14:02.760 align:middle line:90%
And here it is a solvent.

00:14:02.760 --> 00:14:06.830 align:middle line:90%
So it's the predominant.

00:14:06.830 --> 00:14:09.750 align:middle line:84%
You could also apply
it for pure water.

00:14:09.750 --> 00:14:15.280 align:middle line:90%
In this case, y1f is equal to 1.

00:14:15.280 --> 00:14:17.300 align:middle line:90%
Then that is equal--

00:14:17.300 --> 00:14:20.250 align:middle line:84%
and this is Raoult's
law-- to the ratio

00:14:20.250 --> 00:14:22.440 align:middle line:84%
of the pressure
over the saturation

00:14:22.440 --> 00:14:25.550 align:middle line:90%
pressure of pure water.

00:14:25.550 --> 00:14:26.970 align:middle line:90%
This is a formula--

00:14:26.970 --> 00:14:34.760 align:middle line:84%
You can also remember then that
since for an ideal gas mixture--

00:14:34.760 --> 00:14:38.810 align:middle line:84%
ideal Gibbs-Dalton mixture of
ideal gases the partial pressure

00:14:38.810 --> 00:14:43.460 align:middle line:84%
of the component is equal to the
mole fraction times the pressure

00:14:43.460 --> 00:14:44.400 align:middle line:90%
of the mixture.

00:14:44.400 --> 00:14:46.900 align:middle line:84%
This can also be written
as the partial pressure.

00:14:46.900 --> 00:14:49.950 align:middle line:90%


00:14:49.950 --> 00:14:54.330 align:middle line:84%
And so the partial pressure of
the component 1 in the mixture

00:14:54.330 --> 00:14:58.740 align:middle line:84%
is equal to the mole fraction in
the liquid times the saturation

00:14:58.740 --> 00:15:00.755 align:middle line:90%
pressure of component 1.

00:15:00.755 --> 00:15:06.380 align:middle line:90%


00:15:06.380 --> 00:15:09.780 align:middle line:84%
If you apply the
same formula here,

00:15:09.780 --> 00:15:14.240 align:middle line:84%
equation 1, for the case
of liquid and solid--

00:15:14.240 --> 00:15:17.290 align:middle line:90%
f and j for solid--

00:15:17.290 --> 00:15:20.200 align:middle line:90%
so that means ice--

00:15:20.200 --> 00:15:23.900 align:middle line:84%
both the volumes can be
considered constants.

00:15:23.900 --> 00:15:27.500 align:middle line:90%
So they get out of the integral.

00:15:27.500 --> 00:15:29.760 align:middle line:90%
And so you have the difference.

00:15:29.760 --> 00:15:32.600 align:middle line:84%
So the volume of
freezing, the change,

00:15:32.600 --> 00:15:39.410 align:middle line:84%
that little change in density
between pure water and ice

00:15:39.410 --> 00:15:42.840 align:middle line:84%
that makes it so that
ice floats on water,

00:15:42.840 --> 00:15:47.910 align:middle line:90%
that's the volume of freezing.

00:15:47.910 --> 00:15:55.380 align:middle line:84%
And so Raoult's law
takes this other form.

00:15:55.380 --> 00:16:01.650 align:middle line:84%
So it gives you
the mole fraction

00:16:01.650 --> 00:16:07.770 align:middle line:84%
of water in the
solution in terms

00:16:07.770 --> 00:16:10.560 align:middle line:90%
of the difference in pressure.

00:16:10.560 --> 00:16:14.200 align:middle line:84%
Saturation pressure this time--
this is the saturation pressure,

00:16:14.200 --> 00:16:18.720 align:middle line:84%
of course, of the
liquid-ice transition--

00:16:18.720 --> 00:16:25.990 align:middle line:84%
times that volume of
freezing divided by RT.

00:16:25.990 --> 00:16:32.420 align:middle line:84%
So with this, you can
compute these relations.

00:16:32.420 --> 00:16:37.510 align:middle line:84%
And these have to do
with the phase rule.

00:16:37.510 --> 00:16:40.775 align:middle line:84%
For example, in
this case, we have--

00:16:40.775 --> 00:16:44.730 align:middle line:90%


00:16:44.730 --> 00:16:50.936 align:middle line:84%
since it is a binary mixture,
it has just two components.

00:16:50.936 --> 00:17:01.480 align:middle line:90%


00:17:01.480 --> 00:17:09.140 align:middle line:84%
Of course, the second mole
fraction is 1 minus the first.

00:17:09.140 --> 00:17:12.430 align:middle line:84%
And so you have one
relation between these two

00:17:12.430 --> 00:17:16.180 align:middle line:84%
and the same, another
relation between these.

00:17:16.180 --> 00:17:24.290 align:middle line:84%
And then we have the relations
between the chemical potentials

00:17:24.290 --> 00:17:27.920 align:middle line:90%
that remove two other relations.

00:17:27.920 --> 00:17:30.680 align:middle line:84%
And so you are
left with the fact

00:17:30.680 --> 00:17:38.060 align:middle line:84%
that the moment you fix the
temperature and the pressure,

00:17:38.060 --> 00:17:40.840 align:middle line:90%
everything is set.

00:17:40.840 --> 00:17:45.930 align:middle line:84%
So the mole fractions
are fixed, like here.

00:17:45.930 --> 00:17:52.140 align:middle line:84%
You cannot change arbitrarily
the mole fraction and still stay

00:17:52.140 --> 00:17:54.640 align:middle line:90%
in mutual equilibrium.

00:17:54.640 --> 00:17:58.150 align:middle line:90%
And here also, you cannot--

00:17:58.150 --> 00:18:04.000 align:middle line:84%
the ratio of the mole fraction
of the liquid and the vapor

00:18:04.000 --> 00:18:08.490 align:middle line:84%
is fixed by the given
temperature and pressure.

00:18:08.490 --> 00:18:16.610 align:middle line:90%


00:18:16.610 --> 00:18:27.880 align:middle line:84%
You can interpret, in a
molecular way, the Raoult's law.

00:18:27.880 --> 00:18:30.724 align:middle line:90%


00:18:30.724 --> 00:18:37.320 align:middle line:84%
See, Raoult's law
says, for example, here

00:18:37.320 --> 00:18:42.030 align:middle line:84%
we are considering a different
situation than the one

00:18:42.030 --> 00:18:48.230 align:middle line:84%
before in which not
only component 1,

00:18:48.230 --> 00:18:52.600 align:middle line:90%
but also component 2, can go--

00:18:52.600 --> 00:18:54.750 align:middle line:90%
can cross phase.

00:18:54.750 --> 00:18:58.250 align:middle line:90%


00:18:58.250 --> 00:19:01.030 align:middle line:84%
For example, in the
fg case, it can be

00:19:01.030 --> 00:19:05.160 align:middle line:90%
both volatile and condensable.

00:19:05.160 --> 00:19:10.400 align:middle line:84%
This would be,
for example, wine.

00:19:10.400 --> 00:19:19.290 align:middle line:84%
Wine is a solution
of alcohol and water.

00:19:19.290 --> 00:19:24.550 align:middle line:84%
And when you boil that, you'll
get a vapor in which you

00:19:24.550 --> 00:19:27.130 align:middle line:90%
have both alcohol and water.

00:19:27.130 --> 00:19:31.340 align:middle line:84%
But they are in
different compositions.

00:19:31.340 --> 00:19:33.260 align:middle line:84%
So when they are in
mutual equilibrium,

00:19:33.260 --> 00:19:38.800 align:middle line:84%
the mole fraction of
alcohol in the liquid phase

00:19:38.800 --> 00:19:40.900 align:middle line:84%
is different from
the mole fraction

00:19:40.900 --> 00:19:43.335 align:middle line:90%
of alcohol in the vapor phase.

00:19:43.335 --> 00:19:47.630 align:middle line:90%


00:19:47.630 --> 00:19:53.600 align:middle line:84%
So Raoult's law gives you--
which is written here again--

00:19:53.600 --> 00:19:58.910 align:middle line:84%
partial pressure
of, say, water is

00:19:58.910 --> 00:20:03.340 align:middle line:84%
equal to the mole fraction
times the saturation pressure.

00:20:03.340 --> 00:20:05.930 align:middle line:84%
And if you have two components,
you have to Raoult's law,

00:20:05.930 --> 00:20:07.670 align:middle line:84%
one also for the
second component,

00:20:07.670 --> 00:20:10.640 align:middle line:84%
because now also component
2 can be exchanged,

00:20:10.640 --> 00:20:13.240 align:middle line:84%
you should write also the
equality of chemical potential

00:20:13.240 --> 00:20:16.448 align:middle line:90%
for that component.

00:20:16.448 --> 00:20:17.740 align:middle line:90%
So you get the second equation.

00:20:17.740 --> 00:20:20.330 align:middle line:90%


00:20:20.330 --> 00:20:30.830 align:middle line:84%
And you see here the idea,
the molecular interpretation,

00:20:30.830 --> 00:20:31.680 align:middle line:90%
is this.

00:20:31.680 --> 00:20:40.010 align:middle line:84%
And if you have a pure
substance at equilibrium,

00:20:40.010 --> 00:20:44.610 align:middle line:90%
there is somehow an exchange--

00:20:44.610 --> 00:20:47.320 align:middle line:90%


00:20:47.320 --> 00:20:50.560 align:middle line:84%
an equilibrated--
a balanced exchange

00:20:50.560 --> 00:20:54.100 align:middle line:84%
of molecules that are in
the vapor phase that go down

00:20:54.100 --> 00:20:57.898 align:middle line:84%
in the liquid and an
equal amount of molecules

00:20:57.898 --> 00:20:59.440 align:middle line:84%
that, from the
surface of the liquid,

00:20:59.440 --> 00:21:04.210 align:middle line:90%
go up in the vapor phase.

00:21:04.210 --> 00:21:06.070 align:middle line:90%
And this is for component 1.

00:21:06.070 --> 00:21:07.580 align:middle line:90%
And this is for component 2.

00:21:07.580 --> 00:21:13.940 align:middle line:84%
Now, when you have a mixture,
you see that, on the surface,

00:21:13.940 --> 00:21:19.450 align:middle line:84%
you have some dark and
some gray particles.

00:21:19.450 --> 00:21:25.900 align:middle line:84%
So this dynamic exchange that
occurs between the liquid and--

00:21:25.900 --> 00:21:29.890 align:middle line:84%
occurs only proportional
to the amount

00:21:29.890 --> 00:21:38.160 align:middle line:84%
of dark particles for the dark
ones that are on the surface.

00:21:38.160 --> 00:21:42.960 align:middle line:84%
So if you assume that the
fraction of particles that

00:21:42.960 --> 00:21:45.990 align:middle line:84%
are on the surface is
equal to the mole fraction

00:21:45.990 --> 00:21:49.490 align:middle line:84%
in the liquid
phase, then you see

00:21:49.490 --> 00:21:58.130 align:middle line:84%
that the partial pressure,
which is driving this exchange

00:21:58.130 --> 00:22:04.880 align:middle line:84%
with the black
particles, is only

00:22:04.880 --> 00:22:10.340 align:middle line:84%
limited by a fraction,
which is the mole fraction,

00:22:10.340 --> 00:22:15.140 align:middle line:84%
by the fraction with which they
occupy the surface and equally

00:22:15.140 --> 00:22:15.960 align:middle line:90%
the other ones.

00:22:15.960 --> 00:22:19.430 align:middle line:84%
And therefore, since
the pressures add up,

00:22:19.430 --> 00:22:21.950 align:middle line:90%
that's the relation.

00:22:21.950 --> 00:22:27.440 align:middle line:84%
And that is one way
of interpreting this.

00:22:27.440 --> 00:22:34.160 align:middle line:84%
In any case, when we have
the two Raoult's equations,

00:22:34.160 --> 00:22:42.270 align:middle line:84%
we can usefully combine them
by using the obvious fact

00:22:42.270 --> 00:22:46.210 align:middle line:84%
that the mole
fractions add up to 1.

00:22:46.210 --> 00:22:51.130 align:middle line:84%
So for example, I could
add this and that.

00:22:51.130 --> 00:22:57.220 align:middle line:84%
If I add these two, you
see y1g plus y2g is 1.

00:22:57.220 --> 00:22:59.850 align:middle line:90%
So I'm remaining with p.

00:22:59.850 --> 00:23:02.520 align:middle line:84%
So I obtain this
pressure-- this expression

00:23:02.520 --> 00:23:07.350 align:middle line:84%
that the pressure is
equal to this plus that.

00:23:07.350 --> 00:23:10.450 align:middle line:84%
And of course, instead of
y2f, I write 1 minus y1f.

