Distillation

Distillation
Fundamentals
Flash distillation with binary mixture
Continuous distillation with reflux
Material balance in plate column
Number of ideal plates
Reflux ratio
Feed tray location,
Minimum number plates, minimum reflux, optimum reflux ratio
Sieve Trays, valve trays, plate efficiency
Condensers and Reboilers
Batch distillation
Distillation in packed column
Multi component distillation
Distillation - Theory
A. Vaporization and Condensation - One component system
4. When enough energy, in the form of heat, is imparted to the

solution the vapor pressure becomes equal to the atmospheric
pressure and the liquid begins to boil
P < Patm P Patm

A. Vaporization and Condensation One component system
5. The vapor obtained from a boiling liquid, once cooled, will re-
condense to a liquid known as the distillate
6. The complete process is called a distillation

Distillation Back to Theory
D. Separation of Two Liquids
1. For each component: if vapor pressure is plotted versus

temperature an exponential increase of vapor pressure is
observed as the boiling point is approached
Note: even if two compounds have the same ultimate boiling

point, the curvature of this line may be different!
Distillation Theory
2. This relationship of vapor-pressure vs. temperature is given by

the Clausius-Clapeyron equation:
p = po exp [ - H
R
(1/T 1/To) ]
The x,y (independent and dependent variables) for this equation
are the known temperature (T) and the vapor pressure (p)
calculated for that temperature
The constants for this equation:

po and To: known vapor pressure for a known temp. (K)
H: heat of vaporization of the liquid
R: gas constant (8.314 J . mol-1 . K)

1. Above a mixture of two or more volatile liquids each liquid makes

a partial contribution to the overall vapor pressure.
Pmixture = PA + PB +
2. When the sum of these partial pressures equals the atmospheric

pressure (or pressure above the mixture), the mixture boils
3. This Law implies that if a mixture of different volatile liquids is

heated to boiling and the condensed vapors are collected they will
be enriched in the component that is more volatile
more volatile = higher partial pressure, lower boiling point
4. This is the basis for using distillation as a technique for the

separation and purification of liquids.
D. Separation of Two Liquids Raoults Law
1. Raoult extended Daltons Law to illustrate that the contribution of

each components vapor pressure is related to its mole fraction in
the mixture at the interface of the liquid and vapor phases
Pmixture = XAPA + XBPB +
2. Once again at the boiling point:
Patm = XAPA + XBPB (2 component system)
3. The enrichment of a particular component in the condensed

vapors of a boiling mixture is related to both their volatility (P) and
their concentration (X) in the original mixture.
E. Combining Raoult with Clausius-Clapeyron
1. The sum of mole fractions of all components must equal one
1 = XA + XB
2. Substitute the equation for a single component into Raoults Law:
XB = 1 XA so Patm = XAPA + (1-XA)PB
3. Expansion and rearrangement of the expression gives us the

variation of mole fraction versus partial and atmospheric
pressure:
Patm - PB
(PA - PB)
XA = _______________
E. Combining Raoult and Clausius-Clapeyron:

4. If we substitute this expression:
Patm - PB
XA = __________
(PA - PB)
into the Clausius-Clapeyron Equation:
p = po exp [- H
R
(1/T 1/To) ]
We obtain an expression for the mole fraction of each component in
liquid that boils at a given temperature:
[ ]
- HB
(1/T 1/ToB)
Patm - PB exp R
XA = _________________________________________________________
[ ] [ ]
- HA _ - HB
PA exp (1/T 1/ToA) P exp (1/T 1/ToB)
R B R
E. Combining Raoult with Clausius-Clapeyron:
Combining the expressions for each component in a two component

mixture we obtain the following graphical relationship:
bp of pure A
Temperature
Vapor
bp of pure B
Liquid
0.0, 1.0 0.5, 0.5 1.0, 0.0

Mole Fraction, XA, XB
This relationship gives us the boiling point for any mixture of A and B
F. Dalton and Clausius-Clapeyron:
1. We have just described how the liquid composition relates to the

boiling temperature what is happening in the vapor phase?
2. The composition of the vapor is given by Daltons Law:

P = PA + PB
3. Substituting in the Ideal Gas Law for each component:
(PA = nA(RT)/V)
and canceling similar terms we find that the ratio of each
component to the total vapor pressure is given by:
PA/PTOTAL = nA/nTOTAL
4. Substituting mole fractions for number of moles we find that at 760
torr (1 atm) the vapor component for this system is given by:
XA vapor = XA liquid (PA/760)
If we substitute this XA into Clausius-Clapeyron:
G. We get an expression for the composition of the vapor.
Now add this relationship, graphically, of the composition of the vapor

to the mole-fraction to temperature relationship we illustrated earlier,
we arrive at the goal: X =X (P /760)
A vapor A liquid As
bp of pure A
Temperature
Vapor o m position
c
Vapor
bp of pure B
i d co mp osition
Liqu Liquid
0.0, 1.0 0.5, 0.5 1.0, 0.0

