Pdfcoffee.com Distillationpdf 3 PDF Free

NPTEL – Chemical – Mass Transfer Operation 1
MODULE 5: DISTILLATION
LECTURE NO. 1
5.1. Introduction
Distillation is method of separation of components from a liquid mixture which
depends on the differences in boiling points of the individual components and the
distributions of the components between a liquid and gas phase in the mixture.
The liquid mixture may have different boiling point characteristics depending on
the concentrations of the components present in it. Therefore, distillation
processes depends on the vapor pressure characteristics of liquid mixtures. The
vapor pressure is created by supplying heat as separating agent. In the
distillation, the new phases differ from the original by their heat content. During
most of the century, distillation was by far the most widely used method for
separating liquid mixtures of chemical components (Seader and Henley, 1998).
This is a very energy intensive technique, especially when the relative volatility of
the components is low. It is mostly carried out in multi tray columns. Packed
column with efficient structured packing has also led to increased use in
distillation.
5.1.1. Vapor Pressure

Vapor Pressure: The vaporization process changes liquid to gaseous state. The
opposite process of this vaporization is called condensation. At equilibrium, the
rates of these two processes are same. The pressure exerted by the vapor at this
equilibrium state is termed as the vapor pressure of the liquid. It depends on the
temperature and the quantity of the liquid and vapor. From the following
Clausius-Clapeyron Equation or by using Antoine Equation, the vapor pressure
can be calculated.
Joint initiative of IITs and IISc – Funded by MHRD Page 1 of 9

Clausius-Clapeyron Equation:
 p v     1 1 
ln  v       (5.1)
 p1   R  T1 T 
where p v and p1v are the vapor pressures in Pascal at absolute temperature T
and T1 in K. λ is the molar latent heat of vaporization which is independent of
temperature.
Antoine Equation:
B (5.2)
ln p v (Pascal)  A 
T C
Typical representative values of the constants A, B and C are given in the
following Table 5.1 (Ghosal et al., 1993)
Table 5.1: Typical representative values of the constants A, B and C
Components Range of A B C
Temperature (T),
K
Acetone 241-350 21.5439 2940.46 -35.93
Ammonia 179-261 21.8407 2132.50 -32.98
Benzene 280-377 20.7934 2788.51 -52.36
Ethanol 270-369 21.8045 3803.98 -41.68
Methanol 257-364 23.4801 3626.55 -34.29
Toluene 280-410 20.9063 3096.52 -53.67
Water 284-441 23.1962 3816.44 -46.13
Readers are suggested to revise the thermodynamics for more about vapor
pressure and boiling point.
5.1.2. Phase Diagram

For binary mixture phase diagram only two-component mixture, (e.g. A (more
volatile) and B (less volatile)) are considered. There are two types of phase
diagram: constant pressure and constant temperature.
5.1.3. Constant Pressure Phase Diagram

The Figure 5.1 shows a constant pressure phase diagram for an ideal solution
(one that obeys Raoult's Law). At constant pressure, depending on relative

concentrations of each component in the liquid, many boiling point temperatures

are possible for mixture of liquids (solutions) as shown in phase disgram diagram
(Figure 5.1). For mixture, the temperature is called bubble point temperature
when the liquid starts to boil and dew point when the vapor starts to condense.
Boiling of a liquid mixture takes place over a range of boiling points. Likewise,
condensation of a vapor mixture takes place over a range of condensation
points. The upper curve in the boiling point diagram is called the dew-point curve
(DPC) while the lower one is called the bubble-point curve (BPC). At each
temperature, the vapor and the liquid are in equilibrium. The constant pressure
phase diagram is more commonly used in the analysis of vapor-liquid
equilibrium.
Figure 5.1: Phase diagram of binary system at constant pressure
5.1.4. Constant temperature (isothermal) phase diagram
The constant temperature phase diagram is shown in Figure 5.2. The constant
temperature phase diagram is useful in the analysis of solution behaviour. The
more volatile liquid will have a higher vapor pressure (i.e. pA at xA = 1.0)

Figure 5.2: Binary Phase diagram at constant temperature
5.1.5. Relative volatility

Relative volatility is a measure of the differences in volatility between two
components, and hence their boiling points. It indicates how easy or difficult a
particular separation will be. The relative volatility of component ‘A’ with respect
to component ‘B’ in a binary mixture is defined as
y A / xA
 AB  (5.3)
y B / xB
where, yA = mole fraction of component ‘A’ in the vapor, xA = mole fraction of
component ‘A’ in the liquid. In general, relative volatility of a mixture changes with
the mixture composition. For binary mixture, xB = 1-xA. So Equation (5.3) can be
rearranged, simplifying and expressed by dropping subscript 'A' for more volatile
component as:
 ave x
y (5.4)
1  ( ave  1) x
The Equation (5.4) is a non-linear relationships between x and y. This Equation

can be used to determine the equilibrium relationship (y vs. x) provided the

average relative volatility, ave is known. If the system obeys Raoult’s law, i.e
p A  Py A , pB  Py B , the relative volatility can be expressed as:
pA / xA
 AB  (5.5)
p B / xB
where pA is the partial pressure of component A in the vapor, pB is the partial
pressure of component B in the vapor and P is the total pressure of the system.
Thus if the relative volatility between two components is equal to one, separation
is not possible by distillation. The larger the value of , above 1.0, the greater the
degree of separability, i.e. the easier the separation. Some examples of optimal
relative volatility that are used for distillation process design are given in Table
5.2.
Table 5.2: Some optimal relative volatility that are used for distillation process design
More volatile component Less boiling component Optimal Relative

