Extraction

“When separation by distillation is ineffective or very difficult e.g. close-boiling

mixture, liquid extraction is one of the main alternative to consider.”

What is Liquid-liquid extraction (or solvent extraction)?

Liquid-Liquid extraction is a mass transfer operation in which a liquid solution
(feed) is contacted with an immiscible or nearly immiscible liquid (solvent)
that exhibits preferential affinity or selectivity towards one or more of the
components in the feed. Two streams result from this contact:
a) Extract is the solvent rich solution containing the
desired extracted solute.
b) Raffinate is the residual feed solution containing little solute.
Liquid-liquid extraction principle

When Liquid-liquid extraction is carried out in a test tube or flask the two
immiscible phases are shaken together to allow molecules to partition
(dissolve) into the preferred solvent phase.
An example of extraction:

Extract
Acetic acid in H2O Organic layer contains most of acetic acid in ethyl
acetate with a small amount of water.
+
Raffinate
Ethyl acetate Aqueous layer contains a weak acetic acid
solution with a small amount of ethyl acetate.

The amount of water in the extract and ethyl acetate in the raffinate depends
upon their solubilites in one another.
Typical Applications

• Remove products and pollutants from dilute aqueous streams

• Wash polar compounds or acids/bases from organic streams
• Heat sensitive products
• Non-volatile materials
• Azeotropic and close boiling mixtures
• Alternative to high cost distillations
Extraction is Used in a Wide Variety of Industries
•Washing of acids/bases, polar compounds from
Chemical organics
• Recovery of active materials from fermentation
Pharmaceuticals broths
• Purification of vitamin products
• Recovery of phenol, DMF, DMAC
Effluent Treatment • Recovery of acetic acid from dilute solutions
• Recovery of caprolactam for nylon manufacture
Polymer Processing • Separation of catalyst from reaction products
• Lube oil quality improvement
Petroleum • Separation of aromatics/aliphatics (BTX)
• Separation of olefins/parafins
Petrochemicals • Separation of structural isomers
• Decaffeination of coffee and tea
Food Industry • Separation of essential oils (flavors and
fragrances)
• Copper production
Metals Industry • Recovery of rare earth elements

Inorganic Chemicals • Purification of phosphoric acid

Nuclear Industry • Purification of uranium

Removal of Organics From Water
Distillation vs. Extraction
Water Azeotrope Azeotrope Typical
Organic Compound BP [°C]
Solu. [%] B.P. [°C] Water [%] Reduction Level
Methylene Chloride 40 2.0 38.1 1.5 < 50 ppb
Distillation

Acetone 56.2 Infinite Non Azeotropic < 50 ppb

Methanol 64.5 Infinite Non Azeotropic < 50 ppb
Benzene 80.1 0.18 69.4 8.9 < 50 ppb
Toluene 110.8 0.05 85.0 20.2 < 50 ppb
Formaldehyde -21 Infinite Non Azeotropic < 1,000 ppm
Formic Acid 100.8 Infinite 107.1 22.5 < 500 ppm
Acetic Acid 118.0 Infinite Non Azeotropic < 500 ppm
Pyridine 115.5 57 92.6 43 < 10 ppm
Extraction

Aniline 181.4 3.60 99.0 80.8 < 10 ppm

Phenol 181.4 8.20 99.5 90.8 < 10 ppm
Nitrobenzene 210.9 0.04 98.6 88.0 < 10 ppm
Dinitrotoluene
300.0 0.03 99 – 100 > 90 < 10 ppm
(2,4)
Dimethyl
153.0 Infinite Non Azeotropic < 10 ppm
Formamide
Dimethyl
166.1 Infinite Non Azeotropic < 10 ppm
Acetamide
n-
202.0 Infinite Non Azeotropic < 10 ppm
Methylpyrrolidone
Once the solvent has been chosen, the extraction process has to be defined and
designed. Four extraction techniques will be studied in this course:

Single
stage

Cross- Counter-
current current
 Conditions of a choice of solvent
1. Should not mix up with feed solvent.
2. Should be selective.
3. Should have the big capacity in relation to
extractive.
4. The density difference should be high.
5. Should have the minimum viscosity.
6. Should be inexpensive.
7. Cannot be explosive.
Bench Scale Test Apparatus

Variable Speed Drive

Baffle
Thermometer

Tempered
Water Out

1 – Liter Flask

Tempered
Water In
Drain
 Extraction equipment
Batch: Continuous:

single-stage: column:

separatory funnel

mixer-settler
rotating-disk contacter
a. agitator; b. stator disk
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 / x 2S
 R R
x1 / x 2