00:23:10.450 --> 00:23:12.095 align:middle line:90%
This is called the bubble line.

00:23:12.095 --> 00:23:14.780 align:middle line:90%


00:23:14.780 --> 00:23:18.950 align:middle line:84%
It expresses, for
a given pressure

00:23:18.950 --> 00:23:24.920 align:middle line:84%
and temperature, the
composition of the liquid phase.

00:23:24.920 --> 00:23:27.480 align:middle line:84%
So if I give you the
temperature and the pressure,

00:23:27.480 --> 00:23:31.460 align:middle line:84%
you can compute the
mole fraction y1f

00:23:31.460 --> 00:23:35.480 align:middle line:84%
because you can go on the
tables of the pure substances

00:23:35.480 --> 00:23:39.480 align:middle line:84%
for the p-sat of
the two substances.

00:23:39.480 --> 00:23:42.573 align:middle line:84%
And then with this
equation, you solve for y1f.

00:23:42.573 --> 00:23:54.940 align:middle line:90%


00:23:54.940 --> 00:24:00.130 align:middle line:84%
Or, if you solve this
equation for y1f and y2f

00:24:00.130 --> 00:24:02.930 align:middle line:84%
and then sum up so
that it's equal to 1,

00:24:02.930 --> 00:24:08.390 align:middle line:84%
this is the other
expression that you obtain.

00:24:08.390 --> 00:24:09.810 align:middle line:90%
This is the dew line.

00:24:09.810 --> 00:24:14.120 align:middle line:84%
It contains the relation
that gives you--

00:24:14.120 --> 00:24:19.730 align:middle line:84%
that allows you to solve for
the mole fraction of component 1

00:24:19.730 --> 00:24:20.490 align:middle line:90%
in--

00:24:20.490 --> 00:24:28.630 align:middle line:84%
so the composition of the
vapor in the two-phase state.

00:24:28.630 --> 00:24:31.706 align:middle line:90%


00:24:31.706 --> 00:24:35.450 align:middle line:84%
And of course, once you
have the bubble line,

00:24:35.450 --> 00:24:37.010 align:middle line:84%
this is really an
equation of a line.

00:24:37.010 --> 00:24:43.180 align:middle line:84%
So you see that if you plot
pressure versus mole fraction--

00:24:43.180 --> 00:24:45.370 align:middle line:84%
in this case, we're
talking about-- if you're

00:24:45.370 --> 00:24:50.020 align:middle line:84%
doing this equation here, it's
the mole fraction of the liquid.

00:24:50.020 --> 00:24:54.600 align:middle line:84%
You get this straight line
that we call the bubble line.

00:24:54.600 --> 00:24:58.760 align:middle line:84%
An so for a given temperature
and a given pressure,

00:24:58.760 --> 00:25:04.170 align:middle line:84%
you go here and read the
mole fraction of the liquid.

00:25:04.170 --> 00:25:08.190 align:middle line:84%
The dew line is
not a straight line

00:25:08.190 --> 00:25:15.140 align:middle line:84%
because it's 1 over p
that is linear in y.

00:25:15.140 --> 00:25:17.510 align:middle line:90%
So it has this shape.

00:25:17.510 --> 00:25:21.120 align:middle line:84%
And so for that temperature
and the given pressure,

00:25:21.120 --> 00:25:25.570 align:middle line:84%
you go down here and read the
mole fraction of the vapor.

00:25:25.570 --> 00:25:28.000 align:middle line:84%
Of course, you can
eliminate pressure

00:25:28.000 --> 00:25:30.160 align:middle line:84%
from the system of
these two equations.

00:25:30.160 --> 00:25:31.990 align:middle line:84%
And then you get
a direct relation

00:25:31.990 --> 00:25:35.770 align:middle line:84%
between the composition
of the vapor phase

00:25:35.770 --> 00:25:43.730 align:middle line:84%
and the composition of the
liquid phase, where what counts

00:25:43.730 --> 00:25:48.410 align:middle line:84%
is the ratio between the
saturation pressures of the two

00:25:48.410 --> 00:25:51.246 align:middle line:84%
substances, which is called
the relative volatility.

00:25:51.246 --> 00:25:53.950 align:middle line:90%


00:25:53.950 --> 00:25:55.730 align:middle line:90%
And here is a plot.

00:25:55.730 --> 00:25:58.380 align:middle line:90%
So this is y1f.

00:25:58.380 --> 00:26:01.310 align:middle line:90%
This is y1g.

00:26:01.310 --> 00:26:03.490 align:middle line:84%
And this is how
that equation looks.

00:26:03.490 --> 00:26:06.080 align:middle line:90%


00:26:06.080 --> 00:26:10.580 align:middle line:84%
Of course, remember that
here, if we have written

00:26:10.580 --> 00:26:17.840 align:middle line:84%
this for both substance
1 and substance 2,

00:26:17.840 --> 00:26:23.110 align:middle line:84%
it means that we have assumed
that at all compositions,

00:26:23.110 --> 00:26:28.840 align:middle line:84%
the liquid behaves
as an ideal solution,

00:26:28.840 --> 00:26:32.460 align:middle line:90%
as an ideal solution, yes.

00:26:32.460 --> 00:26:36.530 align:middle line:84%
And so this is a very
strong assumption

00:26:36.530 --> 00:26:44.240 align:middle line:84%
because most liquids do
not behave like that.

00:26:44.240 --> 00:26:47.850 align:middle line:84%
But still, this is
a starting point.

00:26:47.850 --> 00:26:51.200 align:middle line:84%
So that's how the bubble
line and the dew line

00:26:51.200 --> 00:26:57.520 align:middle line:84%
look for an ideal model,
a completely ideal model.

00:26:57.520 --> 00:27:01.600 align:middle line:90%


00:27:01.600 --> 00:27:08.210 align:middle line:84%
And that is enough to
give an idea of what is

00:27:08.210 --> 00:27:14.270 align:middle line:90%
done when we do distillation.

00:27:14.270 --> 00:27:18.140 align:middle line:84%
You see, this plot is done
at constant temperature

00:27:18.140 --> 00:27:21.860 align:middle line:84%
because it's easy, because
you fix the temperature.

00:27:21.860 --> 00:27:26.270 align:middle line:84%
You go and read the saturation
pressures from these lines.

00:27:26.270 --> 00:27:36.260 align:middle line:84%
Of course, I can do
also numerically a plot

00:27:36.260 --> 00:27:41.890 align:middle line:84%
in which I fix the pressure,
and then for each temperature.

00:27:41.890 --> 00:27:47.300 align:middle line:84%
So if I fix the pressure
for each temperature,

00:27:47.300 --> 00:27:50.540 align:middle line:84%
I have to go and read the
values that will give me

00:27:50.540 --> 00:27:51.510 align:middle line:90%
the mole fractions.

00:27:51.510 --> 00:27:52.560 align:middle line:90%
And I do a plot.

00:27:52.560 --> 00:27:58.650 align:middle line:84%
And a plot will generally
not produce straight lines

00:27:58.650 --> 00:28:02.545 align:middle line:84%
any longer, if it is a
plot at fixed pressure.

00:28:02.545 --> 00:28:06.070 align:middle line:90%


00:28:06.070 --> 00:28:10.720 align:middle line:84%
This is more useful
because mostly you boil

00:28:10.720 --> 00:28:15.330 align:middle line:90%
things at a fixed pressure.

00:28:15.330 --> 00:28:18.180 align:middle line:90%
So here is the idea.

00:28:18.180 --> 00:28:26.315 align:middle line:84%
So we've got the bubble line
and the dew line, sorry--

00:28:26.315 --> 00:28:29.280 align:middle line:90%


00:28:29.280 --> 00:28:31.830 align:middle line:84%
the dew line here and
the bubble line here.

00:28:31.830 --> 00:28:41.720 align:middle line:84%
So suppose you start with a
mixture in the liquid phase

00:28:41.720 --> 00:28:46.800 align:middle line:90%
at this composition, say 50/50.

00:28:46.800 --> 00:28:48.820 align:middle line:90%
And I heat it up.

00:28:48.820 --> 00:28:54.340 align:middle line:84%
At a certain point, I
reach the bubble line.

00:28:54.340 --> 00:29:02.690 align:middle line:84%
And the first bubble
gets boiled off.

00:29:02.690 --> 00:29:06.570 align:middle line:84%
That bubble is in equilibrium
with my saturated liquid

00:29:06.570 --> 00:29:09.145 align:middle line:84%
and has a composition,
which is here.

00:29:09.145 --> 00:29:11.770 align:middle line:90%


00:29:11.770 --> 00:29:13.400 align:middle line:90%
It has a different composition.

00:29:13.400 --> 00:29:15.130 align:middle line:84%
And we compute it
by that formula

00:29:15.130 --> 00:29:23.267 align:middle line:84%
that we have derived based
on the relative volatility.

00:29:23.267 --> 00:29:25.350 align:middle line:84%
Of course, you can express
the relative volatility

00:29:25.350 --> 00:29:27.700 align:middle line:90%
as a ratio of pressures.

00:29:27.700 --> 00:29:30.720 align:middle line:84%
But it's also related to
the ratio of the saturation

00:29:30.720 --> 00:29:34.170 align:middle line:84%
temperatures because
this T-sat of p

00:29:34.170 --> 00:29:39.520 align:middle line:84%
is the inverse
function of p-sat of T.

00:29:39.520 --> 00:29:44.130 align:middle line:90%
If I keep heating up--

00:29:44.130 --> 00:29:50.970 align:middle line:84%
for example, I have an
equilibrium at this temperature.

00:29:50.970 --> 00:29:54.140 align:middle line:84%
And for an equilibrium
at this temperature,

00:29:54.140 --> 00:29:57.900 align:middle line:84%
the bubble line gives me the
mole fraction of the liquid.

00:29:57.900 --> 00:30:01.260 align:middle line:84%
And the dew line gives me the
mole fraction of the vapor,

00:30:01.260 --> 00:30:05.340 align:middle line:90%
again following this expression.

00:30:05.340 --> 00:30:10.490 align:middle line:84%
If the temperature
goes up to here,

00:30:10.490 --> 00:30:14.670 align:middle line:84%
I have saturated vapor,
which is in equilibrium

00:30:14.670 --> 00:30:19.870 align:middle line:84%
with the first or, say,
last drop of liquid,

00:30:19.870 --> 00:30:21.520 align:middle line:90%
so this is saturated vapor.

00:30:21.520 --> 00:30:24.740 align:middle line:84%
Then if I still heat
it up, it will go up.

00:30:24.740 --> 00:30:28.060 align:middle line:90%


00:30:28.060 --> 00:30:32.630 align:middle line:90%
So the idea is this.

00:30:32.630 --> 00:30:36.470 align:middle line:84%
If I start with a composition
like 50/50 and I want

00:30:36.470 --> 00:30:43.490 align:middle line:84%
to separate, what I could do
is to go up to this temperature

00:30:43.490 --> 00:30:47.850 align:middle line:84%
here, like an intermediate
temperature between

00:30:47.850 --> 00:30:51.150 align:middle line:84%
the temperature-- the
saturation liquid temperature

00:30:51.150 --> 00:30:54.540 align:middle line:84%
and the saturation
vapor temperature.

00:30:54.540 --> 00:31:05.340 align:middle line:84%
And notice that whereas
for a pure substance,

00:31:05.340 --> 00:31:08.010 align:middle line:84%
the dew line and the bubble
line coincide and there

00:31:08.010 --> 00:31:13.740 align:middle line:84%
is only one temperature at
which the pure substance becomes

00:31:13.740 --> 00:31:18.480 align:middle line:90%
liquid and is also saturated.

00:31:18.480 --> 00:31:26.720 align:middle line:84%
So this lens-shaped thing
will reduce to just a line.

00:31:26.720 --> 00:31:29.540 align:middle line:90%


00:31:29.540 --> 00:31:34.830 align:middle line:84%
But here, the boiling occurs
over a range of temperatures

00:31:34.830 --> 00:31:38.390 align:middle line:90%
between this and that.

00:31:38.390 --> 00:31:40.320 align:middle line:84%
This is not only for
liquid and solid.

00:31:40.320 --> 00:31:47.330 align:middle line:84%
For example, in also metals,
if you go metallurgy,

00:31:47.330 --> 00:31:49.590 align:middle line:84%
this would be solid,
this would be liquid,

00:31:49.590 --> 00:31:54.660 align:middle line:84%
and this would be the
"solidus" temperature.

00:31:54.660 --> 00:31:56.250 align:middle line:84%
And that's the
"liquidus" temperature.

00:31:56.250 --> 00:32:01.670 align:middle line:90%
And this gap is like here.