Mole Fraction, XA, XB
G. Now for any mixture of liquids, we can determine:
1. The boiling point of the mixture (liquid line)
2. The composition of the vapor (vapor line) which shows how much
enrichment in the lower boiling component occurs
110 Vapor
line
Liquid
Temperature C
100
line
90
80
Mole % Toluene 0 20 40 60 80 100

Mole % Benzene 100 80 60 40 20 0
Composition (mole%)
Distillation Simple
A. What is it?
1. A simple distillation uses one vaporization-condensation cycle to

effect a separation:
What we will
be The cooling jacket and
discussing vacuum adapter function
occurs only only to cool the vapors to
in this part liquid efficiently and direct
of the them into the receiver
apparatus flask
The distilling
flask is directly
attached to the
distillation head
Distillation Simple
B. How efficient is it?
Lets use the graphical representation we derived earlier to illustrate

what occurs in a simple distillation:
110 Vapor
line
Liquid
Temperature C
100
line
90
80
Mole % Toluene 0 20 40 60 80 100

Mole % Benzene 100 80 60 40 20 0
Composition (mole%)
Distillation Simple
B. How efficient is it?
Suppose we have a 80:20 mixture of benzene and toluene and we

subject it to a simple distillation technique:
110 Vapor
line
From our
Liquid
Temperature C
graph, we see 100

line
that this
mixture will 90
boil at ~100 C
80
Mole % Toluene 0 20 40 60 80 100

Mole % Benzene 100 80 60 40 20 0
Composition (mole%)
Distillation Simple
Vapor
110
line
Temperature C
100 Liquid
line
90
80
Mole % Toluene 0 20 40 60 80 100

Mole % Benzene 100 80 60 40 20 0
Composition (mole%)
Distillation Fractional
A. What is it?
A fractional distillation utilizes two or more vaporization-
condensation cycles, in succession, to effect a separation.
This is accomplished by what distinguishes a

fractional distillation apparatus:
the fractionating column
The fractionating column causes the

vaporization-condensation cycle to repeat
by providing multiple surfaces for the cycle
to take place
Using our graphical representation of the

benzene-toluene mixture as an example
lets see how this works.
110 Vapor
line
Liquid
Temperature C
100
line
90
80
Mole % Toluene 0 20 40 60 80 100

Mole % Benzene 100 80 60 40 20 0
Composition (mole%)
As the hot vapors leave the distilling flask,

they condense on the first cold surface, completing
one vaporization-condensation cycle.
Vapors from the Suppose we distill the same 80:20 mixture of toluene
Distilling flask to benzene we did in the simple distillation example
110 Vapor
line
Liquid
Temperature C
100
line
90
80
Mole % Toluene 0 20 40 60 80 100

Mole % Benzene 100 80 60 40 20 0
Composition (mole%)
This surface begins to heat from the condensed vapors

which are now 55:45 toluene-benzene
This benzene enriched liquid now has a boiling point of ~94

C (lower than the incoming vapors) and it begins to boil off
Vapors from the
this higher surface
Distilling flask
110 Vapor
line
Liquid
Temperature C
100
line
90
80
Mole % Toluene 0 20 40 60 80 100

Mole % Benzene 100 80 60 40 20 0
Composition (mole%)
These vapors are even further enriched in benzene (now

30:70, toluene:benzene) and condense on the next cold
surface
Vapors from the
Distilling flask
110 Vapor
line
Liquid
Temperature C
100
line
90
80
Mole % Toluene 0 20 40 60 80 100

Mole % Benzene 100 80 60 40 20 0
Composition (mole%)
This condensed liquid has an even lower boiling point (86

C) and as this surface heats it begins to boil off this next
higher surface
Vapors from the
Distilling flask
110 Vapor
line
Liquid
Temperature C
100
line
90
80
Mole % Toluene 0 20 40 60 80 100

Mole % Benzene 100 80 60 40 20 0
Composition (mole%)
This vapor now condenses on the next cold surface (now

20:80, toluene:benzene) and the cycle continues
Vapors from the
Distilling flask
1:99 toluene:benzene
110 Vapor
line
Liquid
Temperature C
100
line
90
80
Mole % Toluene 0 20 40 60 80 100

Mole % Benzene 100 80 60 40 20 0
Composition (mole%)
This cycle will continue until the top of the column is reached
The liquid collected after seven cycles is now 99% benzene!
Vapors from the

Distilling flask
80:20 toluene-benzene
1:99 toluene:benzene
110 Vapor
line
Liquid
Temperature C
100
line
90
80
Mole % Toluene 0 20 40 60 80 100