(normal boiling point oC) (normal boiling point, oC) volatility
Benzene (80.1) Toluene (110.6) 2.34
Toluene (110.6) p-Xylene (138.3) 2.31
Benzene (80.6) p-Xylene (138.3) 4.82
m-Xylene (139.1) p-Xylene (138.3) 1.02
Pentane (36.0) Hexane (68.7) 2.59
Hexane (68.7) Heptane (98.5) 2.45
Hexane (68.7) p-Xylene (138.3) 7.0
Ethanol (78.4) iso-Propanol (82.3) 1.17
iso-Propanol (82.3) n-Propanol (97.3) 1.78
Ethanol (78.4) n-Propanol (97.3) 2.10
Methanol (64.6) Ethanol (78.4) 1.56
Methanol (64.6) iso-Propanol (82.3) 2.26
Chloroform (61.2) Acetic acid (118.1) 6.15
5.1.6. Vapor-Liquid Equilibrium (VLE)

It is useful for graphical design in determining the number of theoretical stages
required for a distillation column. A typical equilibrium curve for a binary mixture
on x-y plot is shown in Figure 5.3. It can be plotted by the Equations (5.4) or (5.5)
as discussed earlier section. It contains less information than the phase diagram
(i.e. temperature is not included), but it is most commonly used. The VLE plot

expresses the bubble-point and the dew-point of a binary mixture at constant

pressure. The curved line in Figure 5.3 is called the equilibrium line and it
describes the compositions of the liquid and vapor in equilibrium at some fixed
pressure. The equilibrium line can be obtained from the Equation (5.4) once the
relative volatility is known.
y = 0.671x3 - 1.883x2 + 2.211x + 0.003

R² = 0.999
Mole fraction of benzene in vapour, y
Equilibrium curve
45o diagonal line
Mole fraction of benzene in liquid , x
Figure 5.3: Equilibrium diagram for a benzene-toluene mixture at 1 atmosphere
This particular VLE plot (Figure 5.3) shows a binary ideal mixture that has a
uniform vapor-liquid equilibrium that is relatively easy to separate. On the other
hand, the VLE plots shown in Figure 5.4 represented for non-ideal systems.
These non-ideal VLE systems will present more difficult separation.

Figure 5.4: Vapor-liquid equilibrium curve for non-ideal systems
The most challenging VLE curves are generated by azeotropic systems. An

azeotrope is a liquid mixture which when vaporized, produces the same
composition as the liquid. Two types of azeotropes are known: minimum-boiling
and maximum-boiling (less common). Ethanol-water system (at 1 atm, 89.4 mole
%, 78.2 oC; Carbon-disulfide - acetone (61.0 mole% CS2, 39.25 oC, 1 atm) and
Benzene - water (29.6 mole% H2O, 69.25 oC, 1 atm) are minimum-boiling
azeotropes. Hydrochloric acid - water (11.1 mole% HCl, 110 oC, 1 atm); Acetone
- chloroform (65.5 mole% chloroform, 64.5 oC, 1 atm) are the examples of
Maximum-boiling azeotropes. The Figure 5.5 shows two different azeotropic
systems, one with a minimum boiling point (Figure 5.5a) and one with a
maximum boiling point (Figure 5.5b). The points of intersections of the
equilibrium curves with the diagonal lines are called azeotropic points. An
azeotrope cannot be separated by conventional distillation. However, vacuum
distillation may be used as the lower pressures which can shift the azeotropic
point.

Figure 5.5: VLE curves for azeotropic systems: (a) for maximum boiling point, (b) for minimum
boiling point
Although most distillations are carried out at atmospheric or near atmospheric

pressure, it is not uncommon to distill at other pressures. High pressure
distillation (typically 3 - 20 atm) usually occurs in thermally integrated processes.
In those cases the equilibrium curve becomes narrower at higher pressures as

shown in Figure 5.6. Separability becomes less at higher pressures.
Figure 5.6: Variation of equilibrium curve with pressure

LECTURE NO. 2
5.2. Distillation columns and their process calculations

There are many types of distillation columns each of which is designed to
perform specific types of separations. One way of classifying distillation column
type is to look at how they are operated. Based on operation, they are of two
types: batch or differential and continuous columns.
5.2.1. Batch or differential distillation columns and their process

calculation
In batch operation, the feed is introduced batch-wise to the column. That is, the
column is charged with a 'batch' and then the distillation process is carried out.
When the desired task is achieved, a next batch of feed is introduced. Consider a
binary mixture of components A (more volatile) and B (less volatile). The system
consists of a batch of liquid (fixed quantity) inside a kettle (or still) fitted with
heating element and a condenser to condense the vapor produced as shown in
Figure 5.7. The condensed vapor is known as the distillate. The distillate is
collected in a condensate receiver. The liquid remaining in the still is known as
the residual. The process is unsteady state. The concentration changes can be
analyzed using the phase diagram, and detailed mathematical calculations
carried out using the Rayleigh Equation. Readers are suggested to follow chapter
9 (Page 369) of text book “Mass-transfer operations” by R. E Treybal, third
edition. As the process is unsteady state, the derivation is based on a differential
approach to changes in concentration with time. Let L1 = initial moles of liquid

originally in still, L2 = final moles of liquid remained in still, x1 = initial liquid

composition in still (mole fraction of A), x2 = final liquid composition in still (mole
fraction A). At any time t, the amount of liquid in the still is L, with mole fraction of
A in the liquid being x. After a small differential time (t + dt), a small amount of
vapor dL is produced, and the composition of A in the vapor is y (mole fraction).
The vapor is assumed to be in equilibrium with the residue liquid. The amount of
liquid in the still is thus reduced from L to (L - dL), while the liquid composition
changed from x to (x - dx). Then the material balance on A can be written as:
Figure 5.7: Simple batch or differential distillation process
Initial amount in still = Amount left in still + Amount vaporized
xL  ( x  dx)( L  dL)  ydL (5.6)

or
xL  xL  xdL  Ldx  dxdL  ydL (5.7)
Neglecting the term (dx)(dL), the Equation (5.7) may be written as:
Ldx  ydL  xdL (5.8)
Re-arranging and Integrating from L1 to L2, and from x1 to x2, one can obtain the
following Equation which is called Rayleigh Equation:
L  x1 1
ln  1    dx (5.9)
x2 ( y  x )
 L2 
The integration of Equation (5.9) can be obtained graphically from the equilibrium
curve, by plotting 1/(y-x) versus x.
Example problem 5.1