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
Simple Extraction Single Stage
Extract (E)
A–0
B – 50
Feed (F) C – 0.8
50.8
A – 99
B–0
C–1
100 A–0 A – 99.0
B – 50
Raffinate (R) B–0
C–0 C – 0.2
50 99.2

Solvent (S)
Solute in Raffinate 0.2
Fraction Unextracted U  0.2
Solute in Feed 1.0
0.8
Conc.Solute in Extract
Distribution Coefficient M  50 7.92
Conc.Solute in Raffinate 0.2
99
Extraction Factor    
E  S M   50 7.92 4.0
F 99
SINGLE-STAGE CALCULATIONS
Raffinate phase
Feed R, XR
F, XF
Stage
solvent phase or
Fresh solvent Extract
S Ys E, Y1
mass C kg C
X   in the feed solution Material balance of solute C
F mass A kg A

X 
mass C
in the raffinate X F A  BYS  X 1 A  Y1 B
1 mass A
mass C  A( X 1  X F ) B (Y1  Ys )
Y  in the extract
1 mass B
Y1  Ys A
Y 
mass C
in the extraction solvent

s mass B X1  X F B

B is Solvent (Fresh Solvent adding)

C is Solute
A is Solvent present in the solution
 Y1  Ys A

X1  X F B

e
lin
.
Eq
(X1,Y1) •

slo
pe
=-
A/
B
•(XF,YS)

B is Solvent (Fresh Solvent adding)

C is Solute
A is Solvent present in the solution
 Cross Flow Extraction
E1 E2 E3 E4
Y1
Y2 Y3 Y14
A+B
1 2 3 4
F, XF A

C C C
S1 S2 S3 S4
B1 , Ys 0 B2 , Ys 0 B3 , Ys 0
B4 , Ys 0
Material balance of solute C
2nd Stage
1st Stage
X F A  B1Ys1  X 1 A  Y1 B1 Y2  Ys 2 A
 A( X 1  X F ) B1 (Y1  Ys1 ) 
X 2  X1 B2
Y1  Ys1 A

X1  X F B1
 e
lin
• Increase overall efficiency by introducing

.
Eq
fresh extracting solvent at each stage.
• Each stage has its own mass balance and
(X1,Y1) •
operating line

slo
• Uses much more solvent than counter-

pe
(X2,Y2)
current cascade (requires much more •

=-
-A

A/
solvent recovery) (X3,Y3)•

/B 2

B1
-A
/B 3
• A mixer-settler is just one cross-flow stage.
•(x2,y3,in) •(x1,y2,in) •(x0,y1,in)
X F A  B1Ys1  X 1 A  Y1 B1 •
x3
 A( X 1  X F ) B1 (Y1  Ys1 )
Y1  Ys1 A Y2  Ys 2 A
X1  X F

B1

X 2  X1 B2
 Countercurrent Flow Extraction
B Ys
E1 En+1 Yn+1
C
Y1 Y3 E3
X 1 R1 X 2 R2 X 3 R3 X n Rn
1 2 3 N
F, XF A

E2 Y2 En Yn

X F A  B Y2  X 1 A  Y1 B
Material balance of the solute C over the entire unit
X F A  B Yn 1  X n A  Y1 B
A( X F  X n ) B1 (Y1  Yn 1 )
Y1  Yn 1 A

XF  Xn B
A
Yn 1  X F  X n   Y1
B
 e
lin
.
Eq
•X0

)
mi
n
/B
(A
1• •(XF,Y1)

2• • e
lin ctual
n g /B a
rati = A
e e
3• • op lop
s

•(XN,YN+1)

Yn+1= Inlet solvent X F Feed

X n Final Raffinate
 Reading ternary phase diagrams

Consider the point M: •

water content (xA) is ? 0.19
ethylene glycol content (xB) is ? 0.20
furfural content (xC) is ? 0.61
check: xA + xB + xC = 1
Read the mole/mass fraction of each
component on the axis for that component,
using the lines parallel to the edge opposite the
corner corresponding to the pure component.