00:32:01.670 --> 00:32:06.130 align:middle line:84%
OK, so let's say we
put the temperature--

00:32:06.130 --> 00:32:09.730 align:middle line:84%
we added the temperature
intermediate between

00:32:09.730 --> 00:32:13.630 align:middle line:84%
the "liquidus" and the
"vaporus," so to speak,

00:32:13.630 --> 00:32:14.800 align:middle line:90%
temperature.

00:32:14.800 --> 00:32:18.760 align:middle line:84%
Then you see that
the two phases are

00:32:18.760 --> 00:32:24.160 align:middle line:84%
one enriched in component
2 and the other enriched

00:32:24.160 --> 00:32:25.170 align:middle line:90%
in component 1.

00:32:25.170 --> 00:32:28.050 align:middle line:90%


00:32:28.050 --> 00:32:35.410 align:middle line:84%
So this is what happens in one
stage of a distillation column.

00:32:35.410 --> 00:32:38.740 align:middle line:84%
Then what you do is
you take the liquid

00:32:38.740 --> 00:32:44.540 align:middle line:84%
and you send it down to
another stage, which works

00:32:44.540 --> 00:32:47.270 align:middle line:90%
at a different temperature.

00:32:47.270 --> 00:32:52.190 align:middle line:84%
And then you take the
vapor and you send it up

00:32:52.190 --> 00:33:05.740 align:middle line:84%
to condense into the upper stage
and so on so that eventually you

00:33:05.740 --> 00:33:06.240 align:middle line:90%
can--

00:33:06.240 --> 00:33:09.460 align:middle line:90%


00:33:09.460 --> 00:33:14.120 align:middle line:84%
since in the upper level,
you collected a mixture,

00:33:14.120 --> 00:33:24.900 align:middle line:90%
which is more rich in alcohol.

00:33:24.900 --> 00:33:30.990 align:middle line:84%
Then you could
play the same game,

00:33:30.990 --> 00:33:34.730 align:middle line:84%
go to an intermediate
temperature, like this one.

00:33:34.730 --> 00:33:43.620 align:middle line:84%
And then you generate an even
more enriched vapor and so on.

00:33:43.620 --> 00:33:45.770 align:middle line:84%
You can continue as much
as you need in order

00:33:45.770 --> 00:33:51.710 align:middle line:90%
to obtain up to 99%.

00:33:51.710 --> 00:34:01.640 align:middle line:84%
For example, this is how they
get 96 volume percent alcohol

00:34:01.640 --> 00:34:04.740 align:middle line:90%
that is used to make liqueurs.

00:34:04.740 --> 00:34:14.270 align:middle line:90%


00:34:14.270 --> 00:34:18.230 align:middle line:84%
All right, so this is for
the distillation principle.

00:34:18.230 --> 00:34:24.154 align:middle line:84%
Now, another thing that can
be explained by Raoult's law

00:34:24.154 --> 00:34:28.389 align:middle line:84%
is, as I mentioned,
this idea that--

00:34:28.389 --> 00:34:36.659 align:middle line:84%
this fact that when
you add salt on water,

00:34:36.659 --> 00:34:41.030 align:middle line:84%
that water will freeze
at a lower temperature

00:34:41.030 --> 00:34:44.130 align:middle line:90%
than 0 Celsius.

00:34:44.130 --> 00:34:47.820 align:middle line:84%
Or when you add salt
in water, that water

00:34:47.820 --> 00:34:51.706 align:middle line:84%
will boil at a higher
temperature than 100 degrees

00:34:51.706 --> 00:34:52.206 align:middle line:90%
Celsius.

00:34:52.206 --> 00:34:55.730 align:middle line:90%


00:34:55.730 --> 00:34:57.510 align:middle line:90%
How do we derive that?

00:34:57.510 --> 00:35:02.250 align:middle line:84%
OK, we start from the Raoult's
law for the specific case.

00:35:02.250 --> 00:35:07.340 align:middle line:84%
So this is the Raoult's
law for the liquid-ice case

00:35:07.340 --> 00:35:09.790 align:middle line:84%
that we have seen a
couple of slides ago.

00:35:09.790 --> 00:35:12.720 align:middle line:90%


00:35:12.720 --> 00:35:16.990 align:middle line:84%
And we take the
differential of this.

00:35:16.990 --> 00:35:21.600 align:middle line:90%


00:35:21.600 --> 00:35:31.570 align:middle line:84%
And then realize that when we
take the differential of p-sat,

00:35:31.570 --> 00:35:32.840 align:middle line:90%
we actually--

00:35:32.840 --> 00:35:35.390 align:middle line:90%


00:35:35.390 --> 00:35:37.860 align:middle line:84%
this is p-sat of
the pure substance.

00:35:37.860 --> 00:35:40.340 align:middle line:84%
So for the pure
substance, we can

00:35:40.340 --> 00:35:45.200 align:middle line:84%
remember the Clausius-Clapeyron
relation that gives you

00:35:45.200 --> 00:35:49.350 align:middle line:90%
the slope of this curve.

00:35:49.350 --> 00:35:54.450 align:middle line:84%
On the pT diagram, this is
the saturation pressure curve.

00:35:54.450 --> 00:35:58.080 align:middle line:84%
And the slope of
that curve is given

00:35:58.080 --> 00:35:59.890 align:middle line:84%
by the Clausius-Clapeyron
relation.

00:35:59.890 --> 00:36:02.940 align:middle line:84%
And it is related
to the enthalpy

00:36:02.940 --> 00:36:05.550 align:middle line:84%
of the change of phase
divided by the volume

00:36:05.550 --> 00:36:07.740 align:middle line:90%
of the change of phase.

00:36:07.740 --> 00:36:11.190 align:middle line:84%
Actually, it should be
the entropy of the change

00:36:11.190 --> 00:36:12.727 align:middle line:90%
of phase divided by the volume.

00:36:12.727 --> 00:36:14.310 align:middle line:84%
But the entropy of
the change of phase

00:36:14.310 --> 00:36:16.830 align:middle line:84%
is equal to the enthalpy
over the temperature

00:36:16.830 --> 00:36:20.170 align:middle line:90%
of the change of phase.

00:36:20.170 --> 00:36:23.710 align:middle line:84%
So you use this
in this relation.

00:36:23.710 --> 00:36:31.420 align:middle line:84%
Neglect this term
here, which is good

00:36:31.420 --> 00:36:34.710 align:middle line:90%
if you have a dilute solution.

00:36:34.710 --> 00:36:39.510 align:middle line:84%
Therefore, the mole
fraction is close to 1.

00:36:39.510 --> 00:36:42.910 align:middle line:84%
And therefore, the
logarithm is close to 0.

00:36:42.910 --> 00:36:46.240 align:middle line:84%
And you're left with
these three terms.

00:36:46.240 --> 00:36:51.770 align:middle line:84%
Then if you consider that you're
not changing the pressure--

00:36:51.770 --> 00:36:54.200 align:middle line:84%
so if you remain at
constant pressure

00:36:54.200 --> 00:37:02.680 align:middle line:84%
and change the composition
of the solution,

00:37:02.680 --> 00:37:04.930 align:middle line:84%
this tells you how the
temperature will change.

00:37:04.930 --> 00:37:07.630 align:middle line:84%
So essentially, from
this equation, the p,

00:37:07.630 --> 00:37:10.780 align:middle line:84%
you can compute
the derivative of T

00:37:10.780 --> 00:37:15.290 align:middle line:84%
with respect to a change of
mole fraction of constituent 1.

00:37:15.290 --> 00:37:17.830 align:middle line:84%
But the change of mole
fraction of constituent 1

00:37:17.830 --> 00:37:21.970 align:middle line:84%
is also equal to the negative of
the change of the constituent 2,

00:37:21.970 --> 00:37:30.390 align:middle line:84%
in this case the salt. And
then you have h and RT.

00:37:30.390 --> 00:37:40.290 align:middle line:84%
So in the end, you find that
if you add a little bit of salt

00:37:40.290 --> 00:37:44.370 align:middle line:84%
to the liquid solution, the
temperature at which you have

00:37:44.370 --> 00:37:51.090 align:middle line:84%
equilibrium, therefore
the freezing temperature,

00:37:51.090 --> 00:37:55.770 align:middle line:84%
changes from before
by an amount that

00:37:55.770 --> 00:37:57.780 align:middle line:90%
is related by these things--

00:37:57.780 --> 00:38:00.660 align:middle line:84%
RT squared, the enthalpy,
so all properties

00:38:00.660 --> 00:38:03.720 align:middle line:84%
that you can look up for the
pure substance with a minus

00:38:03.720 --> 00:38:04.680 align:middle line:90%
sign.

00:38:04.680 --> 00:38:10.140 align:middle line:84%
So this is the freezing
point depression case.

00:38:10.140 --> 00:38:14.700 align:middle line:90%


00:38:14.700 --> 00:38:17.670 align:middle line:84%
You can also plot
it on this diagram.

00:38:17.670 --> 00:38:18.956 align:middle line:90%
So if you have a solution--

00:38:18.956 --> 00:38:22.400 align:middle line:90%


00:38:22.400 --> 00:38:24.650 align:middle line:90%
here, I enlarged this area.

00:38:24.650 --> 00:38:29.210 align:middle line:84%
So the black curves
are the pure substance.

00:38:29.210 --> 00:38:34.320 align:middle line:84%
If you add a little bit
of salt in the solution,

00:38:34.320 --> 00:38:39.560 align:middle line:84%
then this black curve
shifts to the blue curve

00:38:39.560 --> 00:38:42.780 align:middle line:90%
at lower temperature.

00:38:42.780 --> 00:38:48.940 align:middle line:84%
And then also the same
for the other side,

00:38:48.940 --> 00:38:55.140 align:middle line:84%
for the other case,
liquid over vapor,

00:38:55.140 --> 00:38:59.433 align:middle line:90%
so boiling of a salted solution.

00:38:59.433 --> 00:39:02.450 align:middle line:90%


00:39:02.450 --> 00:39:06.200 align:middle line:84%
Here, we have to start from
the Raoult's law written

00:39:06.200 --> 00:39:10.530 align:middle line:84%
for these two phases, f
and g, which is this one.

00:39:10.530 --> 00:39:12.000 align:middle line:90%
Again, we do the differential.

00:39:12.000 --> 00:39:15.280 align:middle line:84%
Again, we use the
Clausius-Clapeyron relation.

00:39:15.280 --> 00:39:20.380 align:middle line:84%
And again, we obtain a
partial derivative of T

00:39:20.380 --> 00:39:27.900 align:middle line:84%
at constant pressure, how that
boiling temperature changes

00:39:27.900 --> 00:39:34.120 align:middle line:84%
when you add a little amount of
component 2, which is the salt.

00:39:34.120 --> 00:39:36.530 align:middle line:90%
This time, you have a plus sign.

00:39:36.530 --> 00:39:38.660 align:middle line:84%
The ingredients on
the left-hand side

00:39:38.660 --> 00:39:41.010 align:middle line:84%
look similar, except
that, of course,

00:39:41.010 --> 00:39:45.300 align:middle line:84%
you have to put here the
enthalpy of vaporization,

00:39:45.300 --> 00:39:51.210 align:middle line:84%
whereas here you have
the enthalpy of melting,

00:39:51.210 --> 00:39:53.270 align:middle line:90%
and the plus sign.

00:39:53.270 --> 00:39:56.480 align:middle line:84%
So the curve, the
shift is on the right.

00:39:56.480 --> 00:39:59.275 align:middle line:84%
So the boiling
temperature is higher.

00:39:59.275 --> 00:40:04.170 align:middle line:90%


00:40:04.170 --> 00:40:17.770 align:middle line:84%
Actually, these boiling
and freezing effects

00:40:17.770 --> 00:40:23.890 align:middle line:84%
can also be used as
means to measure--

00:40:23.890 --> 00:40:28.070 align:middle line:84%
are actually used as means
to measure the salt content.

00:40:28.070 --> 00:40:30.880 align:middle line:84%
For example, if
you take seawater

00:40:30.880 --> 00:40:33.410 align:middle line:84%
and if you want to
measure the salt content,

00:40:33.410 --> 00:40:37.630 align:middle line:84%
you run the experiment of
freezing it and boiling it.

00:40:37.630 --> 00:40:41.890 align:middle line:84%
And then with that,
you can compute--

00:40:41.890 --> 00:40:49.120 align:middle line:84%
you see here, by measuring
the delta T, that is related

00:40:49.120 --> 00:40:55.970 align:middle line:90%
to the amount of salt content.

00:40:55.970 --> 00:41:00.830 align:middle line:84%
And therefore, you infer
from those measurements,

00:41:00.830 --> 00:41:04.425 align:middle line:84%
these composition
and salt content.