A mixture of 40 mole % isopropanol in water is to be batch-distilled at 1 atm until
70 mole % of the charge has been vaporized. Calculate the composition of the
liquid residue remaining in the still pot, and the average composition of the
collected distillate. VLE data for this system, in mole fraction of isopropanol, at 1
atm are (Seader and Henley, 1998):
Temp.(K) 366 357 355.1 354.3 353.6 353.2 353.3 354.5
y 0.220 0.462 0.524 0.569 0.593 0.682 0.742 0.916
x 0.012 0.084 0.198 0.350 0.453 0.679 0.769 0.944
Solution 5.1:
Calculate 1/(y-x)
As per Rayleigh Equation

L  1
ln  1   
x1
dx
2 ( y  x)
 L2  x
x1= 0.4
Feed L1 = 100
Distillate (D) = 70
Liquid residue as L2 = L1-D = 30
L  1
Find x2 by equating ln  1   
x1
dx
2 ( y  x)
 L2  x
The value of x2 = 0.067

L1 x1  L2 x2
yD   0.543
D

LECTURE NO. 3
5.2.2. Continuous distillation columns

In contrast, continuous columns process a continuous feed stream. No
interruptions occur unless there is a problem with the column or surrounding
process units. They are capable of handling high throughputs. Continuous
column is the more common of the two types.
Types of Continuous Columns

Continuous columns can be further classified according to the nature of the feed
that they are processing:
Binary distillation column: feed contains only two components
Multi-component distillation column: feed contains more than two
components the number of product streams they have
Multi-product distillation column: column has more than two product streams
where the extra feed exits when it is used to help with the separation,
Extractive distillation: where the extra feed appears in the bottom product
stream
Azeotropic distillation: where the extra feed appears at the top product stream
the type of column internals.
Tray distillation column: where trays of various designs are used to hold up the
liquid to provide better contact between vapor and liquid, hence better
separation. The details of the tray column are given in Module 4.
Packed distillation column: where instead of trays, 'packings' are used to
enhance contact between vapor and liquid. The details of the packed column are
given in Module 4.

4.2.2.1. A single-stage continuous distillation (Flash distillation):
A single-stage continuous operation occurs where a liquid mixture is partially

vaporized. The vapor produced and the residual liquids are in equilibrium in the
process are separated and removed as shown in Figure 5.8. Consider a binary
mixture of A (more volatile component) and B (less volatile component). The feed
is preheated before entering the separator. As such, part of the feed may be
vaporized. The heated mixture then flows through a pressure-reducing valve to
the separator. In the separator, separation between the vapor and liquid takes
place. The amount of vaporization affects the concentration (distribution) of A in
vapor phase and liquid phase. The relationship between the scale of vaporization
and mole fraction of A in vapor and liquid (y and x) is known as the Operating
Line Equation. Define f as molal fraction of the feed that is vaporized and
withdrawn continuously as vapor. Therefore, for 1 mole of binary feed mixture,
(1- f) is the molal fraction of the feed that leaves continuously as liquid. Assume,
yD = mole fraction of A in vapor leaving, xB = mole fraction of A in liquid leaving,
xF = mole fraction of A in feed entering. Based on the definition for f, the greater
the heating is, the larger the value of f. If the feed is completely vaporized, then f
= 1.0 Thus, the value of f can varies from 0 (no vaporization) to 1 (total
vaporization). From material balance for the more volatile component (A) one can
write
1.x F  fy D  (1  f ) x B (5.10)
Or,
fy D  x F  (1  f ) x B (5.11)

Figure 5.8: Flash distillation process
The Equation (5.11) on rearranging becomes:
1 f  x
y D    x B  F (5.12)
 f  f
The fraction f depends on the enthalpy of the liquid feed, the enthalpies of the
vapor and liquid leaving the separator. For a given feed condition, and hence the
known value of f and xF, the Equation (5.12) is a straight line Equation with slope
- (1-f)/f and intercept xF/f as shown in Figure 5.9. It will intersect the equilibrium
line at the point (xB, yD). From this value, the composition of the vapor and liquid
leaving the separator can be obtained.

Figure 5.9: Graphical presentation of flash veporization
5.2.2.2. Multi-stage Continuous distillation-Binary system

A general schematic diagram of a multistage counter-current binary distillation
operation is shown in Figure 4.10. The operation consists of a column containing
the equivalent N number of theoretical stages arranged in a two-section cascade;
a condenser in which the overhead vapor leaving the top stage is condensed to
give a liquid distillate product and liquid reflux that is returned to the top stage; a
reboiler in which liquid from the bottom stage is vaporized to give a liquid bottom
products and the vapor boil off returned to the bottom stage; accumulator is a
horizontal (usually) pressure vessel whereby the condensed vapor is collected;
Heat exchanger where the hot bottoms stream is used to heat up the feed stream

before it enters the distillation column. The feed enters the column at feed stage
contains more volatile components (called light key, LK) and less volatile
components (called heavy key, HK). At the feed stage feed may be liquid, vapor
or mixture of liquid and vapor. The section above the feed where vapor is
washed with the reflux to remove or absorb the heavy key is called enriching or
rectifying section. The section below the feed stage where liquid is stripped of the
light key by the rising vapor is called stripping section.
Figure 5.10: Multi-stage binary distillation column

LECTURE NO. 4
5.2.2.3. Analysis of binary distillation in tray towers: McCabe-

Thiele Method
McCabe and Thiele (1925) developed a graphical method to determine the
theoretical number of stages required to effect the separation of a binary mixture
(McCabe and Smith, 1976). This method uses the equilibrium curve diagram to
determine the number of theoretical stages (trays) required to achieve a desired
degree of separation. It assumes constant molar overflow and this implies that: (i)
molal heats of vaporization of the components are roughly the same; (ii) heat
effects are negligible. The information required for the systematic calculation are
the VLE data, feed condition (temperature, composition), distillate and bottom
compositions; and the reflux ratio, which is defined as the ratio of reflux liquid
over the distillate product. For example, a column is to be designed for the
separation of a binary mixture as shown in Figure 5.11.

