ASME PRESSURE VESSEL CODE

MARCH 29, 2016

MT-II LAB (CHE-316B)

LIQUID-LIQUID EXTRACTOR DESIGN

• During the pilot plant tests, the extractor separation
performance is measured over a range of flow rates,
solvent to feed ratios and effective heights. If the
extractor selected is agitated then the agitation speed
is also varied. At each flow rate, the agitation is
increased until the flood point is reached. From this
data, the conditions that give the optimum volumetric
efficiency are determined.

• Any scale up procedure to commercial size must

determine
(1) how diameter varies with throughput and
(2) how stage efficiency varies with diameter. For the agitated
columns, there is an additional parameter to be determined
(3) how the power input varies with column size.
 Types of columns
• Agitated Columns
• KARR COLUMN
• SCHEIBEL COLUMN
• ROTATING DISC COLUMN (RDC)
• PULSED COLUMN

• Static Columns
• SIEVE TRAYS
• RANDOM PACKING
• STRUCTURED PACKING
 KARR Columns
Characteristics:
• Highest capacity: 30-60
m3/m2.h
• Good efficiency
• Good turndown capability
(25%)
• Uniform shear mixing
• Best suited for systems that
emulsify
 SCHEIBEL Columns
Characteristics
• Reasonable capacity: 15-25
m3/m2.h
• High efficiency due to
internal baffling
• Good turndown capability
(25%)
• Best suited when many
stages are required
• Not recommended for highly
fouling systems or systems
that tend to emulsify
 Rotating Disc Contactor (RDC)
Columns
Characteristics
• Reasonable capacity: 20-30
m3/m2.h
• Limited efficiency due to
axial backmixing
• Suitable for viscous materials
• Suitable for fouling materials
• Sensitive to emulsions due
to high shear mixing
• Reasonable turndown (40%)
 Pulsed Column
Characteristics
• Reasonable capacity: 20-30
m3/m2.h
• Best suited for nuclear
applications due to lack of seal
• Suited for corrosive
applications when constructed
out of non-metals
• Limited stages due to
backmixing
• Limited diameter/height due to
pulse energy required
 Packed – Random
• High capacity: 20-30 m3/m2.h
• Poor efficiency due to backmixing and wetting
• Limited turndown flexibility
• Affected by changes in wetting characteristics
• Limited as to which phase can be dispersed
 Packed – Structured
• High capacity: 40-80 m3/m2.h
• Poor efficiency due to back
mixing and wetting
• Limited turndown flexibility
• Affected by changes in wetting
characteristics
• Limited as to which phase can
be dispersed
 Sieve Tray
Characteristics
• High capacity: 30-50 m3/m2.h
• Good efficiency due to minimum back
mixing
• Multiple interfaces can be a problem
• Limited turndown flexibility
• Affected by changes in wetting
characteristics
• Limited as to which phase can be
dispersed
 Dispersed-phase droplet size
• In packed columns, it is generally desirable to disperse
the inlet stream with the higher volumetric flowrate
into the continuous phase in order to maximize the
interfacial area for mass transfer.
• In agitated columns, the liquid phase with the lower
volumetric flowrate is typically dispersed, so that both
phases (dispersed and continuous) have similar
residence times.
• It is generally desirable to disperse the liquid
containing the solute in the extracting solvent, so that
the direction of mass transfer is out of the dispersed
droplets and into the continuous phase.
• The Marangoni effect explains how mass transfer
of solute from one phase to the other creates
interfacial tension gradients on the surface of the
dispersed phase. Transfer of solute from the
droplet outward into the continuous phase tends
to create larger droplets and higher interfacial
area of contact between the two phases, and is
represented by a larger proportionality constant.
Transfer of solute from the continuous phase into
the droplets tends to form smaller droplets and
the proportionality constant is smaller.
Liquid-liquid extraction is a useful method to separate components
(compounds) of a mixture
Let's see an example.

Suppose that you have a mixture of sugar in vegetable oil (it tastes
sweet!) and you want to separate the sugar from the oil. 

You
observe that the sugar particles are too tiny to filter and you
suspect that the sugar is partially dissolved in the vegetable oil.

What will you do?