The mixture M lies inside the miscibility

boundary, and will spontaneously separate • •
into two phases. Their compositions (E and
R) are given by the tie-line through M. region of partial miscibility A-C
The compositions of E and R converge at the plait point, P (i.e., no separation).
A 2-component mixture of furfural and water is partially miscible over the composition
range from about 8 % furfural to 95 % furfural. Separation by extraction requires a
furfural/water ratio in this range (otherwise – single phase).
Single-stage liquid-liquid extraction processes
Triangular coordinates and equilibrium data

Each of the three corners

represents a pure component A, B,
or C.
Point M represents a mixture of A,
B, and C.
The perpendicular distance from
the point M to the base AB
represents the mass fraction xC. The
distance to the base CB represents
xA, and the distance to base AC
represents xB.

xA + xB + xC = 0.4 + 0.2 + 0.4 = 1

xB = 1.0 - xA - xC

Equilateral triangular diagram yB = 1.0 - yA - yC

(A and B are partially miscible.) 22

Wt fraction of C

Wt fraction of B
Major Types of Extraction Equipment

Mixer Column
Settlers Contactors Centrifugal

Used primarily in the metals Used primarily in the

industry due to: pharmaceutical industry due to:
- Large flows Static Agitated - Large flows
- Intense mixing - Intense mixing
- Long Residence time - Long Residence time
- Corrosive fluids - Corrosive fluids
- History - History

Spray Packed Tray Pulsed Rotary

Reciprocating

Rarely used Used in: Used in: Used in: Used in:
- Refining - Refining - Nuclear - Chemicals
- Petrochemicals - Petrochemicals - Inorganics - Petrochemicals
- Chemicals - Refining
- Pharmaceutical

Example: Example: Example: Example: Example:

- Random - Sieve - Packed - RDC - Karr
- Structured - Tray - Scheibel
- SMVPTM - Disc & Donut
Mix / Decant Tank

Characteristics
Feed Inlet
• Mix – Settle – Phase separate in a single tank
• Batch Processing only
• Requires multiple solvent additions for more
than one stage (crossflow operation)
• Typically used for small capacity operations
or intermittent processing

Sight Glass

Outlet
Mixer / Settlers

Characteristics
• Handle very high flowrates
• Good for processes with
relatively slow reactions
(residence time required)
• Provide intense mixing to
promote mass transfer
• Require large amount of floor
space
• Suitable when few theoretical
stages required
• Large solvent inventory (and
losses)

Mixer-settlers operate with a purely stage-

wise contact. After every mixer there is a
settler. Mixer-settlers can be operated in a
multistage, co- or countercurrent fashion.
Mixer-Settles for Extraction

Separate mixer-settler Combined mixer-settler

39
Centrifugal Extractor

Characteristics
• Countercurrent flow via centrifugal
force
• Low residence time ideally suited for
some pharmaceutical applications
• Handles low density difference
between phases
• Provide up to several theoretical stages
per unit
• High speed device requires
maintenance
• Susceptible to fouling and plugging due
to small clearances
Countercurrent Extraction

B+C Extract (E):

Solute Rich Stream

A+C Primary Interface

Feed (F)

Continuous Phase

Dispersed Phase

B
Solvent (S)

Raffinate (R):
A +C Solute Lean Stream
Packed Column

Characteristics
Extract (E)
• High capacity:
Feed (F) 20-30 M3/M2-hr (Random)
500-750 gal/ft2-hr (Random)
40-80 M3/M2-hr (Structured)
1,000-2,000 gal/ft2-hr (Structured)
• Poor efficiency due to backmixing and
wetting
• Limited turndown flexibility
• Affected by changes in wetting
characteristics
• Limited as to which phase can be dispersed
Solvent (S)
• Requires low interfacial tension for
economic usefulness
Raffinate (R)
• Not good for fouling service
Sieve Tray Column

Characteristics
Extract (E)
Feed (F) • High capacity: 30-50 M3/M2-hr
750-1,250 gal/ft2-hr
Primary
• Good efficiency due to minimum
Interface
backmixing
• Multiple interfaces can be a problem
• Limited turndown flexibility
• Affected by changes in wetting
characteristics
• Limited as to which phase can be dispersed

Solvent (S)

Raffinate (R)
RDC Extractor

Characteristics Drive Motor Gearbox

• Reasonable capacity:
20-30 M3/M2-hr
Light
• Limited efficiency due to axial Phase Out
Heavy
backmixing Phase In
• Suitable for viscous materials
Vessel Shaft
• Suitable for fouling materials Walls
• Sensitive to emulsions due to
high shear mixing
• Reasonable turndown (40%)

Stators Rotors
Light
Phase In

Interface Interface
Control

Heavy
Phase Out
Reciprocating Column

Drive
Characteristics Assembly
• Highest capacity: Seal Light
Phase Out
30-60 M3/M2-hr
750-1,500 gal/ft2-hr
• Good efficiency Heavy
Phase Inlet Spider Plate