Mole % Benzene 100 80 60 40 20 0
Composition (mole%)
Each vapor-condensation (or mini-distillation) cycle is

known as one theoretical plate
The length of distillation column required to provide one

Vapors from the theoretical plate of separation is known as the height
Distilling flask equivalent theoretical plate (HETP)
80:20 toluene-benzene
Important 110 Vapor

line
What we have discussed is
only true for the first drop
Liquid
Temperature C
100
of distillate! line
As the distillation flask 90

loses what the vapor is
enriched in,
The starting point for the 80
next drop of distillate will

be slightly different! Mole % Toluene 0 20 40 60 80 100
Mole % Benzene 100 80 60 40 20 0
Composition (mole%)
In our example, there will
be more and more toluene
in the distillation flask
more heat will need to be More and
applied to get the liquid to more
boil, and heat the toluene as
distillation column distillation
proceeds
Industrially,
fractional
distillation is
very
common and
is typically
run as a
continuous
process
Design consideration
Assumptions
The McCabe-Thiele method is based on the assumption of constant molar

overflow which requires that:
the molar heats of vaporization of the feed components are equal
for every mole of liquid vaporized, a mole of vapour is condensed
heat effects such as heats of solution and heat transfer to and from the
distillation column are negligible.
Y = (q/q-1)x ZF/(q-1)
Location of feed stage
a)Optimal feed stage location

b)Below optimal
c)Above optimal stage
Distillation with Multiple feeds and side stream
Extended use of McCabe Thiele method
In absence of side stream Ls, this

arrangement has no effect on the material
balance associated with the rectifying
section of the column above the upper feed
point F1
The section between upper feed

point and lower feed point F2 (in
absence of feed F) is represented
by an operating line of slope L/V,
it intersects the rectifying sections
operating line
A similar argument hold for the

stripping section of the column
For two feed points the
molar vapour flow rate is A saturated liquid side stream of composition
V = V F1 xs and molar flow rate Ls is withdrawn from the
and rectifying section above feed F
L = L + F2 = L + F2 In the section of stages between the side
stream withdrawal stage and feed stage
For xB, zF2, zF1, xD and L/D three L = L Ls while V = V
operating lines are
Two feeds saturated liquid and saturated One feed one side stream (saturated
vapour liquid)
Example
A complex distillation column, equipped with a partial reboiler and total

condensor and operating at steady state with a saturated liquid feed has a liquid
side stream drawn-off in the enriching (rectifying ) section. Making the usual
simplifying assumptions of the McCabe Thiele method :
(a)Derive an equation for the two operating lines in the enriching section
(b)Find the point of intersection of these operating lines
(c)Find the intersection of the operating line between F and Ls with the
diagonal . (d) Show the construction of x-y diagram
Solution
Multi-component distillation
Theoretical model for an equilibrium stage
Assumptions:
Phase equilibrium at each stage
No chemical reactions
Entrainment of liquid drops in vapour and

occlusion of vapour bubble in liquid are
negligible
MESH equations
M equations - Material balance of each

component (C equation for each stage)
E equations phase equilibrium relation for each

component (C equations for each stage)
Where Ki,j is the phase equilibrium ratio

S equations mole fraction summations (one for each stage)
H equation energy balance (one for each stage)
Where kinetic and potential energy changes are ignored
In general Ki,j = Ki,j (Tj, Pj, xj,yj) ; hvj = hvj (Tj, Pj, yj)
hLj = hLj (Tj, Pj, xj)
Total material balance, combining mole fraction

summations over the C components and over stages
1 through j
If N and all Fj, zi,j, TFj, PFj, Pj, Uj, Wj, and
Qj are specified the model is represented
by N(2C+3) simultaneous algebric
For each stage 2C + 3 MESH equation equations with N(2C+3) unknown / output
variable
Mathematical solution
Equation portioning in conjunction with equation tearing and / or linearization by
Newton Rapshon techniques
Lewis Matheson 1932 stage by- stage, equation by equation calculational
procedure
Thiele-Geddes 1933- equation tearing for solving simple fractionators with one feed
and two products
Holland improved Thiele Gaddes procedure (theta method)
Amundson and Pontinen 1958 solved component by-component from
simultaneous liner equation sets for all N stages by an equation tearing procedure
using same tear variable as the Thiele-Geddes
Friday and Smith 1964 systematically analyzed a number of tearing techniques for
solving the MESH equations. No one tearing technique could solve all types of
problems
Bubble point (BP) method recommended for separators where the feed(s) contains
only component of similar volatility (narrow boiling case)
Sum-rates (SR) method was suggested for a feed(s) containing components of widely
different volatility (wide-boiling case) or solubility
For intermediate case Friday and Smith indicated that either a Newton Raphson
method or a combined tearing and Newton Raphson method was necessary
Boston and Sullivan 1974 presented an alternative robust approach(inside-outside
method)
Present practice is based on BP, SR, Newton-Raphson and Inside outside methods
BP method for distillation
For most problems it is sufficient to