00:41:04.425 --> 00:41:12.960 align:middle line:90%


00:41:12.960 --> 00:41:13.460 align:middle line:90%
OK.

00:41:13.460 --> 00:41:16.490 align:middle line:84%
Again, the problem
is that, of course,

00:41:16.490 --> 00:41:25.490 align:middle line:84%
ideal mixtures are good
models for the gaseous phases,

00:41:25.490 --> 00:41:29.740 align:middle line:84%
not so good for
the liquid phases,

00:41:29.740 --> 00:41:33.250 align:middle line:84%
and therefore not so good
for the liquid equilibrium

00:41:33.250 --> 00:41:39.280 align:middle line:84%
cases that we are interested
in because the reality is more

00:41:39.280 --> 00:41:41.635 align:middle line:90%
complex than the ideal models.

00:41:41.635 --> 00:41:47.410 align:middle line:90%


00:41:47.410 --> 00:41:51.070 align:middle line:84%
So we remind ourselves
of some results

00:41:51.070 --> 00:41:55.720 align:middle line:84%
that are useful in any
problem with mixtures.

00:41:55.720 --> 00:41:57.580 align:middle line:90%
They're going to be useful here.

00:41:57.580 --> 00:42:01.260 align:middle line:84%
And they are going to
be useful also from,

00:42:01.260 --> 00:42:04.120 align:middle line:84%
I think, the next time
when we start the systems

00:42:04.120 --> 00:42:07.530 align:middle line:90%
with chemical reactions.

00:42:07.530 --> 00:42:13.260 align:middle line:84%
And these expressions
are important,

00:42:13.260 --> 00:42:15.930 align:middle line:84%
namely that the
properties are given

00:42:15.930 --> 00:42:18.780 align:middle line:84%
by the weighted sum
weighted on the amounts

00:42:18.780 --> 00:42:23.830 align:middle line:84%
of the partial properties,
denoted lowercase

00:42:23.830 --> 00:42:28.540 align:middle line:90%
with a single subscript.

00:42:28.540 --> 00:42:31.720 align:middle line:84%
And then these
partial properties

00:42:31.720 --> 00:42:36.970 align:middle line:84%
can be related to the specific
properties of pure substances

00:42:36.970 --> 00:42:44.925 align:middle line:84%
if you add what we called
the mixing property.

00:42:44.925 --> 00:42:48.830 align:middle line:90%


00:42:48.830 --> 00:42:53.280 align:middle line:84%
But now, what I
want to go back to,

00:42:53.280 --> 00:42:58.110 align:middle line:84%
is this result that
we haven't used yet.

00:42:58.110 --> 00:43:04.230 align:middle line:84%
It's the Duhem-Margules relation
that we obtained when we

00:43:04.230 --> 00:43:08.100 align:middle line:90%
did this for the simple system.

00:43:08.100 --> 00:43:17.800 align:middle line:84%
It was one of the various, say,
byproducts of this derivation.

00:43:17.800 --> 00:43:26.880 align:middle line:84%
So the Duhem-Margules
relation is this one.

00:43:26.880 --> 00:43:29.430 align:middle line:90%


00:43:29.430 --> 00:43:35.940 align:middle line:84%
And it is important when
you try to develop models

00:43:35.940 --> 00:43:40.470 align:middle line:84%
for non-ideal behavior of
the stable equilibrium state

00:43:40.470 --> 00:43:43.940 align:middle line:90%
properties of mixtures.

00:43:43.940 --> 00:43:45.420 align:middle line:90%
Why is it important?

00:43:45.420 --> 00:43:52.000 align:middle line:84%
Because it imposes rules
about the interrelations

00:43:52.000 --> 00:43:54.180 align:middle line:90%
that must exist.

00:43:54.180 --> 00:43:56.370 align:middle line:84%
You see, this is a
system of equations.

00:43:56.370 --> 00:43:57.600 align:middle line:90%
It's not just one.

00:43:57.600 --> 00:44:01.830 align:middle line:84%
There is one for
every component.

00:44:01.830 --> 00:44:05.970 align:middle line:90%
And so it's not that you--

00:44:05.970 --> 00:44:10.120 align:middle line:84%
you see, when you want to
model properties of substances,

00:44:10.120 --> 00:44:22.590 align:middle line:84%
you have to come up
with correlations

00:44:22.590 --> 00:44:24.160 align:middle line:90%
for these functions.

00:44:24.160 --> 00:44:27.360 align:middle line:90%


00:44:27.360 --> 00:44:35.160 align:middle line:84%
And you might be tempted to
use any sort of analytical form

00:44:35.160 --> 00:44:37.610 align:middle line:90%
of these interrelations.

00:44:37.610 --> 00:44:43.310 align:middle line:84%
But Duhem-Margules says no
or, at least, you may try,

00:44:43.310 --> 00:44:46.910 align:middle line:84%
but you should check the
consistency with this relation

00:44:46.910 --> 00:44:51.260 align:middle line:84%
because not all relations
are good for that purpose

00:44:51.260 --> 00:44:56.200 align:middle line:84%
because some violate
this condition.

00:44:56.200 --> 00:45:01.060 align:middle line:84%
Actually, this condition--
remember that mu i comma j

00:45:01.060 --> 00:45:04.390 align:middle line:84%
is the derivative of the
chemical potential of component

00:45:04.390 --> 00:45:08.470 align:middle line:84%
i with respect to the
amount of component j.

00:45:08.470 --> 00:45:10.270 align:middle line:84%
And so it is the
second derivative

00:45:10.270 --> 00:45:13.070 align:middle line:84%
of the Gibbs free
energy with respect

00:45:13.070 --> 00:45:19.850 align:middle line:84%
to two different components
or also equal components.

00:45:19.850 --> 00:45:23.560 align:middle line:90%


00:45:23.560 --> 00:45:27.850 align:middle line:84%
Now, you can view this
system of equations

00:45:27.850 --> 00:45:32.590 align:middle line:84%
as a homogeneous system
of linear equations

00:45:32.590 --> 00:45:43.220 align:middle line:84%
in the unknowns, n, where
mu i j is the matrix

00:45:43.220 --> 00:45:48.260 align:middle line:90%
of the linear coefficients.

00:45:48.260 --> 00:45:51.840 align:middle line:84%
And if you go back
into your math courses,

00:45:51.840 --> 00:45:54.560 align:middle line:84%
you will find that the system
of homogeneous equations

00:45:54.560 --> 00:45:59.150 align:middle line:84%
has a solution only
if the determinant

00:45:59.150 --> 00:46:05.110 align:middle line:84%
of the matrix of the
coefficients is equal to 0.

00:46:05.110 --> 00:46:10.450 align:middle line:84%
But this determinant--
this matrix

00:46:10.450 --> 00:46:17.800 align:middle line:84%
is the matrix that we call
the Hess because, you see,

00:46:17.800 --> 00:46:21.080 align:middle line:84%
it's the matrix of the
second derivatives.

00:46:21.080 --> 00:46:25.400 align:middle line:84%
And its determinant
is also something

00:46:25.400 --> 00:46:26.850 align:middle line:90%
that we have encountered.

00:46:26.850 --> 00:46:32.560 align:middle line:84%
It's the Jacobian of
the chemical potentials

00:46:32.560 --> 00:46:35.140 align:middle line:84%
with respect to
the compositions,

00:46:35.140 --> 00:46:40.040 align:middle line:84%
of course, done at
constant T and p.

00:46:40.040 --> 00:46:43.200 align:middle line:84%
So you need that Jacobian
to be equal to 0.

00:46:43.200 --> 00:46:44.730 align:middle line:90%
So that is the condition.

00:46:44.730 --> 00:46:48.930 align:middle line:84%
So you cannot arbitrarily
select these functions.

00:46:48.930 --> 00:46:51.955 align:middle line:84%
You have to check that the
Jacobian be equal to 0.

00:46:51.955 --> 00:46:54.730 align:middle line:90%


00:46:54.730 --> 00:47:02.720 align:middle line:84%
For example, as
you know, we have

00:47:02.720 --> 00:47:13.060 align:middle line:84%
decided to call this
functional relation the ideal.

00:47:13.060 --> 00:47:17.490 align:middle line:90%


00:47:17.490 --> 00:47:33.720 align:middle line:84%
And in fact, we're going to call
this mu i ideal of T, p, and y.

00:47:33.720 --> 00:47:35.915 align:middle line:84%
This is a particular
functional relation.

00:47:35.915 --> 00:47:38.630 align:middle line:90%


00:47:38.630 --> 00:47:51.460 align:middle line:84%
So we have to check that it
satisfies the Duhem-Margules

00:47:51.460 --> 00:47:53.460 align:middle line:90%
relation.

00:47:53.460 --> 00:47:58.236 align:middle line:84%
So here is the ideal
mixture assumption.

00:47:58.236 --> 00:48:03.040 align:middle line:90%


00:48:03.040 --> 00:48:06.160 align:middle line:84%
So in order to do
this derivative,

00:48:06.160 --> 00:48:09.370 align:middle line:84%
computing this
derivative, you have

00:48:09.370 --> 00:48:14.870 align:middle line:84%
to be a bit careful in
doing those derivatives.

00:48:14.870 --> 00:48:22.370 align:middle line:90%


00:48:22.370 --> 00:48:24.660 align:middle line:84%
I forget now if we
have done this already.

00:48:24.660 --> 00:48:28.370 align:middle line:84%
But the key point
here is that you

00:48:28.370 --> 00:48:32.000 align:middle line:84%
have to take the derivative
of mu with respect--

00:48:32.000 --> 00:48:33.690 align:middle line:84%
at fixed temperature
and pressure,

00:48:33.690 --> 00:48:35.130 align:middle line:90%
so this doesn't contribute.

00:48:35.130 --> 00:48:38.740 align:middle line:84%
It's only the logarithm
of the mole fraction.

00:48:38.740 --> 00:48:42.460 align:middle line:84%
So the logarithm of
the mole fraction,

00:48:42.460 --> 00:48:48.533 align:middle line:84%
you have to take its derivative
with respect to the nj's.

00:48:48.533 --> 00:48:53.190 align:middle line:90%


00:48:53.190 --> 00:48:59.420 align:middle line:84%
Now nj, it's easier
to write it this way.

00:48:59.420 --> 00:49:02.210 align:middle line:84%
So I'm taking the
derivative with respect

00:49:02.210 --> 00:49:05.970 align:middle line:84%
to nj of the logarithm
of the mole fraction.

00:49:05.970 --> 00:49:14.120 align:middle line:84%
The mole fraction of component i
is ni divided by the sum of all

00:49:14.120 --> 00:49:14.830 align:middle line:90%
the n's.

00:49:14.830 --> 00:49:19.740 align:middle line:90%
Let's use another subscript, k.

00:49:19.740 --> 00:49:24.570 align:middle line:84%
All right, so that
means that sometimes you

00:49:24.570 --> 00:49:30.870 align:middle line:84%
find you have a contribution
of the numerator when

00:49:30.870 --> 00:49:34.750 align:middle line:90%
j is equal to i.

00:49:34.750 --> 00:49:39.010 align:middle line:84%
And that's why I use
the Kronecker delta.

00:49:39.010 --> 00:49:46.050 align:middle line:84%
And then that
contribution is simply

00:49:46.050 --> 00:49:55.465 align:middle line:84%
1 over the mole fraction,
something like that.

00:49:55.465 --> 00:50:08.650 align:middle line:90%


00:50:08.650 --> 00:50:11.330 align:middle line:90%
Sorry, not correct.

00:50:11.330 --> 00:50:16.360 align:middle line:84%
It's n sub i because
it's the logarithm.

00:50:16.360 --> 00:50:23.110 align:middle line:84%
The derivative of the logarithm
is 1 over the argument.

00:50:23.110 --> 00:50:29.140 align:middle line:84%
And then sometimes,
the derivative

00:50:29.140 --> 00:50:32.320 align:middle line:84%
picks up something
from the denominator.

00:50:32.320 --> 00:50:35.630 align:middle line:84%
The denominator, this could be
viewed as-- that is a constant.

00:50:35.630 --> 00:50:40.130 align:middle line:90%
So that is minus log of this.

00:50:40.130 --> 00:50:45.375 align:middle line:90%
So it would be minus 1 over n.

00:50:45.375 --> 00:50:48.550 align:middle line:90%


00:50:48.550 --> 00:50:52.010 align:middle line:90%
Yeah, so that's written here.

00:50:52.010 --> 00:50:55.820 align:middle line:84%
Now that you've taken
this derivative,

00:50:55.820 --> 00:50:59.100 align:middle line:90%
you can substitute into here.

00:50:59.100 --> 00:51:03.920 align:middle line:84%
So you sum over n and
verify that, indeed, that

00:51:03.920 --> 00:51:07.050 align:middle line:90%
is equal to 0 identically.