Figure 5.11: Schematic of column for separation of binary mixture
The feed has a concentration of xF (mole fraction) of the more volatile

component, and a distillate having a concentration of xD of the more volatile
component and a bottoms having a concentration of xB is desired. In its essence,
the method involves the plotting on the equilibrium diagram three straight lines:
the rectifying section operating line (ROL), the feed line (also known as the q-
line) and the stripping section operating line (SOL). An important parameter in
the analysis of continuous distillation is the Reflux Ratio, defined as the quantity
of liquid returned to the distillation column over the quantity of liquid withdrawn as
product from the column, i.e. R = L / D. The reflux ratio R is important because
the concentration of the more volatile component in the distillate (in mole fraction
xD) can be changed by changing the value of R. The steps to be followed to
determine the number of theoretical stages by McCabe-Thiele Method:
 Determination of the Rectifying section operating line (ROL).

 Determination the feed condition (q).
 Determination of the feed section operating line (q-line).
 Determination of required reflux ratio (R).
 Determination of the stripping section operating line (SOL).

 Determination of number of theoretical stage.
Determination of the Rectifying section operating line (ROL)
Consider the rectifying section as shown in the Figure 5.12. Material balance can
be written around the envelope shown in Figure 5.12:
Overall or total balance:
Vn1  Ln  D (5.13)
Component balance for more volatile component:
Vn1 y n1  Ln xn  DxD (5.14)
From Equations (4.13) and (4.14), it can be written as
( Ln  D) y n1  Ln xn  DxD (5.15)
Consider the constant molal flow in the column, and then one can write: L1 = L2 =
.......... Ln-1 = Ln = Ln+1 = L = constant and V1 = V2 = .......... Vn-1 = Vn = Vn+1 = V =
constant. Thus, the Equation (5.15) becomes:
( L  D) y n1  Lxn  DxD (5.16)

Figure 5.12: Outline graph of rectifying section
After rearranging, one gets from Equation (5.16) as:
 L   D 
y n 1    xn    xD (5.17)
L D L D
Introducing reflux ratio defined as: R = L/D, the Equation (5.17) can be expressed
as:
 R   1 
y n 1    xn    xD (5.18)
 R 1  R  1
The Equation (5.18) is the rectifying section operating line (ROL) Equation having
slope R/(R+1) and intercept, xD/(R+1) as shown in Figure 5.13. If xn = xD, then
yn+1 = xD, the operating line passed through the point (xD, xD) on the 45o diagonal
line. When the reflux ratio R changed, the ROL will change. Generally the

rectifying operating line is expressed without subscript of n or n+1. Without

subscript the ROL is expresses as:
 R   1 
y x    xD (5.19)
 R  1  R  1
Figure 5.13: Representation of the rectifying operating line
Determination the feed condition (q):
The feed enters the distillation column may consists of liquid, vapor or a mixture
of both. Some portions of the feed go as the liquid and vapor stream to the
rectifying and stripping sections. The moles of liquid flow in the stripping section
that result from the introduction of each mole of feed, denoted as ‘q’. The
limitations of the q-value as per feed conditions are shown in Table 5.3.

Table 5.3: Limitations of q-value as per feed conditions
Feed condition Limit of q-value

cold feed (below bubble point) q>1
feed at bubble point (saturated liquid) q=1
feed as partially vaporized 0<q<1
feed at dew point (saturated vapor) q=0
feed as superheated vapor q<0
feed is a mixture of liquid and vapor q is the fraction of the feed that is liquid
Calculation of q-value
When feed is partially vaporized:
Other than saturated liquid (q = 1) and saturated vapor (q = 0), the feed condition
is uncertain. In that case one must calculate the value of q. The q-value can be
obtained from enthalpy balance around the feed plate. By enthalpy balance one
can obtain the q-value from the following form of Equation:
HV  H F
q (5.20)
HV  H L
where HF, HV and HL are enthalpies of feed, vapor and liquid respectively which
can be obtained from enthalpy-concentration diagram for the mixture.
When feed is cold liquid or superheated vapor:
q can be alternatively defined as the heat required to convert 1 mole of feed from
its entering condition to a saturated vapor; divided by the molal latent heat of
vaporization. Based on this definition, one can calculate the q-value from the
following Equations for the case whereby q > 1 (cold liquid feed) and q < 0
(superheated vapor feed) as:

For cold liquid feed:
C p , L (Tbp  TF )  
q (5.21)

For superheated vapor feed:
C p ,V (Tdp  TF )
q (5.22)

where Tbp is the bubble point,  is the latent heat of vaporization and Tdp is the
dew point of the feed respectively.
Determination of the feed section operating line (q-line):
Consider the section of the distillation column (as shown in Figure 5.11) at the
tray (called feed tray) where the feed is introduced. In the feed tray the feed is
introduced at F moles/hr with liquid of q fraction of feed and vapor of (1-f) fraction
of feed as shown in Figure 5.14. Overall material balance around the feed tray:
L  L  qF and V  V   (1  q) F (5.23)
Component balances for the more volatile component in the rectifying and
stripping sections are:
For rectifying section:
Vy  Lx  DxD (5.24)
For stripping section:
V y  Lx  Bx B (5.25)
At the feed point where the two operating lines (Equations (5.24) and (5.25)
intersect can be written as:

(V  V ) y  ( L  L) x  DxD  Bx B (5.26)
Figure 5.14: Feed tray with fraction of liquid and vapor of feed
From component balance around the entire column, it can be written as
Fx F  DxD  Bx B (5.27)
Substituting L-L’ and V-V’ from Equations (5.23) and (5.24) into Equation (5.26)
and with Equation (5.27) one can get the q-line Equation after rearranging as:
 q   1 
y    x    x F (5.28)
1 q  1 q 
For a given feed condition, xF and q are fixed, therefore the q-line is a straight
line with slope -q / (1-q) and intercept xF/(1-q). If x = xF , then from Equation