establish an initial set of Vj values
based on the assumption of constant
molar inter stage flows using the
specified reflux, distillate feed and side
stream flow rates
Adequate initial set of Tj values can be
provided by computing or assuming
both the bubble point of the estimated
bottom product and the dew point of an
(10-2) assumed vapour distillate product; or
computing or assuming bubble point
temperature if distillate is liquid or a
temperature in between the dew point
and bubble point temperature if
distillate is mixed (both vapour and
liquid) ; and then other stage
temperatures by assuming a liner
variation of temperature with stage
location
Elimination of y and L in MESH equations (substitution
Tridigonal matrix algorithm of y of E equation in M equation ; substitution of L of
total material balance on M equn.
Ajxi,j-1 + Bjxi,j + Cjxi,j+1 = Dj
Where
Aj
2jN
Bj = [
+ Uj + (Vj + Wj)Ki,j] 1 j N
Cj = Vj+1 Ki,j+1 ], 1 j N
(10-12)
Dj = - Fjzi,j, 1 j N
To solve xi, ki,j values are required. When these are

composition dependent assumptions for all xi,j and yi,j
values are also needed unless ideal K-values are employed
for the first iteration.
For each iteration , the computed set of xi,j values for each
stage will , in general, not satisfy the summation constraint
Normalized xi,j
These normalized values are used
for all subsequent calculations
involving xi,j during the iteration
(10-19)
A new set of temperatures Tj is computed stage by stage by computing bubble

point temperatures from the normalized xi,j values
Friday and Smith showed that bubble point calculations for stage temperatures are
particularly effective for mixtures having narrow range of K values because
temperature are not then sensitive to composition.
Hence for limiting case where all components have identical K values , the
temperature corresponds to the conditions of Ki,j = 1 and is not dependant on xi,j
values.
Bubble point equation which is non linear in Tj, iterative sol.

(10-20)
Liq. Mol. Fr. of CH4 0.000 0.018 0.054 0.093
BP 0F 275 250 200 150

Muller s iterative method
Does not require the calculation of derivatives
It requires three initial assumptions of Tj
For each assumption the value of Sj is computed from Sj =
Three sets of Tj and Sj are fitted in quadratic equation for Sj in terms of Tj, the
quadratic equation is then employed to predict Tj for Sj = 0
The validity of this value of Tj is checked by using it to compute Sj

The quadratic fit and Sj check are repeated with the three best sets of (Tj, Sj) until
some convergence tolerance is achieved say Tj (n) Tj (n-i) / Tj (n) 0.0001
T in absolute degree and n is the iteration number for the temperature loop in the
bubble point calculation
Or Sj 0.0001C
Values of yi,j are determined along with the calculation of stage temperatures using
the E equations
With a consistent set of values for xi,j, Tj and yi,j , molar enthalpies are computed for
each liquid and vapour stream leaving a stage
Since F1, v1, U1, W1 and L1 are specified, V2 is readily obtained from
Condenser duty (+) is obtained from
(10-5)
Reboiler duty (-) is obtained from
(10-22)
Combining the following equations twice to eliminate Lj-1 and Lj
and
After rearrangement
,
and
and enthalpies are computed at

stage temperature
Equation
In didiagonal matrix form applied

over 2 to N-1 stage :
The matrix equation is readily solved

one equation at a time by starting at the
top where V2 known and working down
recursively
Thus or in
general (10-30)
Corresponding liquid flow rates are obtained from

(10-6)
The solution procedure is considered to be converged when sets of Tj(k) and Vj(k)
are within some prescribed tolerance of corresponding sets of Tj(k-1) and Vj(k-1)
values
One possible convergence criterion is
Where T is absolute temperature and

is some prescribed tolerance
Wang and Henke suggests following simple criteria which is based on successive
sets of Tj values only
(10-32)
To prevent oscillating of the iterations, damping can be employed to limit changes

in the values of Vj and Tj from one iteration to the next to say 10 %
Thomas algorithm for solving the linearized matrix equation set
Gaussian elimination procedure that involves forward elimination starting

from stage 1 and working toward stage N to finally isolate xi,N. Other values of
xi,j are then obtained starting with xi,N-1 by backward substitution
For five stages the matrix equations at the

beginning, middle and end of the procedure are as
shown in the following figure
Ajxi,j-1 + Bjxi,j + Cjxi,j+1 = Dj
Bjxi,j + Cjxi,j+1 = Dj (for stage 1)
p1 = C1/B1 and
q1 = D1/B1 Xi,1 = q1 p1xi,2
generalized
pj = Cj/(Bj-Ajpj-1)
And
qj = (Dj-Aj qj-1)/ ((Bj Ajpj-1)
Then xi,j = qj pjxi,j+1

Xi,j-1 = qj-1 pj-1xi,j = rj-1
yi = Kixi or Ki = yi/xi
yiP = xiPi sat or Ki = Pisat/P
logPisat = Ai Bi/(Ci+T) (Antoine Equ.)
A, B, C values of different components from literature
P = Pixi T values (bubble point feed temp.)