00:51:07.050 --> 00:51:13.380 align:middle line:84%
So obviously, people
have been careful in--

00:51:13.380 --> 00:51:15.970 align:middle line:84%
people and also ourselves,
when we derived this,

00:51:15.970 --> 00:51:18.330 align:middle line:84%
we did it using the
rules of the game.

00:51:18.330 --> 00:51:27.370 align:middle line:84%
And therefore, it does obey
the Duhem-Margules relation.

00:51:27.370 --> 00:51:31.900 align:middle line:84%
But when you start
correlating experimental data

00:51:31.900 --> 00:51:37.600 align:middle line:84%
for non-ideal behavior,
you have to be

00:51:37.600 --> 00:51:46.426 align:middle line:84%
careful in cooking up functional
relations that are compatible.

00:51:46.426 --> 00:51:49.110 align:middle line:90%


00:51:49.110 --> 00:51:57.520 align:middle line:84%
And if you remember now
that-- so here it is.

00:51:57.520 --> 00:52:02.880 align:middle line:84%
So if we are going to decide
to write the actual chemical

00:52:02.880 --> 00:52:09.280 align:middle line:84%
potential as a sum of
the ideal plus an excess.

00:52:09.280 --> 00:52:12.240 align:middle line:84%
So in other words, we are
going to call excess chemical

00:52:12.240 --> 00:52:16.290 align:middle line:84%
potential the difference between
the actual chemical potential

00:52:16.290 --> 00:52:21.900 align:middle line:90%
and what is this ideal one.

00:52:21.900 --> 00:52:29.370 align:middle line:84%
So essentially, we're going to
say, for the real substance,

00:52:29.370 --> 00:52:40.350 align:middle line:84%
for the real mixture, we need to
define how the behavior departs

00:52:40.350 --> 00:52:42.960 align:middle line:90%
from ideal.

00:52:42.960 --> 00:52:45.340 align:middle line:84%
And this is called excess
chemical potential.

00:52:45.340 --> 00:52:49.630 align:middle line:90%


00:52:49.630 --> 00:53:00.440 align:middle line:84%
Remember also this-- chemical
potentials of the pure substance

00:53:00.440 --> 00:53:09.510 align:middle line:84%
are equal to the
Gibbs free energy,

00:53:09.510 --> 00:53:12.160 align:middle line:84%
to the specific
Gibbs free energy.

00:53:12.160 --> 00:53:13.880 align:middle line:90%
This will come up.

00:53:13.880 --> 00:53:16.440 align:middle line:90%


00:53:16.440 --> 00:53:33.470 align:middle line:84%
Also, since we have defined
these properties of mixing,

00:53:33.470 --> 00:53:35.870 align:middle line:84%
by the way we define
them, it turns out

00:53:35.870 --> 00:53:40.490 align:middle line:84%
that what we decided to call
the excess chemical potential

00:53:40.490 --> 00:53:45.440 align:middle line:84%
is equal to the difference
between the enthalpy of mixing

00:53:45.440 --> 00:53:49.800 align:middle line:84%
and the entropy of mixing
multiplied by the temperature.

00:53:49.800 --> 00:53:54.800 align:middle line:90%


00:53:54.800 --> 00:54:01.160 align:middle line:84%
Not quite, only the part of
the entropy of mixing that

00:54:01.160 --> 00:54:07.720 align:middle line:84%
is not already included
in the RT log of y.

00:54:07.720 --> 00:54:14.210 align:middle line:84%
So you see, here, we
call that delta s ex.

00:54:14.210 --> 00:54:21.520 align:middle line:84%
So it is an excess
entropy of mixing

00:54:21.520 --> 00:54:26.140 align:middle line:84%
because, in the ideal solution,
there is a mixing effect that

00:54:26.140 --> 00:54:27.490 align:middle line:90%
accounts for that.

00:54:27.490 --> 00:54:30.265 align:middle line:84%
Remember the
surprise discussion?

00:54:30.265 --> 00:54:33.440 align:middle line:90%
It accounts with R log of y.

00:54:33.440 --> 00:54:34.460 align:middle line:90%
This is the surprise.

00:54:34.460 --> 00:54:43.570 align:middle line:84%
And this is a part of the
entropy, of the mixing entropy.

00:54:43.570 --> 00:54:46.630 align:middle line:84%
It's the only mixing
entropy that we

00:54:46.630 --> 00:54:50.390 align:middle line:90%
have under ideal behavior.

00:54:50.390 --> 00:54:56.520 align:middle line:84%
And you may have some
extra, some excess one.

00:54:56.520 --> 00:55:03.780 align:middle line:84%
So that's why we say here
chemical potential excess is

00:55:03.780 --> 00:55:07.560 align:middle line:84%
equal to the entire
enthalpy of mixing

00:55:07.560 --> 00:55:12.670 align:middle line:84%
because, for the ideal mixture,
there is no enthalpy of mixing

00:55:12.670 --> 00:55:18.620 align:middle line:84%
and only this excess part
of the entropy of mixing.

00:55:18.620 --> 00:55:26.059 align:middle line:90%


00:55:26.059 --> 00:55:32.650 align:middle line:84%
So with that in mind,
of course, another thing

00:55:32.650 --> 00:55:36.760 align:middle line:84%
that we should remember is
that the Gibbs free energy,

00:55:36.760 --> 00:55:39.760 align:middle line:84%
if we are interested
in the Gibbs

00:55:39.760 --> 00:55:49.885 align:middle line:84%
free energy of our entire
mixture, that is equal to this.

00:55:49.885 --> 00:55:54.080 align:middle line:90%


00:55:54.080 --> 00:56:01.940 align:middle line:84%
And so if you want, you can also
define parallel with the fact

00:56:01.940 --> 00:56:06.890 align:middle line:84%
that we decided to write
the chemical potential as mu

00:56:06.890 --> 00:56:11.390 align:middle line:90%
i ideal plus mu i excess.

00:56:11.390 --> 00:56:21.980 align:middle line:84%
You can also plug that into here
and obtain an overall mixture,

00:56:21.980 --> 00:56:26.645 align:middle line:84%
Gibbs free energy as a sum of an
ideal part and an excess part.

00:56:26.645 --> 00:56:32.726 align:middle line:90%


00:56:32.726 --> 00:56:38.120 align:middle line:90%
With that in mind, here is--

00:56:38.120 --> 00:56:40.600 align:middle line:84%
for example, for
a binary mixture,

00:56:40.600 --> 00:56:44.876 align:middle line:84%
let's fix these
limit for ourselves.

00:56:44.876 --> 00:56:48.140 align:middle line:90%


00:56:48.140 --> 00:56:52.950 align:middle line:84%
Binary mixture, here
is the expression

00:56:52.950 --> 00:56:57.420 align:middle line:84%
for the specific
Gibbs free energy.

00:56:57.420 --> 00:57:05.350 align:middle line:84%
So now, if I take this
G and divide it by n,

00:57:05.350 --> 00:57:07.500 align:middle line:90%
that gives me G over n.

00:57:07.500 --> 00:57:10.610 align:middle line:90%


00:57:10.610 --> 00:57:14.750 align:middle line:84%
Of course, that summation
becomes the mole fractions'

00:57:14.750 --> 00:57:17.330 align:middle line:90%
weighted chemical potentials.

00:57:17.330 --> 00:57:21.305 align:middle line:90%
And this becomes gid plus gex.

00:57:21.305 --> 00:57:35.290 align:middle line:90%


00:57:35.290 --> 00:57:40.950 align:middle line:84%
Now, gid is therefore
the weighted sum

00:57:40.950 --> 00:57:45.150 align:middle line:84%
of the ideal chemical potentials
weighted on the mole fractions.

00:57:45.150 --> 00:57:47.390 align:middle line:84%
And the ideal chemical
potentials are these ones.

00:57:47.390 --> 00:57:51.010 align:middle line:90%


00:57:51.010 --> 00:57:53.780 align:middle line:90%
So if I take that--

00:57:53.780 --> 00:57:58.150 align:middle line:84%
let's see if I can do
this game like this.

00:57:58.150 --> 00:58:06.860 align:middle line:84%
Take the weighted sum of these
ones, that gives me g small id.

00:58:06.860 --> 00:58:11.280 align:middle line:90%


00:58:11.280 --> 00:58:16.430 align:middle line:84%
So you see that
in this summation,

00:58:16.430 --> 00:58:18.240 align:middle line:90%
I have two contributions.

00:58:18.240 --> 00:58:21.500 align:middle line:84%
One is the weighted sum
on the mole fractions

00:58:21.500 --> 00:58:24.020 align:middle line:84%
of the pure component
chemical potentials, which

00:58:24.020 --> 00:58:27.350 align:middle line:84%
are equal to the pure
component Gibbs free energies.

00:58:27.350 --> 00:58:34.700 align:middle line:84%
And that is this plus this
because, in this case,

00:58:34.700 --> 00:58:40.032 align:middle line:84%
binary mixture, the sum is
just limited to two components.

00:58:40.032 --> 00:58:44.280 align:middle line:84%
And then I have this other
summation that is here.

00:58:44.280 --> 00:58:47.010 align:middle line:84%
It's RT times the
sum of y log y.

00:58:47.010 --> 00:58:51.295 align:middle line:84%
That's the entropy of mixing,
isothermobaric mixing.

00:58:51.295 --> 00:58:53.850 align:middle line:90%


00:58:53.850 --> 00:58:54.890 align:middle line:90%
So that's the gid.

00:58:54.890 --> 00:59:00.070 align:middle line:90%


00:59:00.070 --> 00:59:03.520 align:middle line:84%
And these are the ideal
chemical potentials.

00:59:03.520 --> 00:59:04.970 align:middle line:90%
And then we recall that.

00:59:04.970 --> 00:59:07.706 align:middle line:90%


00:59:07.706 --> 00:59:12.296 align:middle line:84%
Now, below here, I have listed
some of the historical--

00:59:12.296 --> 00:59:16.820 align:middle line:90%


00:59:16.820 --> 00:59:22.730 align:middle line:84%
guesses or semi-empirical
theories that

00:59:22.730 --> 00:59:28.350 align:middle line:84%
have been developed
in order to come up

00:59:28.350 --> 00:59:34.380 align:middle line:84%
with ways of representing
the excess part,

00:59:34.380 --> 00:59:40.460 align:middle line:84%
so the non-ideal part of
the behavior of mixtures.

00:59:40.460 --> 00:59:46.710 align:middle line:90%
The simplest one, by Margules--

00:59:46.710 --> 00:59:53.490 align:middle line:90%
you see, this is 1895.

00:59:53.490 --> 01:00:08.884 align:middle line:84%
Yeah, my grandfather was
1889, so he was six years old.

01:00:08.884 --> 01:00:18.340 align:middle line:84%
So gex is-- so it's assumed
to be-- it's like symmetric--

01:00:18.340 --> 01:00:20.730 align:middle line:84%
a single parameter
which may depend

01:00:20.730 --> 01:00:23.530 align:middle line:90%
on temperature and pressure.

01:00:23.530 --> 01:00:26.840 align:middle line:84%
But the dependence
on composition

01:00:26.840 --> 01:00:29.730 align:middle line:84%
is just the product of
the two mole fractions.

01:00:29.730 --> 01:00:33.810 align:middle line:90%


01:00:33.810 --> 01:00:39.480 align:middle line:84%
And if you do the proper
derivatives in order to obtain

01:00:39.480 --> 01:00:46.930 align:middle line:84%
the excess chemical potential
that derives from that,

01:00:46.930 --> 01:00:51.340 align:middle line:84%
you find that the excess
chemical potential of component

01:00:51.340 --> 01:00:54.940 align:middle line:84%
1 is equal to that
function-- to that--

01:00:54.940 --> 01:01:02.270 align:middle line:84%
let's call it constant
A, constant in terms

01:01:02.270 --> 01:01:04.220 align:middle line:84%
that it doesn't
depend on composition,

01:01:04.220 --> 01:01:09.000 align:middle line:84%
times the square of the mole
fraction of the other component

01:01:09.000 --> 01:01:09.860 align:middle line:90%
and the same here.

01:01:09.860 --> 01:01:13.630 align:middle line:90%


01:01:13.630 --> 01:01:19.780 align:middle line:84%
If you go down below here,
I have written an expression

01:01:19.780 --> 01:01:29.110 align:middle line:84%
that we have derived a
while ago that relates

01:01:29.110 --> 01:01:35.790 align:middle line:84%
the enthalpy of mixing to
the chemical potential.