 How about shaking the mixture
with water
Will it separate the sugar from the
oil? Sugar is much more soluble in
water than in vegetable oil, and,
as you know, water is immiscible
(=not soluble) with oil.

Did you see the result?The water

phase is the bottom layer andthe oil
phase is the top layer, because
water is denser than oil.

*You have not shaken the mixture

yet, so sugar is still in the oil phase.
By shaking the layers (phases) well, you
increase the contact area between the
two phases.The sugar will move to the
phase in which it is most soluble: the
water layer

Now the water phase tastes

sweet,
because the sugar is moved
to the water phase upon
shaking.**You extracted sugar from
the oil with water.**In this
example,water was the extraction
solvent ;the original oil-sugar mixture
was the solution to be extracted; and
sugar was the compound extracted
from one phase to another.
Separating the two layers
accomplishes the separation of the
sugar from the vegetable oil
Did you get it? .....the concept of liquid-liquid extraction?

Liquid-liquid extraction is based on the transfer of a solute

substance from one liquid phase into another liquid phase according
to the solubility.Extraction becomes a very useful tool if you choose
a suitable extraction solvent.
You can use extraction to separate a
substance selectively from a mixture, or to remove unwanted
impurities from a solution.In the practical use, usually one phase is a
water or water-based (aqueous) solution and the other an organic
solvent which is immiscible with water.

The success of this method depends upon the difference in solubility

of a compound in various solvents. For a given compound, solubility
differences between solvents is quantified as the "distribution
coefficient"
Partition Coefficient Kp (Distribution Coefficient Kd)

When a compound is shaken in a separatory funnel with two immiscible

solvents, the compound will distribute itself between the two solvents.

Normally one solvent is water

and the other solvent is a
water-immiscible organic
solvent.

Most organic compounds are

more soluble in organic solvents,
while some organic compounds
are more soluble in water.
Here is the universal rule:

At a certain temperature, the ratio of concentrations of a solute

in each solvent is always constant.ﾊAnd this ratio is called the
distribution coefficient, K.

(when solvent1 and solvent2 are immiscible liquids

For example,Suppose the

compound has a distribution
coefficient K = 2 between
solvent1 and solvent2

By convention the organic

solvent is (1) and waater is
(2)
 (1) If there are 30 particles
of compound , these are
distributed between equal
volumes of solvent1 and solvent2..

(2) If there are 300

particles of compound , the
same distribution ratio is
observed in solvents 1 and 2

(3) When you double the

volume of solvent2 (i.e., 200
mL of solvent2 and 100 mL of
solvent1),
the 300 particles
of compound distribute as
shown
If you use a larger amount of extraction solvent, more solute is
extracted
What happens if you extract twice with 100 mL of solvent2 ?
In this case, the amount of extraction solvent is the same volume as was
used in Figure 3, but the total volume is divided into two portions and
you extract with each.

As seen previously, with 200 mL

of solvent2 you extracted 240
particles of compound . One
extraction with 200 mL gave a
TOTAL of 240 particles
You still have 100 mL of solvent1,
containing 100 particles. Now you
add a second 100 mL volume of
fresh solvent2. According to the
distribution coefficient K=2, you
can extract 67 more particles
from the remaining solution
An additional 67 particles are
extracted with the second portion
of extraction solvent
(solvent2).The total number of
particles extracted from the first
(200 particles) and second (67
particles) volumes of extraction
solvent is 267.This is a greater
number of particles than the
single extraction (240 particles)
using one 200 mL portion of
solvent2!

It is more efficient to carry out

two extractions with 1/2 volume
of extraction solvent than one
large volume!
If you extract twice with 1/2 the volume, the extraction is more
efficient than if you extract once with a full volume. Likewise,
extraction three times with 1/3 the volume is even more efficient….
four times with 1/4 the volume is more efficient….five times with 1/5
the volume is more efficient…ad infinitum

The greater the number of small extractions, the greater the

quantity of solute removed. However for maximum efficiency the
rule of thumb is to extract three times with 1/3 volume
 Liquid-Liquid Extraction
CHOICE OF SOLVENT

Choosing the best solvent is the most critical aspect of developing a liquid-liquid
extraction process. The solvent should have a high selectivity for the extracted solute.
The selectivity  of a solvent is similar to relative volatility and is given by

x1S / x2S
 R R
x1 / x2

x1S = weight fraction of component 1 (solute) in the solvent phase

x2S = weight fraction of component 2 in the solvent phase
x1R = weight fraction of component 1 (solute) in the raffinate phase
x2R = weight fraction of component 2 in the raffinate phase
 Liquid-Liquid Extraction
SINGLE-STAGE CALCULATIONS
In one-stage liquid-liquid extractor as shown in the next Figure, Component 1 is the solute,
component 2 is the other component in the feed that we are trying to separate from
component 1, and component 3 is the solvent.