• Good turndown capability (4:1) Sparger

Center Shaft
• Uniform shear mixing & Spacers
Metal Baffle
• Best suited for systems that emulsify
Plate Tie Rods
Perforated & Spacers
Plate
Teflon
Light
Baffle Plate
Phase Inlet

Sparger

Interface Interface
Control

Heavy
Phase Out
Pulsed Extractor

Characteristics
Compressed
• Reasonable capacity: Light
Air
Phase Out
20-30 M3/M2-hr Heavy
Phase In
• Best suited for nuclear
applications due to lack of seal Timer

• Also suited for corrosive Solenoid

Valves
applications since can be
constructed out or non-metals Air
• Limited stages due to backmixing
• Limited diameter/height due Exhaust
Liquid
to pulse energy required Light
Pulse
Phase In
Leg
Interface Interface
Control

Heavy
Phase Out
Comparison Plot of Various Commercial Extractors

20
Scheibel
10
RZE
Kuhni Key
6
Efficiency / Stages per Meter

Graesser Karr Graesser = Raining Bucket

4 PFK MS = Mixer Settler
RDC PSE SE = Sieve Plate
FK = Random Packed
2 FK PFK = Pulsed Packed
MS PSE = Pulsed Sieve Plate
SE
1 RDC = Rotating Disc Contactor
RZE = Agitated Cell
.06 Karr = Karr Recipr. Plate
0.4 Kuhni = Kuhni Column
Scheibel = Scheibel Column

0.2
1 2 4 6 10 20 40 60 100
Capacity M3/(M2 HR)
Column Selection Criteria Static Column

A static column design may be appropriate when:

• Interfacial tension is low to medium: up to 10-
15 dynes/cm
• Only a few theoretical stages are required, and
reduction in S/F is not an economic benefit
• No operational flexibility required
• There is a large difference in solvent to feed
rates
Column Selection Criteria
Agitated Column
Agitated columns are generally more economical when:
• More than 2-3 theoretical stages are required
• Interfacial tension is moderate to high, although
low interfacial tensions may also be economical
• A reduction in solvent usage is beneficial to the
process economics
• The process requires a wide turndown as well as
the ability to handle a range of S/F ratios
Column Selection Criteria
Rotating Disc Contactor (RDC)
• Systems with moderate to high viscosity, i.e. >
100 cps
• Systems that are residence time controlled, for
example, slow mass transfer rate with few
theoretical stages required
• Systems with a high tendency towards fouling
Column Selection Criteria
Karr Reciprocation Plate Column
• Difficult systems that tend to emulsify and/or
flood easily
• Systems in which the hydraulic behavior varies
significantly through length of the column
• Sometimes requiring non-metallic internals,
such as Teflon due to wetting characteristics or
corrosive materials
• Fouling applications that may have tars
formations and/or solids precipitation
The Three Cornerstones of Successful
Extraction Applications

Successful
Application

Proper Solvent Meaningful Accurate

Selection Pilot Tests Scale-Up
Selection Based on: Testing Based on: Scale-Up Based on:

• Sound thermodynamic • Actual feed stocks • Proven techniques

principles
• Full process including • Proper safety factors
• Sound economic solvent recovery
principles
• Availability • Wide range of operating
• Recoverability conditions

• Sound environmental
principles
• Toxicity
• Safety
Liquid-Liquid Extraction Scale-Up

• Theoretical scale-up is difficult

• Complexity of processes taking place within an extractor
 Droplet Breakup
 Coalescence
 Mass Transfer
 Axial and radial mixing
 Effects of impurities
• Best method of design:
Pilot testing followed by empirical scale-up
Typical Extraction System

Feed B+C+(A)
A+B

Stripping
Raffinate

Recovery
Solvent
Extraction

C C
Solvent (A) (A+B)
C
(A+B)

A+(B+C)

A (B+C) B (C)
Possible Extraction Column Configurations
Solvent is Light Phase
B+C E B+C E
F F
A+B Primary A+B
Interface
Solvent Solvent
Primary
Dispersed Continuous Interface

S S
C C
R R
A A
Solvent is Heavy Phase A A
R R
S S
C C
Primary
Solvent Solvent Interface
Dispersed Primary Continuous
Interface
F F
A+B E A+B E
B+C B+C
Factors Effecting which Phase is Dispersed
Flow Rate
• For Sieve Tray and Packed Columns – disperse the higher flowing phase
• For all other columns – disperse lower flowing phase
Viscosity
• For efficiency – disperse less viscous phase

Viscous drop

Diffusion rate inside the drop is

inhibited by viscosity
• For capacity – disperse more viscous phase

Viscous continuous phase

Drop rise or fall will

be inhibited
Factors Effecting which Phase is Dispersed
Surface Wetting
• Want the continuous phase to preferentially set the internals – this
minimizes coalescence and therefore maximizes interfacial area.
Droplets coalesce. Droplets retain shape.
Interfacial area lost. Maximizes interfacial area.