01:01:35.790 --> 01:01:41.260 align:middle line:90%


01:01:41.260 --> 01:01:48.177 align:middle line:84%
Now, actually, since for
the ideal chemical potential

01:01:48.177 --> 01:01:49.510 align:middle line:90%
there is no enthalpy of mixing--

01:01:49.510 --> 01:01:57.200 align:middle line:90%


01:01:57.200 --> 01:02:06.650 align:middle line:84%
OK, if I take this expression
here and do this derivative,

01:02:06.650 --> 01:02:08.990 align:middle line:90%
I obtain--

01:02:08.990 --> 01:02:09.490 align:middle line:90%
let's see.

01:02:09.490 --> 01:02:13.660 align:middle line:90%


01:02:13.660 --> 01:02:36.230 align:middle line:84%
So that's A y2 over T
with respect to 1 over T.

01:02:36.230 --> 01:02:39.290 align:middle line:84%
So it's equal to the
chemical potential itself.

01:02:39.290 --> 01:02:41.370 align:middle line:90%
Yeah, there it is.

01:02:41.370 --> 01:02:45.220 align:middle line:90%


01:02:45.220 --> 01:02:48.310 align:middle line:84%
Because you divide by T and
take the derivative with respect

01:02:48.310 --> 01:02:51.240 align:middle line:84%
to 1 over T, it's
just the 1 over T.

01:02:51.240 --> 01:02:53.010 align:middle line:90%
So the derivative is simply 1.

01:02:53.010 --> 01:02:59.210 align:middle line:90%


01:02:59.210 --> 01:03:02.860 align:middle line:90%
The enthalpy of mixing is Ay2.

01:03:02.860 --> 01:03:09.450 align:middle line:84%
So in the limit when
y2 is equal to 1,

01:03:09.450 --> 01:03:13.670 align:middle line:84%
it means that the
constant A represents

01:03:13.670 --> 01:03:16.460 align:middle line:90%
the enthalpy of mixing.

01:03:16.460 --> 01:03:19.600 align:middle line:90%


01:03:19.600 --> 01:03:28.770 align:middle line:84%
And remember, the
enthalpy of mixing

01:03:28.770 --> 01:03:30.900 align:middle line:84%
is the change in
enthalpy when you

01:03:30.900 --> 01:03:44.460 align:middle line:84%
add a little bit of substance
1 into a mixture in which

01:03:44.460 --> 01:03:49.380 align:middle line:90%
the predominant component is 2.

01:03:49.380 --> 01:04:01.045 align:middle line:84%
That's why we evaluate
it at y2 equals 1.

01:04:01.045 --> 01:04:03.920 align:middle line:90%


01:04:03.920 --> 01:04:07.760 align:middle line:84%
But you can also find the
same for the component 1.

01:04:07.760 --> 01:04:10.580 align:middle line:84%
So it turns out that
constant A is also

01:04:10.580 --> 01:04:17.480 align:middle line:84%
equal to the mixing enthalpy
of component 2 in pure 1.

01:04:17.480 --> 01:04:21.620 align:middle line:84%
So essentially, this says that
the Margules one-parameter model

01:04:21.620 --> 01:04:28.130 align:middle line:84%
is able to model only liquid
mixtures in which the effect

01:04:28.130 --> 01:04:33.620 align:middle line:84%
of adding component 1 to pure 2
is equal to the effect of adding

01:04:33.620 --> 01:04:35.540 align:middle line:90%
component 2 to pure 1.

01:04:35.540 --> 01:04:40.540 align:middle line:84%
So in other words, this
mixture is very symmetric.

01:04:40.540 --> 01:04:43.655 align:middle line:84%
Molecules are very
similar to one another.

01:04:43.655 --> 01:04:46.400 align:middle line:90%


01:04:46.400 --> 01:04:50.390 align:middle line:84%
So he developed a
two-parameter model.

01:04:50.390 --> 01:04:52.350 align:middle line:90%
It's slightly more complicated.

01:04:52.350 --> 01:04:54.530 align:middle line:84%
But at least, I
have two parameters.

01:04:54.530 --> 01:04:56.450 align:middle line:84%
And these two
parameters turn out

01:04:56.450 --> 01:05:01.340 align:middle line:84%
to be related to the mixing
enthalpies of the two

01:05:01.340 --> 01:05:07.460 align:middle line:84%
components, which
makes more sense

01:05:07.460 --> 01:05:12.530 align:middle line:84%
because, if I have
a bunch of water

01:05:12.530 --> 01:05:16.680 align:middle line:84%
and put some alcohol in it,
I obtain a different effect

01:05:16.680 --> 01:05:20.410 align:middle line:84%
than if I had a bunch of alcohol
and I put some water in it.

01:05:20.410 --> 01:05:26.980 align:middle line:90%


01:05:26.980 --> 01:05:28.940 align:middle line:84%
And so this is Margules
two parameters.

01:05:28.940 --> 01:05:32.390 align:middle line:84%
Van Laar also is
a two parameter,

01:05:32.390 --> 01:05:38.170 align:middle line:84%
some sort of an inverse
of Margules two parameter.

01:05:38.170 --> 01:05:49.860 align:middle line:84%
It also, of course, satisfies
the Duhem-Margules condition.

01:05:49.860 --> 01:05:55.350 align:middle line:84%
And again, this A and B
are related to the delta--

01:05:55.350 --> 01:05:57.960 align:middle line:84%
when you write it this
way, they are related

01:05:57.960 --> 01:05:59.736 align:middle line:90%
to the mixing enthalpies.

01:05:59.736 --> 01:06:06.820 align:middle line:90%


01:06:06.820 --> 01:06:12.580 align:middle line:84%
But two parameters,
for many mixtures

01:06:12.580 --> 01:06:18.000 align:middle line:84%
are not enough because
molecules may be very different.

01:06:18.000 --> 01:06:20.640 align:middle line:84%
You may have long
molecules, small molecules.

01:06:20.640 --> 01:06:25.880 align:middle line:84%
You have to account for the fact
that maybe a long molecule is

01:06:25.880 --> 01:06:28.650 align:middle line:84%
surrounded by many
small molecules.

01:06:28.650 --> 01:06:32.800 align:middle line:84%
And so you need more
parameters for reality.

01:06:32.800 --> 01:06:35.670 align:middle line:90%


01:06:35.670 --> 01:06:40.500 align:middle line:84%
And people over and over
have developed models

01:06:40.500 --> 01:06:41.770 align:middle line:90%
with more parameters.

01:06:41.770 --> 01:06:44.400 align:middle line:84%
They all satisfy the
Duhem-Margules relation.

01:06:44.400 --> 01:06:49.230 align:middle line:84%
For example, this one was
considered by Guggenheim.

01:06:49.230 --> 01:06:52.630 align:middle line:84%
And then it became famous under
the name of Redlich-Kister.

01:06:52.630 --> 01:06:55.450 align:middle line:90%


01:06:55.450 --> 01:07:00.330 align:middle line:84%
This other one was
developed by Wilson.

01:07:00.330 --> 01:07:06.750 align:middle line:84%
It's not the same Wilson
as the cloud chamber.

01:07:06.750 --> 01:07:14.680 align:middle line:84%
And based on this model,
developments and improvements

01:07:14.680 --> 01:07:18.566 align:middle line:84%
of the Wilson model, by
Prausnitz and co-workers.

01:07:18.566 --> 01:07:21.670 align:middle line:90%


01:07:21.670 --> 01:07:25.540 align:middle line:84%
You have the models
that you find now

01:07:25.540 --> 01:07:35.490 align:middle line:84%
in your softwares
that do modeling,

01:07:35.490 --> 01:07:43.070 align:middle line:84%
which have several parameters
which are related--

01:07:43.070 --> 01:07:50.570 align:middle line:84%
these epsilons are related to
the Lennard-Jones potential,

01:07:50.570 --> 01:07:55.190 align:middle line:84%
to the strength with
which molecule 1 is

01:07:55.190 --> 01:07:57.650 align:middle line:90%
attracted by molecule 1.

01:07:57.650 --> 01:08:06.430 align:middle line:84%
And this is molecule 1 attracted
by molecule 2 and 2 by 2.

01:08:06.430 --> 01:08:10.120 align:middle line:90%
And so you see, there is some--

01:08:10.120 --> 01:08:17.050 align:middle line:84%
the physics behind this model
is still related to the fact

01:08:17.050 --> 01:08:20.290 align:middle line:84%
that we're trying to
describe intermolecular

01:08:20.290 --> 01:08:23.479 align:middle line:90%
forces and their effects.

01:08:23.479 --> 01:08:28.145 align:middle line:84%
And these are the specific
volume of the pure components.

01:08:28.145 --> 01:08:36.600 align:middle line:90%


01:08:36.600 --> 01:08:44.960 align:middle line:84%
So in practice, you
describe chemical potentials

01:08:44.960 --> 01:08:51.140 align:middle line:84%
and, therefore, the Gibbs free
energy of a mixture in this way,

01:08:51.140 --> 01:08:54.510 align:middle line:90%
by adding up contributions.

01:08:54.510 --> 01:08:58.520 align:middle line:84%
So here are some
examples, I mean

01:08:58.520 --> 01:09:03.060 align:middle line:84%
examples of how these
contributions look,

01:09:03.060 --> 01:09:09.569 align:middle line:84%
for a binary mixture, as a
function of the mole fraction

01:09:09.569 --> 01:09:10.594 align:middle line:90%
of component 1.

01:09:10.594 --> 01:09:15.590 align:middle line:90%


01:09:15.590 --> 01:09:30.900 align:middle line:84%
For example, just the term y log
y is this pinkish curve here.

01:09:30.900 --> 01:09:34.460 align:middle line:90%


01:09:34.460 --> 01:09:37.645 align:middle line:84%
If you add also
this other term--

01:09:37.645 --> 01:09:41.330 align:middle line:90%


01:09:41.330 --> 01:09:44.569 align:middle line:84%
and of course, it depends
on what assumptions

01:09:44.569 --> 01:09:50.609 align:middle line:84%
you take for the Gibbs free
energies of the pure components.

01:09:50.609 --> 01:09:54.810 align:middle line:84%
Here, I assumed 0
for component 2.

01:09:54.810 --> 01:10:06.490 align:middle line:84%
And this number here may be
0.5 or 0.6 for component 1,

01:10:06.490 --> 01:10:10.240 align:middle line:90%
just to make this.

01:10:10.240 --> 01:10:12.870 align:middle line:90%
And this is the linear part.

01:10:12.870 --> 01:10:15.790 align:middle line:84%
The first two terms are
linear in the mole fraction.

01:10:15.790 --> 01:10:18.370 align:middle line:90%
And that's the blue curve.

01:10:18.370 --> 01:10:22.420 align:middle line:84%
So you add up the blue
curve and the pinkish one.

01:10:22.420 --> 01:10:25.070 align:middle line:90%
And you get the red curve.

01:10:25.070 --> 01:10:31.010 align:middle line:90%
That is g-ideal, just like that.

01:10:31.010 --> 01:10:33.960 align:middle line:90%


01:10:33.960 --> 01:10:39.530 align:middle line:84%
And then here, I plotted
the Margules one parameter,

01:10:39.530 --> 01:10:46.120 align:middle line:84%
that's the green; Margules
two parameters, the black;

01:10:46.120 --> 01:10:48.765 align:middle line:90%
and van Laar, the orange.

01:10:48.765 --> 01:10:57.280 align:middle line:90%


01:10:57.280 --> 01:11:02.880 align:middle line:84%
Now, when you add up the
ideal and the excess Gibbs

01:11:02.880 --> 01:11:05.150 align:middle line:90%
free energies--

01:11:05.150 --> 01:11:07.530 align:middle line:84%
see, this is negative,
this is positive--

01:11:07.530 --> 01:11:08.490 align:middle line:90%
they add up.

01:11:08.490 --> 01:11:12.970 align:middle line:90%
And here is what you obtain.

01:11:12.970 --> 01:11:14.950 align:middle line:90%
These are for the three cases--

01:11:14.950 --> 01:11:20.110 align:middle line:84%
Margules one parameter,
two parameters, van Laar.

01:11:20.110 --> 01:11:22.610 align:middle line:84%
You obtain shapes
that are non-trivial.

01:11:22.610 --> 01:11:25.970 align:middle line:90%


01:11:25.970 --> 01:11:31.280 align:middle line:84%
And actually, in
some cases, they're

01:11:31.280 --> 01:11:35.810 align:middle line:84%
also no good in the sense that
they violate the stability

01:11:35.810 --> 01:11:48.660 align:middle line:84%
conditions because the
second derivative of g

01:11:48.660 --> 01:11:53.010 align:middle line:90%
must always be positive.

01:11:53.010 --> 01:12:00.660 align:middle line:84%
We'll come back to
this in a short while.