Feed Raffinate phase

F, z1, z2 , z3 R1, x1R1, x2R1 , x3R1
Stage

Fresh solvent solvent phase

So, x1So, x2So , x3So S1, x1S1, x2S1 , x3S1

In the normal situation, the process feed contains no solvent, so that z3 = 0. We need to
specify only two weight fractions in this ternary system.

If the solvent is essentially pure, as is often the case, x3So = 1, x1So = 0, x2So = 0
 Liquid-Liquid Extraction

The total mass balance at this stage is

(4)

A component balance on the jth component yields

(5)

Now we define the parameter

(6)

Where the point M must lie on the straight line joining F and So. It also must lie on a
straight line joining S1 and R1
 Liquid-Liquid Extraction
(7)
(8)

(9)

Since the two liquid phases leaving the system are in phase
equilibrium, the points R1 and S1 must be connected
by an LLE tie-line.

the procedure for determining the compositions and

flow rates of the two liquid streams leaving the
system is as follows:
1. Calculate M from equation (7).
2. Calculate x1M and x3M from equations (8) and (9).
3. Locate the point M using x1M and x3M .
4. Find the LLE tie-line that passes through the point M.
 Liquid-Liquid Extraction
5. The points at the two ends of the LLE tie-line give the compositions of the two phases
leaving the system: the raffinate-rich phase with composition x1R1 and x3R1 and the
solvent-rich phase with composition x1S1 and x3S1
6. Calculate the flow rates S1 and R1 by solving equations (4) and (5) simultaneously:

Splitting the mixture of F and So into raffinate-rich phase R1 and

solvent-rich phase S1 at ends of a tie-line.
 Liquid-Liquid Extraction
Example: An organic stream, with composition 30 weight percent acetone and 70 weight
percent methyl isobutyl ketone and flow rate 10,000 kg/h, is mixed with a pure water
solvent with flow rate 5,000 kg/h. What are the compositions and flow rates of the two
liquid phases leaving a single-stage liquid-liquid extractor operating at 25oC.

Solution:
Raffinate phase
z1 = 0.3 x1 S o = 0 Feed= 10000 R 1 , x 1 R1 , x 2 R1 , x 3 R1
z1=0.3 z2= 0.7, z3=0
z2 =0.7 x2 S o = 0 Stage
z3 = 0 x3So= 1.0 Fresh solvent=5000 solvent phase
x1So,=0x2So =0, x3So=1.0 S1, x1S1, x2S1 , x3S1
 Liquid-Liquid Extraction

Fz1  S 0 x1S0
M
x
1 
M

Fz3  S 0 x3S0
x M
3 
M
 Liquid-Liquid Extraction

The point M is thus located at (0.333,0.20), as shown in Figure

Note that M lies on the straight line connecting the points F and So
 Liquid-Liquid Extraction

The LLE conjugate line is used to determine the other end of the LLE tie-line on the solvent
phase solubility curve. If the tie-line goes through the point M, the compositions of
the water solvent phase and organic raffinate phase have been found. We have:
 Liquid-Liquid Extraction
MULTIPLE STAGES WITH CROSSFLOW OF SOLVENT

If the process liquid stream from the first stages fed into a second extractor and
mixed with more fresh solvent, as shown in the Figure, we have what is called
cross-flow extraction.

The process can be described the same as for a single stage .it is simply repeated
again for each stage, using the raffinate phase from the upstream stage as the
feed to each stage.
Liquid-Liquid Extraction

Mix points for cross-flow extractor.