(5.28) y = xF. At this condition the q-line passes through the point (xF, xF) on the
45o diagonal. Different values of q will result in different slope of the q-line.
Different q-lines for different feed conditions are shown in Figure 5.15.
Figure 5.15: Different q-lines for different feed conditions

LECTURE NO. 5
Determination of the stripping section operating line (SOL):
The stripping section operating line (SOL) can be obtained from the ROL and q-
line without doing any material balance. The SOL can be drawn by connecting
point xB on the diagonal to the point of intersection between the ROL and q-line.
The SOL will change if q-line is changed at fixed ROL. The change of SOL with
different q-lines for a given ROL at constant R and xD is shown in Figure 5.16.
Figure 5.16: Stripping section operating line with different q-lines

The stripping section operating line can be derived from the material balance
around the stripping section of the distillation column. The stripping section of a
distillation column is shown in Figure 5.17. The reboiled vapor is in equilibrium
with bottoms liquid which is leaving the column.
Figure 5.17: Schematic of the stripping section
Consider the constant molal overflow in the column. Thus L'm = L'm+1 = .... = L' =
constant and V'm = V'm+1 = ..... = V' = constant.
Overall material balance gives
L  V   B (5.29)
More volatile component balance gives:
Lxm  V y m1  Bx B (5.30)

Substituting and re-arranging the Equation (5.30) yields
 L  B
y m1    xm    x B (5.31)
V   V  
Dropping the subscripts "m+1" and "m" it becomes:
 L   B 
y    x    xB (5.32)
V   V  
Substituting V' = L' – B from Equation (5.29), the Equation (5.32) can be written
as:
 L   B 
y x    xB (5.33)
 L  B   L  B 
The Equation (5.33) is called the stripping operating line (SOL) which is a straight
line with slope ( L' / L' - B) and intercept ( BxB / L' - B ). When x = xB , y = xB, the
SOL passes through (xB, xB ) on the 45o diagonal line.
Determination of number of theoretical stage
Suppose a column is to be designed for the separation of a binary mixture where

the feed has a concentration of xF (mole fraction) of the more volatile component
and a distillate having a concentration of xD of the more volatile component
whereas the bottoms having a desired concentration of xB. Once the three lines
(ROL, SOL and q-line) are drawn, the number of theoretical stages required for a
given separation is then the number of triangles that can be drawn between
these operating lines and the equilibrium curve. The last triangle on the diagram
represents the reboiler. A typical representation is given in Figure 5.18.

Figure 5.18: A typical representation of identifying number of theoretical stages
Reflux Ratio, R
The separation efficiency by distillation depends on the reflux ratio. For a given
separation (i.e. constant xD and xB) from a given feed condition (xF and q), higher
reflux ratio (R) results in lesser number of required theoretical trays (N) and vice
versa. So there is an inverse relationship between the reflux ratio and the
number of theoretical stages. At a specified distillate concentration, xD, when R
changes, the slope and intercept of the ROL changes (Equation (5.19)). From the
Equation (5.19), when R increases (with xD constant), the slope of ROL becomes
steeper, i.e. (R/R+1) and the intercept (xD/R+1) decreases. The ROL therefore
rotates around the point (xD, xD). The reflux ratio may be any value between a
minimum value and an infinite value. The limit is the minimum reflux ratio (result
in infinite stages) and the total reflux or infinite reflux ratio (result in minimum
stages). With xD constant, as R decreases, the slope (R/R+1) of ROL (Equation

(5.19)) decreases, while its intercept (xD/R+1) increases and rotates upwards
around (xD, xD) as shown in Figure 5.19. The ROL moves closer to the
equilibrium curve as R decreases until point Q is reached. Point Q is the point of
intersection between the q-line and the equilibrium curve.
Equilibrium line
Q
ROL
xD/(Rm+1)
y
q-line
xD/(R+1)
SOL
xF xD
xB
Figure 5.19: Representation of minimum reflux rtio
At this point of intersection the driving force for mass transfer is zero. This is also
called as Pinch Point. At this point separation is not possible. The R cannot be
reduced beyond this point. The value of R at this point is known as the minimum
reflux ratio and is denoted by Rmin. For non-ideal mixture it is quite common to
exhibit inflections in their equilibrium curves as shown in Figure 5.20 (a, b). In
those cases, the operating lines where it becomes tangent to the equilibrium
curve (called tangent pinch) is the condition for minimum reflux. The ROL cannot

move beyond point P, e.g. to point K. The condition for zero driving force first
occurs at point P, before point K which is the intersection point between the q-
line and equilibrium curve. Similarly it is the condition SOL also. At the total reflux
ratio, the ROL and SOL coincide with the 45 degree diagonal line. At this
condition, total number of triangles formed with the equilibrium curve is equal
minimum number of theoretical stages. The reflux ratio will be infinite.
Figure 5.20 (a): Representation of minimum reflux ratio for non-ideal mixture

Figure 5.20 (b): Representation of minimum reflux ratio for non-ideal mixture
Tray Efficiency
For the analysis of theoretical stage required for the distillation, it is assumed that
the the vapor leaving each tray is in equilibrium with the liquid leaving the same
tray and the trays are operating at 100% efficiency. In practice, the trays are not
perfect. There are deviations from ideal conditions. The equilibrium with
temperature is sometimes reasonable for exothermic chemical reaction but the
equilibrium with respect to mass transfer is not often valid. The deviation from the
ideal condition is due to: (1) Insufficient time of contact
(2) Insufficient degree of mixing. To achieve the same degree of desired
separation, more trays will have to be added to compensate for the lack of

perfect separability. The concept of tray efficiency may be used to adjust the the
actual number of trays required.
Overall Efficiency
The overall tray efficiency, EO is defined as:
No. of theoratica l trays