01:12:00.660 --> 01:12:07.460 align:middle line:84%
So here, we will find
where the convexity is--

01:12:07.460 --> 01:12:09.240 align:middle line:84%
the second derivative
is negative,

01:12:09.240 --> 01:12:12.820 align:middle line:90%
this will have unstable states.

01:12:12.820 --> 01:12:17.350 align:middle line:84%
And this will drive us
to exactly the same ideas

01:12:17.350 --> 01:12:19.940 align:middle line:84%
that we have seen for
the van der Waals case,

01:12:19.940 --> 01:12:21.940 align:middle line:90%
for the pure substance.

01:12:21.940 --> 01:12:26.625 align:middle line:84%
Unstable states and also
metastable states are possible.

01:12:26.625 --> 01:12:29.860 align:middle line:90%


01:12:29.860 --> 01:12:33.015 align:middle line:84%
But we postpone
that for a while.

01:12:33.015 --> 01:12:36.640 align:middle line:90%


01:12:36.640 --> 01:12:46.430 align:middle line:84%
When we are in the
business of mixtures,

01:12:46.430 --> 01:12:49.010 align:middle line:84%
there are other
terms that I only

01:12:49.010 --> 01:12:55.260 align:middle line:84%
want to mention because they
are used by the practitioners.

01:12:55.260 --> 01:12:57.250 align:middle line:90%
And that might be confusing.

01:12:57.250 --> 01:13:00.755 align:middle line:84%
So I want to link
to what we're doing.

01:13:00.755 --> 01:13:03.900 align:middle line:90%


01:13:03.900 --> 01:13:10.250 align:middle line:84%
These are called-- for
the condensed phases,

01:13:10.250 --> 01:13:14.300 align:middle line:84%
we use activities and
activity coefficients.

01:13:14.300 --> 01:13:20.970 align:middle line:84%
It's just another way of
writing the chemical potential,

01:13:20.970 --> 01:13:23.920 align:middle line:90%
so to speak.

01:13:23.920 --> 01:13:26.400 align:middle line:84%
You could define the
activity this way.

01:13:26.400 --> 01:13:29.170 align:middle line:90%


01:13:29.170 --> 01:13:32.430 align:middle line:84%
So you take the chemical
potential, this one.

01:13:32.430 --> 01:13:35.170 align:middle line:90%


01:13:35.170 --> 01:13:42.460 align:middle line:84%
You subtract the same at
some reference conditions

01:13:42.460 --> 01:13:51.400 align:middle line:84%
of pressure and composition,
so not necessarily

01:13:51.400 --> 01:13:55.658 align:middle line:84%
the pure substance,
divided by RT.

01:13:55.658 --> 01:13:58.410 align:middle line:90%


01:13:58.410 --> 01:14:02.140 align:middle line:84%
And that means that
essentially you're

01:14:02.140 --> 01:14:07.290 align:middle line:84%
rewriting your chemical
potential as a term evaluated

01:14:07.290 --> 01:14:13.740 align:middle line:84%
at those reference conditions
plus RT log of the activity

01:14:13.740 --> 01:14:17.630 align:middle line:84%
so that you have an
equation that resembles

01:14:17.630 --> 01:14:22.190 align:middle line:84%
this one, except that
here you have the activity

01:14:22.190 --> 01:14:27.200 align:middle line:84%
and here you have taken a
reference chemical potential not

01:14:27.200 --> 01:14:34.040 align:middle line:84%
for the pure substance, but for
the mixture at some conditions

01:14:34.040 --> 01:14:37.770 align:middle line:84%
that you decide to be
useful as a reference.

01:14:37.770 --> 01:14:40.930 align:middle line:90%


01:14:40.930 --> 01:14:43.310 align:middle line:84%
And you can also define
the activity coefficient,

01:14:43.310 --> 01:14:47.310 align:middle line:84%
which is the ratio of the
activity to the mole fraction,

01:14:47.310 --> 01:14:55.230 align:middle line:84%
so that here you have
the logarithm of y,

01:14:55.230 --> 01:15:01.530 align:middle line:84%
which represents the ideal
part of the contribution.

01:15:01.530 --> 01:15:05.940 align:middle line:84%
And the non-ideal,
or the excess,

01:15:05.940 --> 01:15:13.440 align:middle line:84%
is all contained in these
activity coefficients.

01:15:13.440 --> 01:15:17.470 align:middle line:84%
So it's just changing
names, defining variables.

01:15:17.470 --> 01:15:22.280 align:middle line:84%
But the substance of the
matter is what we've done.

01:15:22.280 --> 01:15:25.390 align:middle line:90%


01:15:25.390 --> 01:15:35.410 align:middle line:84%
In the same way, people
sometimes use fugacities.

01:15:35.410 --> 01:15:40.065 align:middle line:84%
This is useful for
the gaseous phases.

01:15:40.065 --> 01:15:44.300 align:middle line:90%


01:15:44.300 --> 01:15:58.210 align:middle line:84%
So fugacities play the role of
generalized partial pressures.

01:15:58.210 --> 01:16:05.682 align:middle line:90%
But the idea is that here you--

01:16:05.682 --> 01:16:08.240 align:middle line:90%


01:16:08.240 --> 01:16:12.740 align:middle line:84%
so you define-- you
see these relations.

01:16:12.740 --> 01:16:15.920 align:middle line:84%
Let's start from
this relation here.

01:16:15.920 --> 01:16:20.220 align:middle line:84%
The chemical potential, you
write it, again, as usual,

01:16:20.220 --> 01:16:21.560 align:middle line:90%
like we did.

01:16:21.560 --> 01:16:26.300 align:middle line:90%


01:16:26.300 --> 01:16:30.390 align:middle line:84%
We take as reference
the pure substance.

01:16:30.390 --> 01:16:35.170 align:middle line:90%
Then you have RT log of y.

01:16:35.170 --> 01:16:40.940 align:middle line:84%
And then you have plus RT log
of p over p0 plus RT log of phi.

01:16:40.940 --> 01:16:48.980 align:middle line:84%
So essentially, you rename
the excess chemical potential

01:16:48.980 --> 01:16:55.090 align:middle line:84%
in terms of this
fugacity coefficient.

01:16:55.090 --> 01:17:03.040 align:middle line:84%
Or if you wanted to start
from the chemical potential,

01:17:03.040 --> 01:17:06.800 align:middle line:84%
you define the
fugacity by selecting,

01:17:06.800 --> 01:17:11.940 align:middle line:84%
again, a reference
condition for your mixture,

01:17:11.940 --> 01:17:17.050 align:middle line:84%
not necessarily like we do
here, the pure substance.

01:17:17.050 --> 01:17:23.000 align:middle line:90%


01:17:23.000 --> 01:17:33.800 align:middle line:84%
You select this coefficient,
this prefactor here as something

01:17:33.800 --> 01:17:37.490 align:middle line:84%
that in the limit of
ideal gas behavior

01:17:37.490 --> 01:17:41.920 align:middle line:90%
reduces to the partial pressure.

01:17:41.920 --> 01:17:46.160 align:middle line:84%
So the fugacity in those
limits is the partial pressure

01:17:46.160 --> 01:17:50.090 align:middle line:90%
times the exponential of this.

01:17:50.090 --> 01:17:56.740 align:middle line:84%
So again, this is just because
I wanted to mention it.

01:17:56.740 --> 01:17:59.350 align:middle line:84%
For example, if you
compute the activities

01:17:59.350 --> 01:18:03.320 align:middle line:84%
for the Margules, van Laar,
and Margules two parameters,

01:18:03.320 --> 01:18:06.820 align:middle line:90%
here are the expressions.

01:18:06.820 --> 01:18:13.750 align:middle line:84%
And here they are
for the van Laar.

01:18:13.750 --> 01:18:22.490 align:middle line:84%
I think this is for the van Laar
and how the activity depends

01:18:22.490 --> 01:18:27.620 align:middle line:84%
on mole fraction and on that
parameter that characterizes

01:18:27.620 --> 01:18:29.690 align:middle line:84%
the model-- or the
parameters A and B

01:18:29.690 --> 01:18:31.950 align:middle line:90%
that characterize the model.

01:18:31.950 --> 01:18:38.424 align:middle line:90%


01:18:38.424 --> 01:18:45.400 align:middle line:84%
Now, that means that when
we talk about liquid-vapor

01:18:45.400 --> 01:18:47.740 align:middle line:90%
equilibrium--

01:18:47.740 --> 01:18:52.600 align:middle line:84%
for example, let's suppose
we use the van Laar model.

01:18:52.600 --> 01:18:59.390 align:middle line:84%
And you remember that
we have to apply--

01:18:59.390 --> 01:19:01.280 align:middle line:84%
if we have two
substances that are

01:19:01.280 --> 01:19:06.260 align:middle line:84%
both volatile and
condensable, we

01:19:06.260 --> 01:19:09.230 align:middle line:84%
have to apply twice
the Raoult's law.

01:19:09.230 --> 01:19:17.670 align:middle line:84%
Now, Raoult's law rewrites
for a non-ideal mixture--

01:19:17.670 --> 01:19:21.180 align:middle line:84%
well, you substitute
the mole fraction

01:19:21.180 --> 01:19:23.370 align:middle line:84%
with the activity,
the mole fraction

01:19:23.370 --> 01:19:28.140 align:middle line:84%
for the components in the
liquid with their activities.

01:19:28.140 --> 01:19:30.390 align:middle line:84%
And we've given the
expressions for the activities

01:19:30.390 --> 01:19:33.150 align:middle line:84%
in the previous slide,
also for the van Laar.

01:19:33.150 --> 01:19:39.400 align:middle line:84%
So if you do that
substitution, here

01:19:39.400 --> 01:19:46.010 align:middle line:84%
is what the equation
of the bubble line is.

01:19:46.010 --> 01:19:48.285 align:middle line:90%
And this is for the dew line.

01:19:48.285 --> 01:19:57.380 align:middle line:90%


01:19:57.380 --> 01:20:01.250 align:middle line:84%
One of them used to
be a straight line.

01:20:01.250 --> 01:20:08.640 align:middle line:84%
But with this non-ideal model,
it becomes very different.

01:20:08.640 --> 01:20:12.690 align:middle line:90%


01:20:12.690 --> 01:20:16.602 align:middle line:84%
And the dew line also
becomes sort of interesting.

01:20:16.602 --> 01:20:21.860 align:middle line:90%


01:20:21.860 --> 01:20:25.490 align:middle line:84%
So just two parameters
introduce a lot

01:20:25.490 --> 01:20:30.680 align:middle line:84%
of structure in the
behavior of your mixture,

01:20:30.680 --> 01:20:33.240 align:middle line:84%
including the
possibility of this,

01:20:33.240 --> 01:20:36.716 align:middle line:90%
which is called the azeotrope.

01:20:36.716 --> 01:20:41.370 align:middle line:90%


01:20:41.370 --> 01:20:45.210 align:middle line:84%
The azeotrope is a
condition in which

01:20:45.210 --> 01:20:51.133 align:middle line:90%
you go from liquid to vapor.

01:20:51.133 --> 01:20:55.700 align:middle line:90%


01:20:55.700 --> 01:21:01.930 align:middle line:84%
And that change is not across a
finite temperature difference,

01:21:01.930 --> 01:21:04.090 align:middle line:90%
but at a single temperature.

01:21:04.090 --> 01:21:10.910 align:middle line:84%
So it behaves as if it
were a pure substance.

01:21:10.910 --> 01:21:13.910 align:middle line:84%
Like a pure substance
it boils and condenses

01:21:13.910 --> 01:21:17.680 align:middle line:90%
at the same unique temperature.

01:21:17.680 --> 01:21:21.010 align:middle line:84%
But this occurs only at
that particular composition.

01:21:21.010 --> 01:21:23.290 align:middle line:84%
At other compositions,
instead you

01:21:23.290 --> 01:21:24.920 align:middle line:90%
have a range of temperatures.

01:21:24.920 --> 01:21:32.440 align:middle line:84%
So the temperature
of the "liquidus"

01:21:32.440 --> 01:21:39.010 align:middle line:84%
is different than
that of the "vaporus."

01:21:39.010 --> 01:21:46.070 align:middle line:90%


01:21:46.070 --> 01:21:51.000 align:middle line:84%
This is done by
combining these two.

01:21:51.000 --> 01:21:55.260 align:middle line:84%
You can construct a plot
at a constant temperature.

01:21:55.260 --> 01:21:57.770 align:middle line:84%
But you can also,
the same equations,

01:21:57.770 --> 01:22:01.435 align:middle line:84%
construct the plot
at constant pressure.