Importance of maintaining droplets

Assume – 30% holdup of dispersed phase in 1 M3 of solution

Droplet Droplet Number Droplet Interfacial

Diameter Volume Droplets SA [M2] Area
[m] [M3] [M2/M3]
100 0.3 7.16x1010 1.26x10-7 9022
300 0.3 2.65x109 1.13x10-6 2995
500 0.3 5.73x108 3.14x10-6 1796
Factors Effecting which Phase is Dispersed
Marangoni Effect
• Coalescence is enhanced by mass transfer from
droplets continuous phase

C Mass Transfer Direction

A+B A+B Dispersed Continuous
(d c)
• Droplets tend to coalesce
A
• Must be counteracted by additional energy

C C Continuous Dispersed
A+B (c d)

C+B C+B
• Droplets tend to repel each other
• Less energy required to maintain
dispersion
Interface Behavior
Actions to control unstable interface Light Phase Heavy Phase
Dispersed Dispersed
As extraction proceeds, interface normally
grows in thickness and forms a “rag” layer
that stabilizes at some thickness
Rag
If rag layer continues to grow, some action Layer
must be taken
1. Rag Draw
Continuously withdraw a portion of the Growing
interface and pass through a filter to Uncontrolled
remove interfacial contamination Interface
2. Reverse Phases Filter
Often a stable interface can be controlled
by reversing which phase is dispersed
1 2
Entrainment
Entrainment involves carrying over a small portion of one phase out the wrong end
of the column.

Entrainment is controlled by:

1.) Increased settling time inside the column
2.) Coalescer inside the column
3.) Coalescer external to the column
E
E E
E

F F F F

OR OR
1 2 3
S S S S

R
R
R R
Flooding
Flooding – the point where the upward or downward flow of the dispersed phase
ceases and a second interface is formed in the column.

Flooding can be caused by:

• Increased continuous phase flow rate which increases drag on droplets
F2 > F1
f f
Primary Interface Primary Interface

E E
F1 F2

Second
Interface

S S

R R
Flooding
Flooding can be caused by:
• Increased agitation speed which forms smaller droplets which cannot
overcome flow of the continuous phase
• Decreased interfacial tension – forms smaller drops – same effect as
increased agitation

f2 > f 1
f1 f2
Primary Interface Primary Interface

E E
F1 F2

Second
Interface

S S

R R
Pilot Tests
Static Columns Agitated Columns
(Packed, Tray) (Scheibel, Karr)
N, S/F Process Factors N, S/F
D, H Column Variable D, H
(F+S) Variable (F+S),f
f
F

F
H

S D H

Flood S D
HETS
F+S
F+S
HETS MIN
HETS

f F+S
Extractor Flow Patterns
Ideal Plug Flow Actual Flow

Y Y

X X

This “axial” or “back” mixing causes

concentration gradients that decrease driving
force and therefore increase HETS
Generalized Scale-up Procedure
Pilot Scale Commercial Scale
f1 f2

Q1
Q2
Feed Rate
Feed Rate
H1

H2

D1

Basic Scale-up Relationships:

D2/D1 = K1(Q2/Q1 )^M1
D2
H2/H1 = K2(D2/D1 )^M2
f2/f1 = K3(D2/D1)^M3 Where:
K1, M1 = Capacity Scale-up Factors
K2, M2 = Efficiency Scale-up Factors
K3, M3 = Power Scale-up Factors
Source: Mass transfer Theory and Practice by N.Anantharaman and K.M. Meera Sheriffa Begum
Source: Mass transfer Theory and Practice by N.Anantharaman and K.M. Meera Sheriffa Begum
Source: Mass transfer Theory and Practice by N.Anantharaman and K.M. Meera Sheriffa Begum
Source: Mass transfer Theory and Practice by N.Anantharaman and K.M. Meera Sheriffa Begum
Source: Mass transfer Theory and Practice by N.Anantharaman and K.M. Meera Sheriffa Begum