Example
For the distillation column as shown below, do one iteration of the BP method upto and
including the calculation of a new set of Tj values. Use composition independent K
values
Matrix equation for the first component is developed with V1 =0 , W =0 as:
with V1 =0 , W =0
Substituting these values in matrix equation
Solution through
Thomas algorithm
After these compositions are normalized bubble point temperatures at 100 psi
are computed iteratively from
Approximate methods for multi component multi stage separation
Short cut method (FUG method)
For multi component distillation, select two key component, operating pressure
and type condenser
For specified separation between two key components in multi componet distillation
column
Fenske equation Minimum number of equilibrium stages and distribution of non

key components
Underwood method Minimum reflux ratio
Gilliland correlation Number of reflux ratio greater than minimum
Krikbride equation Feed stage location

Distribution of non key component at total reflux
i = non key component

r = reference / heavy
key component
di: Xi ; bi = 1-Xi
dr: XB; br: 1-XB
= m
For actual reflux ratio
and theoretical stages
Design aspect of atmospheric distillation
LPG
Hynap
Preheat Desalter
Stabiliser
Preflash Heater Kerosene
Gas oil Lt nap
Steam Atm
residue
Major section of a atmospheric distillation unit
LPG
LIGHT NAPHTHA
HEAVY NAPHTHA
Major products AVIATION TURBINE FUEL (ATF)
KEROSENE
LIGHT GAS OIL HIGH
SPEED DIESEL
HEAVY GAS OIL
COMPONENTS
Broad product specification
ATF
LPG AROMATICS 22.0 (20)
VAPOUR PRESSURE 16.87 Kg/cm Max
2
VOLATILITY (EVAP. TEMP. 2oC Max

FOR 95% VOL. AT 760 MMHG) DISTILLATION
10% Vol.
GASOLINE 205 Max.
DISTILLATION FBP 300 Max.
REC. UPTO 70oC 10-40%
SMOKE POINT 26
REC. UPTO 100oC 30-65%
MM Min.
REC. UPTO 180 C
o
80% (Min.)
FREEZING POINT -47o
C Max.
FBP 215oC
FLASH POINT (ABEL) 38oC Min.
RVP AT 38 C
o
35-70 KPA/cm 2
RVP AT 38oC 35-70
(SULPHUR AND BENZENE
KPA/cm2
SPEC. BEING REVISED)
Broad product specification
FUEL OIL
KEROSENE
SULPHUR 3.5 TO 4.5 Wt.%
VOL. % BELOW 200oC 20
VISCOSITY @ 50oC, CST 80, 125,
FBP 300 C Max.
o
180, 370
FLASH POINT (ABEL) 35 C Min.
o
FLASH POINT 66oC
SMOKE POINT 18 MM Min.
Min.
SULPHUR 0.25 Wt. % Max.
HIGH SPEED DIESEL

DISTILLATION (REVISED)
90% VOL. 366oC (85% < 350oC)
(95% < 370oC)
FLASH POINT (ABEL) 32oC (35oC)
SULPHUR 0.25 Wt. % (0.05 Wt. % BEING
REVISED
FURTHER DOWN)
CETANE No. 45 (48, 51 LATER)
Atmospheric distillation unit / column
TYPICAL 40 TO 50 ACTUAL TRAYS

No. OF TRAYS FROM BOTTOM
NAPHTHA 4 TO 6 BOTTOM STRIPPING
STEAM 4 TO 6 FLASH ZONE / HEAVY GAS
KERO OIL
STEAM 4 TO 6 HEAVY GAS OIL / LIGHT
LGO GAS OIL
STEAM 4 TO 6 LIGHT GAS OIL /
HGO KEROSENE
CRUDE
4 TO 8 KEROSENE / HEAVY
STEAM
RCO NAPHTHA
4 TO 8 HEAVY NAPHTHA / LIGHT
PRESSURE NAPHTHA
TOP : BUBBLE PRESSURE BASED ON COOLANT TEMPERATURE
2 TO 3 PUMP AROUNDS EACH
PRESSURE DROP IN CONDENSER 0.3 TO 0.5
Kg/cm 2
TRAYS PRESSURE DROP : 0.08 TO 0.1 Kg/cm2/