01:22:01.435 --> 01:22:08.740 align:middle line:90%


01:22:08.740 --> 01:22:12.480 align:middle line:84%
And of course, you can eliminate
the pressure between the two

01:22:12.480 --> 01:22:15.950 align:middle line:84%
and construct, for
a given temperature

01:22:15.950 --> 01:22:18.740 align:middle line:84%
or for a given pressure,
the relation that you

01:22:18.740 --> 01:22:22.220 align:middle line:84%
have between the composition
of the vapor phase

01:22:22.220 --> 01:22:24.250 align:middle line:84%
and the composition
of the liquid phase.

01:22:24.250 --> 01:22:30.160 align:middle line:90%


01:22:30.160 --> 01:22:33.670 align:middle line:84%
So these are all things that you
can implement in an Excel file

01:22:33.670 --> 01:22:39.920 align:middle line:84%
and so on and very easy if
you have few parameters.

01:22:39.920 --> 01:22:47.950 align:middle line:90%


01:22:47.950 --> 01:22:55.610 align:middle line:84%
There is also a Lennard-Jones
model of the mixture,

01:22:55.610 --> 01:22:56.770 align:middle line:90%
of the binary mixture.

01:22:56.770 --> 01:23:00.590 align:middle line:90%


01:23:00.590 --> 01:23:09.100 align:middle line:84%
So here, this model
is taken again,

01:23:09.100 --> 01:23:11.940 align:middle line:84%
from this article on
Wikipedia from which I take

01:23:11.940 --> 01:23:16.710 align:middle line:90%
these plots, very nicely done.

01:23:16.710 --> 01:23:23.890 align:middle line:84%
It explains that you can
assume for the sigma--

01:23:23.890 --> 01:23:27.330 align:middle line:84%
you remember that
in the model of--

01:23:27.330 --> 01:23:30.570 align:middle line:84%
in the Lennard-Jones
model, sigma

01:23:30.570 --> 01:23:38.770 align:middle line:84%
is representation of the
diameter, essentially,

01:23:38.770 --> 01:23:41.500 align:middle line:90%
of the particle.

01:23:41.500 --> 01:23:46.120 align:middle line:84%
So when they collide, the two
centers of the two spheres

01:23:46.120 --> 01:23:50.190 align:middle line:84%
are at a distance, which
is equal to their diameter.

01:23:50.190 --> 01:23:53.540 align:middle line:90%


01:23:53.540 --> 01:23:58.840 align:middle line:84%
If you have spheres
of different sizes,

01:23:58.840 --> 01:24:07.450 align:middle line:84%
the diameter at which they
collide is the sum of the radii.

01:24:07.450 --> 01:24:10.250 align:middle line:84%
So the radii is half
of the diameter.

01:24:10.250 --> 01:24:16.180 align:middle line:84%
So the sum of the halves is
just the average diameter

01:24:16.180 --> 01:24:19.160 align:middle line:90%
of the two proper diameters.

01:24:19.160 --> 01:24:22.360 align:middle line:90%


01:24:22.360 --> 01:24:31.050 align:middle line:84%
So you assume a Lennard-Jones
curve or potential characterized

01:24:31.050 --> 01:24:36.220 align:middle line:84%
by a sigma12 for the
interaction between two

01:24:36.220 --> 01:24:43.380 align:middle line:84%
unlike particles, that is
equal to this mean value.

01:24:43.380 --> 01:24:52.090 align:middle line:84%
And then for the depth of
the Lennard-Jones valley,

01:24:52.090 --> 01:25:03.970 align:middle line:84%
you use the geometric mean
of the pure component ones.

01:25:03.970 --> 01:25:07.550 align:middle line:84%
But then you also use an
adjustable parameter, xi.

01:25:07.550 --> 01:25:15.090 align:middle line:90%


01:25:15.090 --> 01:25:22.620 align:middle line:84%
Now, the geometric--
if xi is equal to 1,

01:25:22.620 --> 01:25:28.980 align:middle line:84%
this geometric mean
gives you the equivalent

01:25:28.980 --> 01:25:33.930 align:middle line:90%
of the ideal solution model.

01:25:33.930 --> 01:25:40.080 align:middle line:84%
And in fact, you get that
the bubble line is straight.

01:25:40.080 --> 01:25:43.400 align:middle line:90%
It's a straight line.

01:25:43.400 --> 01:25:46.390 align:middle line:90%
And this is with xi equal to 1.

01:25:46.390 --> 01:25:52.210 align:middle line:90%


01:25:52.210 --> 01:25:59.650 align:middle line:84%
If you change instead
xi to greater than 1,

01:25:59.650 --> 01:26:05.290 align:middle line:84%
you get a non-ideal
mixture behavior

01:26:05.290 --> 01:26:09.780 align:middle line:84%
with an azeotrope
at this composition.

01:26:09.780 --> 01:26:16.810 align:middle line:84%
If you change xi from 1.2 to
0.85, it goes on the other side.

01:26:16.810 --> 01:26:19.380 align:middle line:90%
So it's an azeotrope--

01:26:19.380 --> 01:26:21.400 align:middle line:84%
upper azeotrope,
lower azeotrope.

01:26:21.400 --> 01:26:27.580 align:middle line:90%


01:26:27.580 --> 01:26:35.410 align:middle line:84%
And also, if you
play with the ratio--

01:26:35.410 --> 01:26:36.410 align:middle line:90%
this is wrong.

01:26:36.410 --> 01:26:40.480 align:middle line:84%
This should be an epsilon 2
over epsilon 1, of course.

01:26:40.480 --> 01:26:43.690 align:middle line:90%
I need to correct that.

01:26:43.690 --> 01:26:50.690 align:middle line:84%
So if you change that from
this one to very different,

01:26:50.690 --> 01:26:58.490 align:middle line:84%
you may find a critical
region, in other words that

01:26:58.490 --> 01:27:05.020 align:middle line:84%
could allow you
to go from liquid

01:27:05.020 --> 01:27:11.580 align:middle line:84%
to vapor without a phase
change, similarly to what

01:27:11.580 --> 01:27:14.955 align:middle line:90%
happens in a pure substance.

01:27:14.955 --> 01:27:20.710 align:middle line:84%
Where if you go above
the critical point,

01:27:20.710 --> 01:27:25.230 align:middle line:84%
you can take a liquid
to vapor and vice

01:27:25.230 --> 01:27:27.406 align:middle line:90%
versa (without a phase change).

01:27:27.406 --> 01:27:36.090 align:middle line:90%


01:27:36.090 --> 01:27:49.030 align:middle line:84%
So this case is one, let's
see, for example here,

01:27:49.030 --> 01:27:52.510 align:middle line:90%
in which you have a--

01:27:52.510 --> 01:27:55.230 align:middle line:84%
well, if you were
in liquid-liquid,

01:27:55.230 --> 01:27:57.410 align:middle line:84%
this would give you
partial miscibility.

01:27:57.410 --> 01:28:01.950 align:middle line:84%
But we'll do that
in another occasion.

01:28:01.950 --> 01:28:08.266 align:middle line:90%


01:28:08.266 --> 01:28:10.080 align:middle line:90%
So now, let's talk about--

01:28:10.080 --> 01:28:13.286 align:middle line:84%
a bit of the
stability conditions.

01:28:13.286 --> 01:28:23.730 align:middle line:90%


01:28:23.730 --> 01:28:27.750 align:middle line:90%
So as we have seen, we can go--

01:28:27.750 --> 01:28:30.810 align:middle line:84%
see, the business here is
to go from the properties

01:28:30.810 --> 01:28:33.690 align:middle line:84%
of a pure substance
to the properties

01:28:33.690 --> 01:28:36.870 align:middle line:84%
of the other substance,
and all the way to all

01:28:36.870 --> 01:28:40.410 align:middle line:90%
the compositions in between.

01:28:40.410 --> 01:28:48.730 align:middle line:84%
And when we do
these diagrams here,

01:28:48.730 --> 01:28:53.560 align:middle line:84%
everything is fixed
by the condition--

01:28:53.560 --> 01:28:55.190 align:middle line:90%
This is a binary composition.

01:28:55.190 --> 01:29:08.490 align:middle line:84%
So it's fixed by the
fact that mu1f must be--

01:29:08.490 --> 01:29:12.240 align:middle line:90%
at T p y.

01:29:12.240 --> 01:29:13.490 align:middle line:90%
Let's say y1f.

01:29:13.490 --> 01:29:17.100 align:middle line:90%


01:29:17.100 --> 01:29:21.510 align:middle line:84%
So for component
1 in the phase f,

01:29:21.510 --> 01:29:25.140 align:middle line:84%
I have to have the same chemical
potential of that component

01:29:25.140 --> 01:29:26.320 align:middle line:90%
in the phase g.

01:29:26.320 --> 01:29:32.606 align:middle line:90%


01:29:32.606 --> 01:29:38.210 align:middle line:84%
And that's one condition of
mutual stable equilibrium.

01:29:38.210 --> 01:29:42.575 align:middle line:90%
The other, mu2f--

01:29:42.575 --> 01:29:55.710 align:middle line:90%


01:29:55.710 --> 01:29:56.210 align:middle line:90%
Yeah.

01:29:56.210 --> 01:30:00.630 align:middle line:90%


01:30:00.630 --> 01:30:04.290 align:middle line:90%
Well, let's call it again 1f.

01:30:04.290 --> 01:30:10.410 align:middle line:84%
Let's choose for variables the
mole fraction of one of them.

01:30:10.410 --> 01:30:12.480 align:middle line:84%
But I need to
impose the condition

01:30:12.480 --> 01:30:16.140 align:middle line:84%
for equality of the
chemical potential also

01:30:16.140 --> 01:30:17.470 align:middle line:90%
of the other component.

01:30:17.470 --> 01:30:20.790 align:middle line:90%
So mu 2 must be equal to mu 2.

01:30:20.790 --> 01:30:26.650 align:middle line:84%
So I have a system of
two equations here.

01:30:26.650 --> 01:30:39.030 align:middle line:84%
And that system of two equations
can be solved, in general,

01:30:39.030 --> 01:30:53.680 align:middle line:84%
to give me this, to
give me-- for a given

01:30:53.680 --> 01:30:56.680 align:middle line:84%
temperature and pressure,
I can solve this system

01:30:56.680 --> 01:31:01.960 align:middle line:84%
for the unknowns that are the
mole fractions of component 1

01:31:01.960 --> 01:31:04.590 align:middle line:84%
in the gaseous and
in the vapor phase.

01:31:04.590 --> 01:31:15.500 align:middle line:90%


01:31:15.500 --> 01:31:18.140 align:middle line:84%
So here, I have pressure
and temperature.

01:31:18.140 --> 01:31:20.900 align:middle line:84%
I fixed the pressure
and the temperature.

01:31:20.900 --> 01:31:24.380 align:middle line:84%
And then that
system of equations

01:31:24.380 --> 01:31:27.620 align:middle line:84%
gives me the mole
fraction of the liquid

01:31:27.620 --> 01:31:29.160 align:middle line:84%
and the mole fraction
of the vapor.

01:31:29.160 --> 01:31:30.690 align:middle line:90%
So I have this surface.

01:31:30.690 --> 01:31:37.800 align:middle line:84%
So that equation here is an
equation of this surface here.

01:31:37.800 --> 01:31:41.250 align:middle line:84%
Actually, this is on the
liquid and on the vapor.

01:31:41.250 --> 01:31:42.420 align:middle line:90%
So it's on the other side.

01:31:42.420 --> 01:31:45.890 align:middle line:84%
But you can do it
for all the phases.

01:31:45.890 --> 01:31:49.730 align:middle line:84%
And you have these gaps
here, these empty spaces

01:31:49.730 --> 01:31:52.580 align:middle line:84%
between these surfaces,
which doesn't represent

01:31:52.580 --> 01:31:59.940 align:middle line:84%
any state, any stable state,
because if you are there,

01:31:59.940 --> 01:32:05.570 align:middle line:84%
the system likes to find stable
equilibrium by separating

01:32:05.570 --> 01:32:09.350 align:middle line:90%
and generating the two phases.

01:32:09.350 --> 01:32:11.665 align:middle line:90%
So that's the idea.

01:32:11.665 --> 01:32:14.330 align:middle line:90%


01:32:14.330 --> 01:32:19.130 align:middle line:84%
And yeah, I think we'll stop
here because there's not

01:32:19.130 --> 01:32:23.690 align:middle line:90%
enough time to do Henry's law.

01:32:23.690 --> 01:32:30.740 align:middle line:84%
But we can-- yeah, so next time,
we will finish up this part.

01:32:30.740 --> 01:32:35.260 align:middle line:84%
And we'll start the systems
with chemical reactions.

01:32:35.260 --> 01:32:44.000 align:middle line:90%