Eo  (5.34)
No. of actual trays
It is applied for the whole column. Every tray is assumed to have the same
efficiency. The overall efficiency depends on the (i) geometry and design of the
contacting trays, (ii) flow rates and flow paths of vapor and liquid streams, (iii)
Compositions and properties of vapor and liquid streams (Treybal, 1981; Seader
and Henley, 1998). The overall efficiency can be calculated from the following
correlations:
The Drickamer-Bradford empirical correlation:
Eo  13.3  66.8 log(  ) (5.35)
The corrrelation is valid for hydrocarbon mixtures in the range of 342 K < T <
488.5 K, 1 atm < P < 25 atm and 0.066 < µ < 0.355 cP
The O'Connell correlation:
Eo  50.3( ) 0.226 (5.36)
Murphree Efficiency
The efficiency of the tray can also be calculated based on semi-theoretical

models which can be interpreted by the Murphree Tray Efficiency E M. In this case
it is assumed that the vapor and liquid between trays are well-mixed and have
uniform composition. It is defined for each tray according to the separation

achieved on each tray based on either the liquid phase or the vapor phase. For a
given component, it can be expressed as:
Based on vapor phase:
y n  y n 1
E MV  (5.37)
y n*  y n 1
Based on liquid phase:
x n  x n 1
E ML  (5.38)
x n*  x n 1
Example problem 5.2:

A liquid mixture of benzene toluene is being distilled in a fractionating column at
101.3 k Pa pressure. The feed of 100 kmole/h is liquid and it contains 45 mole%
benzene (A) and 55 mole% toluene (B) and enters at 327.6 K. A distillate
containing 95 mole% benzene and 5 mole% toluene and a bottoms containing 10
mole% benzene and 90 mole% toluene are to be obtained. The amount of liquid
is fed back to the column at the top is 4 times the distillate product. The average
heat capacity of the feed is 159 KJ/kg mole. K and the average latent heat 32099
kJ/kg moles.
Calculate
i. The kg moles per hour distillate, kg mole per hour bottoms
ii. No. of theoretical stages at the operating reflux.
iii. The minimum no. of theoretical stages required at total reflux
iv. If the actual no. of stage is 10, what is the overall efficiency increased
at operating condition compared to the condition of total reflux?
The equilibrium data:
Temp.(K) 353.3 358.2 363.2 366.7 373.2 378.2 383.8
xA (mole fraction) 1.000 0.780 0.580 0.450 0.258 0.13 0
yA(mole fraction) 1.000 0.900 0.777 0.657 0.456 0.261 0

Solution 5.2:
F=D+B
100 = D + B
F xF = D xD + B xB
Therefore, D = 41.2 kg mole/h, B = 58.8 kg mole/h
y = [R/(R+1)] x + xD/(R+1) = 0.8 x +0.190
q = 1+ cpL (TB-TF)/Latent heat of vaporization
TB = 366.7 K from boiling point of feed, TF = 327.6 K ( inlet feed temp)
Therefore q = 1.195
Slope of q line = 6.12
From the graph (Figure E1), Total no of theoretical stages is 8 at operating reflux
(Red color)
From the graph (Figure E1), Total no of theoretical stages is 6 at total reflux
(Black color)
Overall efficiency at operating conditions: Eo (Operating) = No of ideal stage/ No
of actual trays = 7.9/10 = 0.79
Overall efficiency at total reflux conditions: Eo (total relux) = No of ideal stage/ No
of actual trays = 5.9/10=0.59
Overall efficiency increased: 0.79-0.59 = 0.20

Figure E1: Graph representing the example problem 5.2.

LECTURE NO. 6
5.2.2.4. Analysis of binary distillation by Ponchon-Savarit

Method
Background Principle:
The method is concerned with the graphical analysis of calculating the theoretical
stages by enthalpy balance required for desired separation by distillation process
(Hines and Maddox, 1984). In this method, the enthalpy balances are
incorporated as an integral part of the calculation however it is not considered in
the analysis separation by distillation process by McCabe-Thele method. This
procedure combines the material balance calculations with enthalpy balance
calculations. This method also provides the information on the condenser and
reboiler duties. The overall material balance for the distillation column is as
shown in Figure 5.11:
F  DB (5.39)
For any component the material balance around the column can be written as:
Fx F  DxD  Bx B (5.40)
Overall enthalpy balance for the column yields
FH F  QR  DH D  BH B  QC (5.41)

where H is the enthalpy of the liquid stream, energy/mol, Q R is the heat input to
the reboiler, J/s and QC is the heat removed from condenser, J/s. The heat
balance on rearrangement of Equation (5.41) gives
 Q   Q 
FH F  D H D  C   B H B  R  (5.42)
 D   R 
The Equation (5.41) is rearranged as Equation (5.42) for the convenience to plot
it on the enthalpy-concentration diagram. The points represented by the feed,
distillate and bottom streams can be plotted on the enthalpy-concentration
diagram as shown in Figure 5.21.
Figure 5.21: Representation of feed, distillate and bottom streams on the enthalpy-concentration

Substituting the Equation (5.39) into the Equation (5.40) and (5.42) we get
( D  B) x F  DxD  Bx B (5.43)
 Q   Q 
( D  B) H F  D H D  C   B H B  R  (5.44)
 D   R 
From the Equations (5.43) and (5.44) one can write
QR
H F  (H B  )
D xF  xB R
  (5.45)
B xD  xF Q
(H D  C )  H F
D
The Equation (5.45) can also be written on rearrangement as:
xF  xB xD  xF
 (5.46)
Q Q
H F  (H B  R ) (H D  C )  H F
R D
Comparing the Equation (5.46) with the points plotted on Figure (5.21), it is found
that the left hand side of the Equation (5.46) represents the slope of the straight
line between the points (xB, HB-QR/B) and (xF, HF). The right hand side of the
Equation (5.46) represents the slope of the straight line passing through the
points (xF, HF) and (xD, HD-QC/D). From this it can be said that all three points are
on a same straight line. The amount of distillate as per Equation (5.45) is
______
proportional to the horizontal distance x F x B and the amount of bottoms is
______
proportional to the horizontal distance x D x F , then from the overall material
balance it can be interpreted that the amount of feed is proportional to the
______
horizontal distance x D x B . This leads to the inverse lever rule. Based on this
principle, the analysis of distillation column is called Enthalpy-composition
analysis or Ponchon-Savarit analysis of distillation column.