TRAY
TEMPERATURE
TOP : 8oC TO 10oC ABOVE WATER DEW POINT
(120oC TO 130oC TYPICAL)
Atmospheric distillation column
PUMP AROUND
2 TO 3 TRAYS
60 TO 80oC TEMPERATURE DROP
GOVERN COLUMN SIZING
REDUCED FRACTIONTAION
TYPICAL HEAT TRANSFER COEFFICIENT 7500 TO 10000
Kcal/Hr/M2/oC
TYPICALLY AT PRODUCT DRAW TRAY
INTERNAL REFLUX / HYDROCARBON VAPOUR 0.3 TO 0.4 ON
Stabilizer
MOLAR BASIS
PRESSURE : MIN. BUBBLE PRESSURE AT COOLING MEDIA

TEMPERATURE
LPG SPECS : (A) VAPOUR PRESSURE AT 65oC=16.87 Kg/cm2g
(B) VAPORISATION AT 2oC AND 760 MMHG = 95%
VOL. MIN.
(CORRESPONDS TO 1.0 Wt % C5 PLUS IN LPG)
No.
LPG OF TRAYS
RECOVERY :: 30
90%TOPLUS
40 (VALVE TRAYS) (TYPICAL)
(TYPICAL)
REFLUX RATIO
RVP OF STABILISED : :0.3 TO 0.7
2.5Kg/cm
TO 3.5
2 (TYPICAL)
NAPHTHA (TYPICAL)
Major design variables
DESALTER TEMPERATURE
PREFLASH
HEATER OUTLET TEMPERATURE
FLASH ZONE TEMPERATURE
OVER FLASH
COLUMN / FLASH ZONE PRESSURE
TYPE OF CONDENSER (PARTIAL / TOTAL)
LOCATION AND NUMBER OF PUMP AROUNDS
STRIPPING STEAM
No. OF TRAYS
VACUUM DISTILLATION UNIT
WHY VACUUM UNIT ?

TO VAPORISE AND FRACTIONATE AT TEMPERATURE BELOW
CRACKING BY REDUCING PRESSURE
TYPES OF VACUUM
DRY TOP PRESSURE 10 TO 15 MMHGA, NO STEAM
WET TOP PRESSURE 70 TO 80 MMHGA, STEAM
DAMP LOW STEAM INTERMEDIATE PRESSURE
HOW VACUUM ACHIEVED ?

STEAM EJECTOR
VACUUM PUMP
TYPES
FUEL BASED LUBE BASED
PREPARE FEEDSTOCK FOR PREPARE FEEDSTOCK FOR
FCC, HYDROCRACKER LUBE PLANT
PRODUCTS
DIESEL PRODUCTS
LIGHT VACUUM GAS OIL SPINDLE OIL
(LVGO)
LIGHT NEUTRAL
HEAVY VACUUM GAS OIL
(HVGO) INTERMEDIATE NEUTRAL
SLOP DISTILLATE HEAVY NEUTRAL
VACUUM RESIDUE
VACUUM RESIDUE
TYPICAL HYDROCRACKER FEED SPECIFICATIONS

ASTM END POINT 565oC MAX
ASPHALTENES 0.01% HEPTANE
INSOLUBLES
CCR 0.5% WT. MAX
Ni + Va 1.0 PPM MAX
Na & MATERIALS 1.0 PPM MAX
ASTM COLOUR + 4 MAX
TYPICAL FCCU SPECIFICATIONS

METAL FACTOR 2.5 MAX (10 Ni + V)
CCR 0.5% WT MAX
TYPICAL LUBE PRODUCT SPECIFICATION
VISCOSITY FLASH POINT CCR % WT

CST @ 98.9oC o
C
GAS OIL 1.3-1.6 120 -
SPINDLE OIL 3.5-3.6 160+ -
LIGHT NEUTRAL 6.0 MIN 200+ -
INT. NEUTRAL 9.0-11.0 230-240 -
HEAVY NEUTRAL 19.0-24.0 240-260 2.4 MAX
SHORT RESIDUE 1200 300-320 18-20
LUBE COLUMNS ARE TYPICALLY WET VACUUM AS

PRODUCTS NEED TO MEET FLASH POINT SPEC
BETTER PRODUCT SEPARATION REQUIRED
VACUUM COLUMN
NON-CONDENS.
TO VACUUM SYSTEM
LVGO
SLOP MAX HVGO
WASH OIL
TOPPED CRUDE
VACUUM RESID
PROCESS
STEAM
STEAM
VACUUM UNIT DESIGN ISSUES
PREHEAT TRAIN
TYPE OF VACUUM / PRESSURE
STEAM RATES (EJECTOR / STRIPPING /
VELOCITY)
TYPE OF INTERNALS
TOP TEMPERATURE
NUMBER LOCATION OF PUMP AROUNDS
EJECTOR VS VACUUM PUMPS OR
COMBINATION
VACUUM UNIT DESIGN PARAMETERS
PREHEAT TRAIN
SUITABLE FOR LIGHT / HEAVY FEED
SUITABLE FOR COLD / HOT FEED
OPTIMIZED USING PINCH ANALYSIS
MAXIMUM OUTLET TEMPERATURE 320oC TO 325oC (TYPICAL)