 Liquid-Liquid Extraction
MULTISTAGE COUNTERCURRENT EXTRACTION
Multistage countercurrent extraction is the most commonly encountered liquid-liquid
extraction process. The raffinate and solvent streams travel countercurrent to each other
through N stages. The flow rate of the raffinate leaving the last stage (tray N) is R.

Mass and component balances around the entire cascade give

F  S 0  RN  S1 (10)
z1 F  x1S0 S 0  x1RN R N  x1S1 S1 (11)
z 3 F  x3S0 S 0  x3RN R N  x3S1 S1 (12)
 Liquid-Liquid Extraction

We next define a pseudo flow rate M and pseudo compositions x1M and x3M :

M  F  S0 (13)

x1M M  z1 F  x1S0 S 0 (14)

x3M M  z 3 F  x3S0 S 0 (15)

 Liquid-Liquid Extraction
If the fresh solvent flow rate So and
composition are given along with the
feed flow rate F and composition, we
can locate the point M on the
straight line connecting the points F
and So.
From equations (10) through (12), it is
clear that M must also lie on the
straight line connecting S1 and RN, as
shown in the figure.
 Liquid-Liquid Extraction
In the typical design problem, you will be given:

The point RN will be given, i.e., the concentration of the raffinate phase leaving the
final stage will be specified so as to recover the desired amount of the solute from the
feed.

Note: in any real system we cannot recover all of the solute from the feed

Since the points RN and M are known, a straight line can be drawn to the solubility
curve to determine the composition of the x1S1 and x3S1 of the S1 stream.
Equations (10) and (11) can be used to solve for the flow rates RN and S1.
 Liquid-Liquid Extraction

Since S1 and R1 are in phase equilibrium, we

can use an LLE tie-line to determine the
point R1 on the solubility curve.
 Liquid-Liquid Extraction
Component and mass balances around the first stage can then be used to calculate the flow
rate S2 and compositions x1S2 and x3S2 of the solvent entering stage 1 from stage 2:

F  S 2  R1  S1 (16)
z1 F  x1S2 S 2  x1R1 R1  x1S1 S1 (17)
z 3 F  x3S 2 S 2  x3R1 R1  x3S1 S1 (18)
Once S2 is known, R2 can be found at the other end of the LLE tie-line.
This computational procedure can be repeated from stage to stage until enough stages
have been used to produce a process stream that meets or exceeds the specifications
on RN (the final raffinate phase leaving the unit).
 Liquid-Liquid Extraction
Graphical procedure

First, define another pseudo flow rate Δ and pseudo compositions x1Δ and x3Δ This
fictitious Δ stream serves the same purpose as the operating lines did in binary
distillation. We can calculate the compositions of "passing" streams in the column. It
provides a graphical way to solve the three mass balances that describe this ternary
system simultaneously:

  RN  S 0 (19)

x1   x1RN R N  x1S0 S 0 (20)

x3   x3RN R N  x3S0 S 0 (21)

 Liquid-Liquid Extraction

The pseudo composition xjΔ is entirely fictitious and, therefore, can be less than zero or
greater than unity.

Using equations (10) and (19), we see that

  RN  S 0  F  S1 (22)

Therefore, the Δ point must lie on two straight lines, one through the points RN and S0
and the other through the points F and S1.
 Liquid-Liquid Extraction

Δ can lie either to the left or to the right

of the phase diagram
 Liquid-Liquid Equilibrium

The mass balances for the first stage (equations (16) through (18), the definition of Δ
(equations (19) through (21), and equation(22) can be combined to give

  RN  S 0  F  S1  R1  S 2 (23)

The two streams R1 and S2 that pass each other between the first and second stages are
related to each other by the Δ point. Hence, if we know R1 we can determine S2 by
using the straight line that connects R1 and Δ.
Liquid-Liquid Extraction
 Liquid-Liquid Extraction
Example: A liquid-liquid ternary phase diagram for isopropyl alcohol (IPA), toluene, and water
at 25oC. Feed flow rate is 100 kg/hr, and feed compositions 40 weight percent IPA and 60
weight percent toluene. Fresh solvent is pure water at a flow rate of 100 kghr. Determine
the number of equilibrium stages required to produce a raffinate stream that contains 3
weight percent IPA.
End of the Lecture

Thank You