Analysis of tray column
Consider the theoretical stage shown in Figure 5.12 and the principle by which it
operates is described in the enthalpy-composition diagram as shown in Figure
5.22.
Figure 5.22: Operation principle of stage-wise binary distillation on enthalpy-concentration diagram
The vapor entering to the tray (n), Vn+1 is a saturated vapor of composition yn+1
wheras the liquid entering to the tray is Ln-1 of composition xn-1. The point P in the
Figure 5.22 represents the total flow to the tray. The point P lies in a straight line
______
joining xn-1 and yn+1. The distance y n1 P is proportional to the quantity Ln-1 and
______
the distance x n1 P is proportional to the quantity Vn+1 as per lever rule. So

______
Ln 1 y n 1 P
 ______ (5.47)
Vn 1
x n 1 P
The sum of the liquid and vapor leaving the plate must equal the total flow to the
plate. So the tie line must pass through the point P which represents the addition
point of the vapor Vn and liquid Ln leaving the tray. The intersection of the tie line
with the enthalpy-composition curves will represent the compositions of these
streams.
Around the tray n, the material balances can be written as:
Vn1  Ln  Vn  Ln1 (5.48)
Vn1 y n1  Ln xn  Vn y n  Ln1 xn1 (5.49)
The enthalpy balance around the tray gives
Vn1 H V ,n1  Ln H L,n  Vn H V ,n  Ln1 H L,n1 (5.50)
Where HV is the enthalpy of the vapor and HL is the enthalpy of the liquid. From
the material and enthalpy balance Equations above mean that the stream that
added to Vn+1 to generate Ln is same to the stream that must be added to V n to
generate Ln-1. In the Figure 5.22, Δ represents the difference point above and
below the tray. This point of a common difference point can be extended to a
section of a column that contains any number of theoretical trays.

NPTEL – Chemical Engineering – Chemical Engineering Design - II
Module # 7
PROCESS DESIGN OF MASS TRANSFER COLUMN: DESIGN OF
DISTILLATION AND ABSORPTION COLUMN
1. Introduction
2. Plate contractors
7.1. Definition of tray areas
7.2. Plate types
7.2.1. Bubble cap plates
7.2.2. Valve plates
7.2.3. Sieve plate
7.2.4. Selection of tray type
7.3. Effect of vapor flow conditions on tray design
7.3.1. Flooding consideration
7.3.2. Sieve tray weeping
7.3.3. Liquid entrainment
7.4. Tray hydraulic parameters
7.5. Column sizing approximation
7.6. Provisional plate design
7.6.1. Column diameter
7.6.2. Hole diameter, hole pitch and plate thickness
7.6.3. Weir height and weir length
7.6.4. Calming zones
3. Stepwise design tray procedure
4. Design problem
References

Lecture 1: Introduction
1. Introduction
The typical gas-liquid contacting operations include distillation, absorption, stripping,
leaching and humidification. Distillation and absorption are two most widely used mass
transfer processes in chemical industries. Design of plate column for absorption and
distillation involves many common steps of calculation such as determination of number
of theoretical plates, column diameter, plate hydraulic design, etc.In absorption process, a
soluble component is absorbed in a liquid (called solvent) from a gaseous mixture. The
gas and liquid streams leaving the tray are in equilibrium under the ideal condition. The
separation in distillation is based on the relative volatility of the components. Additional
vapor phase is generated by the vaporization of more volatile components (called
stripping) and by condensation of relatively less volatile components(called absorption)
adds to the liquid phase.
Selection of column type: Plate or Packed: Packed towers (columns) are also used as
the contacting devices for gas absorption, liquid-liquid extraction and distillation. The
gaseous mixture is allowed to contact continuously with the liquid counter-currently in a
packed column. The liquid flows downward over the packing surface, and the gaseous
mixture flows upward through the space in the packing. The performance of the column
strongly depends on the arrangement of the packings to provide good liquid and gas
contact throughout the packed bed. The solute gas is absorbed by the fresh solvent
(liquid) entering at the top of the tower where the lean gas leaves system. The liquid
enriched with absorbed solute gas, leaves the column bottom through the exit port.
In a plate tower, the liquid and gas are contacted in stage-wise manner on the trays; while
gas-liquid contact is continuous in a packed column. There are always some uncertainly
to maintain good liquid distribution in a packed tower. For this reason, it is difficult to
accurately estimate the packed tower efficiency. The course content is limited to design
of plate column only and some typical criterions for the selection of column type are
discussed below.

Plate towers exhibit larger pressure drops and liquid holdup at higher gas flow
rate. While, packed towers are not appropriate for very low liquid flow rates.
Packed column is the preferred choice than a plate column to handle toxic and
flammable liquids due to lower liquid holdup to keep the unit as small as possible
for the sake of safety.
Plate columns are normally suitable for fouling liquids or laden with solids. They
are easier to clean and could handle substantial temperature variation during
operation.
Packed towers are more suitable for foaming and corrosive services.
It is easier to make the provision for the installation of internal cooling coils or
withdrawal of side streams from a plate column.
2. Plate contractors
Plate contractors/ towers are vertical cylindrical columns in which a vertical stack of
trays or plates are installed across the column height as shown in Figure 7.1. The liquid
enters at the top of the column and flows across the tray and then through a downcomer
(cross-flow mode) to the next tray below. The gas/vapor from the lower tray flows in the
upward direction through the opening/holes in the tray to form a gas-liquid dispersion. In
this way, the mass transfer between the phases (gas/vapor-liquid) takes place across the
tray and through the column in a stage-wise manner.

Figure 7.1. Schematic diagram of a plate contractor ([1] page 159).