LOWER THAN CRACKING TEMPERATURE
425oC MAXIMUM WITH STEAM INJECTION (TYPICAL)

SUITABLE TO VAPORIZE PRODUCTS + OVER FLASH
405oC TO 410oC MAXIMUM (TYPICAL)
VACUUM COLUMN - TYPICAL
OPERATING DATA
DRY DAMP WET

TOP PRESSURE MMHGA 8 25
75
FLASH ZONE PRESSURE MMHGA 24 42
95
HEATER OUTLET TEMP. 395 405
420
COLUMN DIA HIGHEST INT LEAST
STEAM NIL INT HIGHEST
TRANSFER LINE HIGHEST LEAST
INT
TYPICAL PRESSURE DROPS

PRECONDENSER 3 TO 5 MMHG
OVERHEAD LINE 2 TO 5 MMHG
PACKED SECTION 1.0 MMHG / 1.0 M PACKING
CHIMNEY TRAY 1.0 TO 1.5 MMHG
DEMISTERS 1 MMHG
VALVE TRAY 2-3 MMHG / TRAY
GRID 1.0 MMHG / 1.5 M
STRIPPING STEAM
INJECTED IN HEATER (VELOCITY STEAM)
INJECTED IN COLUMN BOTTOM (MAX 36 KG / HR / STD M3 OF
RCO)
VACUUM COLUMN
TYPICAL PACKED WITH RANDOM/STRUCTURED PACKING
SECTIONS PURPOSE
#1 DIESEL FRACTIONATION
#2 LVGO HEAT TRANSFER
#3 HVGO HEAT TRANSFER FOR FUEL
REFINERY
#4 WASH ZONE FRACTIONATION
TYPICAL PUMP AROUNDS

TEMPERATURE DROP 60oC TO 80oC
AT DRAW TRAY
AT HVGO / LVGO / TOP
VACUUM DISTILLATION
VACUUM COLUMN
UNIT
WASH ZONE
LIQUID WETTING RATE : 0.2 TO 0.3 GPM/FT2
0.1 MIN FOR
STRUCTURED
PACKING
DISTRIBUTORS
SPRAY TYPE FOR HEAT TRANSFER
TROUGH TYPE FOR MASS TRANSFER
(WASH ZONE FOR DEEP CUTS)
TOP TEMPERATURE
80oC TO 1000C TYPICAL
TYPICAL LOCATION OF PUMP AROUNDS
AT HVGO, LVGO, DIESEL DRAW OFFS.
EJECTOR / VACUUM PUMPS

EJECTOR RELIABLE (LESS MAINTENANCE)
(HIGHER OPERATING COST)
VACUUM PUMPS LESS OPERATING COST

MORE MAINTENANCE
EJECTOR LOAD AND ESTIMATION
NON-CONDENSIBLE (AIR LEAKAGE + CRACKED PRODUCTS)
CONDENSIBLES - STEAM
AIR LEAKAGE ESTIMATION

BASED ON THROUGHPUT
COLUMN VOLUME / No. OF CONNECTION
CRACKED PRODUCTS ESTIMATION

BASED ON HEATER / FLASH ZONE TEMPERATURE
VACUUM HEATER
MIN. MASS VELOCITY SUITABLE TO MINIMIZE FILM

TEMPERATURE
PRESSURE DROP 2 TO 5 Kg/ cm2 TYPICAL
AVERAGE RADIANT 27100 Kcal/Hr/m2
HEAT FLUX
TYPE VERTICAL / BOX
AIR PREHEATER CAST / PLATE / GLASS
EFFICIENCY 85 TO 90%
TUBES NORMALLY SIZE INCREASE AT
OUTLET
VACUUM HEATER
COKE LAYER 5 TO 6 mm
VELOCITY 80% OF SONIC
TRANSFER LINE 10oC to 20o
TEMPERATURE DROP
SIZE OF TUBES 8 INCH MAX.
MOC OF TUBES 9 CR, 1 Mo
INJECTION STEAM WHERE
VAPORISATION STARTS
DEEP CUT VACUUM DISTILLATION
TYPICAL CUT 370oC TO 530oC / 550oC AS VGO

DEEP CUT 370oC TO 570oC / 600oC AS VGO
ADVANTAGES
HIGHER VGO 2 TO 4%
LESSER LOW VALUE RCO
PROBLEMS
AVAILABILITY OF DATA BEYOND 550oC TBP
METALS
PHYSICAL PROPERTIES
ENTRAINMENT
TENDENCY OF COKING HIGHER
SOLUTIONS
RAISE HEATER OUTLET (MAX 425oC) FLASH ZONE
TEMPERATURE
LOWER FLASH ZONE PRESSURE
DAMP VACUUM (STEAM INJECTION)
REDUCE OVER FLASH
TRANSFER LINE SIZE (80 TO 90% OF SONIC)
FEED DE-ENTRAINMENT DEVICES
IMPROVE WASH ZONE INTERNALS (2 TO 3 STAGES)
GRID / STRUCTURED
TOO HIGH COKING
TOO LOW LESS MASS TRANSFER
MORE TEMPERATURE MEASUREMENT IN WASH BED
HETP (M)
GRID 1.5 TO 2.0
3 PALL RINGS 1.4 TO 1.5
STRUCTURED PACKING 1.0 TO 1.2
(25 MM CRIMP)
WASH ZONE
CHIMNEY TRAY MINIMUM RESIDENCE TIME
NORMALLY PACKED SINGLE BED
DOUBLE BED
ONE WITH SLOP WASH
SECOND WITH CLEAN HVGO WASH
OVER FLASH
3 TO 4% VOL. TYPICAL
0.1 GPM / FT2 MIN LIQUID RATE (0.3 REFERRED)
DISTRIBUTOR
SPRAY TYPE
TO HAVE 200 TO 300% OVERLAP
WIDE ANGLE SPRAY AVOID
NOZZLE TO BE 1200 TO 1500 MM ABOVE PACKING
EACH NOZZLE TO COVER 0.5 TO 0.7 M2
BOTTOM ZONE
QUENCH TO MAINTAIN 370oC MAX
MINIMIZE RESIDENCE TIME
HVGO PACKED SECTION
HEAT TRANSFER - PACKED BED
1500 MM MINIMUM
2000 TO 3000 MM HEIGHT TYPICAL
LVGO PACKED SECTION
HEAT TRANSFER - PA
1500 MM MINIMUM
2000 TO 3000 MM HEIGHT TYPICAL
DIESEL SECTION
MASS TRANSFER / HEAT TRANSFER
2 TO 3 THEORETICAL STAGES
200 TO 300 MM HIGH

Distillation

Uploaded by

Copyright:

Available Formats

Distillation

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Distillation

Uploaded by

Copyright:

Available Formats

Distillation

A. Vaporization and Condensation - One component system

4. When enough energy, in the form of heat, is imparted to the

P < Patm P Patm

A. Vaporization and Condensation One component system

6. The complete process is called a distillation

D. Separation of Two Liquids

1. For each component: if vapor pressure is plotted versus

Note: even if two compounds have the same ultimate boiling

D. Separation of Two Liquids

2. This relationship of vapor-pressure vs. temperature is given by

The constants for this equation:

H: heat of vaporization of the liquid

R: gas constant (8.314 J . mol-1 . K)

D. Separation of Two Liquids

1. Above a mixture of two or more volatile liquids each liquid makes

2. When the sum of these partial pressures equals the atmospheric

3. This Law implies that if a mixture of different volatile liquids is

more volatile = higher partial pressure, lower boiling point

4. This is the basis for using distillation as a technique for the

D. Separation of Two Liquids Raoults Law

1. Raoult extended Daltons Law to illustrate that the contribution of

Pmixture = XAPA + XBPB +

2. Once again at the boiling point:

Patm = XAPA + XBPB (2 component system)

3. The enrichment of a particular component in the condensed

E. Combining Raoult with Clausius-Clapeyron

1. The sum of mole fractions of all components must equal one

2. Substitute the equation for a single component into Raoults Law:

XB = 1 XA so Patm = XAPA + (1-XA)PB

3. Expansion and rearrangement of the expression gives us the

E. Combining Raoult and Clausius-Clapeyron:

E. Combining Raoult with Clausius-Clapeyron:

Combining the expressions for each component in a two component

0.0, 1.0 0.5, 0.5 1.0, 0.0

F. Dalton and Clausius-Clapeyron:

1. We have just described how the liquid composition relates to the

2. The composition of the vapor is given by Daltons Law:

G. We get an expression for the composition of the vapor.

Now add this relationship, graphically, of the composition of the vapor

0.0, 1.0 0.5, 0.5 1.0, 0.0

G. Now for any mixture of liquids, we can determine:

1. The boiling point of the mixture (liquid line)

Mole % Toluene 0 20 40 60 80 100

1. A simple distillation uses one vaporization-condensation cycle to

B. How efficient is it?

Lets use the graphical representation we derived earlier to illustrate

Mole % Toluene 0 20 40 60 80 100

B. How efficient is it?

Suppose we have a 80:20 mixture of benzene and toluene and we

graph, we see 100

Mole % Toluene 0 20 40 60 80 100

Mole % Toluene 0 20 40 60 80 100

This is accomplished by what distinguishes a

the fractionating column

The fractionating column causes the

Using our graphical representation of the

Mole % Toluene 0 20 40 60 80 100

As the hot vapors leave the distilling flask,