7.1. Definition of tray areas

The definition of tray areas and its nomenclature illustrated in Figures 7.2 &7.3 are
followed throughout the design procedure.
Total tower cross-section area (𝑨𝑻 ): The empty tower inside cross-sectional area
without trays or downspouts.
Net area (𝑨𝑵 ) (also called free area):The total tower crosssectional area (𝐴 𝑇 )minus the
area at the top of the downcomer (𝐴𝐷𝑇 ). The net area symbolizes the smallest area
available for vapor flow in the inter-tray spacing.
Bubbling area or active area (𝑨𝑨 ): The total tower cross-sectional area minus sum of
the downcomer top area(𝐴𝐷𝑇 ) and downcomer seal area (𝐴𝐷𝐵 )and any other
nonperforated areas on the tray. The bubbling area represents the area available for vapor
flow just above the tray floor.
Hole area (𝑨𝒉 ): The total area of the perforations on the tray. The hole area is the
smallest area available for vapor/gas passage.
Figure 7.2.Schematic of a tray operating in the froth regime ([2] page 14-28).

Figure 7.3. Typical cross-flow plate (sieve) ([3] page 557).

7.2. Plate types
Gas and liquid flow across the tray can either be by cross-flow or counter-flow manner
(Figure 7.4). The cross-flow plates are most widely practiced and the three main types of
cross flow plates are: bubble cap, valve and sieve trays with downcomer.
Figure 7.4. Classification of plate types based on flow mode- side view shown: (a) Cross-flow plate, (b): Counterflow plate.

7.2.1. Bubble cap plates

An enhanced gas-liquid contact can be achieved having bubble caps on the tray at very
low liquid flow rates. A bubble cap consists of a riser (also called chimney) fixed to the
tray through a hole and a cap is mounted over the riser (Figure 7.5). The gas flows up
through the riser, directed downward by the cap through the annular space between riser
and cap. Finally, the gas is dispersed into the liquid. A number of slots in the lower part
of the cap help in gas bubble dispersion. Un-slotted types of cap designs are also common
in application. Bubble caps are especially suitable for higher turndown ratio. Turndown
ratio is the ratio of maximum operating vapor rate to the minimum allowable vapor rate,
below which weeping starts.
Figure 7.5. Bubble caps ([1] page 166).

7.2.2. Valve plates

Valve trays (or floating cap plate) are the modified design of sieve trays where relatively
large plate perforations are covered by movable caps/valves (Figure 7.6). Valves cover
may be round or rectangular. The very common hole diameter is 40 mm but upto 150 mm
are also used. The valve lifts up as the vapor flow rate increases and the valve sits over
the perforation at lower flow rate, thus stops the liquid from weeping. Valve trays provide
good vapor-liquid contact at low flow rates (high turndown ratio).
Figure 7.6. Valve tray ([4] page 14-25).
7.2.3. Sieve plate

The sieve tray (also known as perforated plate) is a flat perforated metal sheet (Figure
7.7). The hole diameter from 1.5 to 25 mm are very commonly used. The sieve tray
layout is a typical square or equilateral triangular pitch holes. The gas/vapor flows
upward through the perforation and disperses into the flowing liquid over the plate. There
is no liquid seal in case of trays without downcomer and the liquid weeps (called
weeping) through the holes at low flow rates, reducing the efficiency of plate. For this
reason, sieve tray has the lowest turndown ratio. Sieve tray construction is simple and
relatively cheap.
Figure 7.7. Sieve tray ([4] page 14-25).

7.2.4. Selection of tray type

The comparative performances of three common types of trays are summarized in Table
7.1.
The capacity, efficiency, pressure drop and entrainment of sieve and valve trays are
almost same.
Bubble cap trays have lower capacity and efficiency and but higher pressure drop and
entrainment compared to valve and sieve trays. The turndown ratio comes in the order of:
bubble cap>valve>sieve. However, valve trays have the best turndown ratio in case of
refinery applications. Sieve trays are the least expensive and suitable for almost all
applications. Valve trays can be considered where higher turndown ratio is needed.
Bubble cap trays should be used at very low liquid flow rate which is not achievable
using sieve trays.
Table 7.1: Comparison of three types of cross-flow trays ([5] page 266).
Tray Capacity Efficiency Pressure Entrainment Turndown Cost
type drop ratio
Bubble Medium Medium High ~3 times than Excellent 100-200 %
cap high high sieve tray more than sieve
tray
Valve High to High Medium to Medium 4 to 10.1 20-50% more
very high high than sieve tray
Sieve High High Medium Medium 2.1 Cheapest of all
types

Lecture 2: Effect of Vapor Flow Conditions on Tray

Design
7.3. Effect of vapor flow conditions on tray design
7.3.1. Flooding consideration
Excessive liquid buildup inside the column leads to column flooding condition. The
nature of flooding depends on the column operating pressure and the liquid to vapor flow
ratio. It may be downcomer backup, spray entrainment or froth entrainment type
floodings.Higher tray pressure drop due to excessive vapor flow rates holds up the liquid
in the downcomer, increases the liquid level on the plate and leads to downcomer
flooding situation. The column flooding conditions sets the upper limit of vapor velocity
for steady operation.
Gas velocity through the net area at flooding conditioncan be estimated using Fair’s
correlation ([4], page 14-26):
𝜍 0.2 𝜌 𝑙 −𝜌 𝑣 0.5
𝑈𝑛𝑓 = 𝐶𝑠𝑏𝑓 [m/s]
20 𝜌𝑣
(7.1)
𝜌𝑣 = vapor density, kg/m3
𝜌𝑙 = liquid density, kg/m3
𝜍 = liquid surface tension, mN/m (dyn/cm)
𝐶𝑠𝑏𝑓 = capacity parameter (m/s) can be calculated([4] page 14-27) in terms of plate
𝐿 𝜌 𝑣 0.5
spacing and flow parameter 𝐹𝐿𝐺 = 𝑉 𝜌𝑙
(7.2)
𝐿 =liquid flow rate, kg/s
𝑉 =vpor flow rate, kg/s
The design gas velocities (𝑈𝑛 ) is generally 80-85% of 𝑈𝑛𝑓 for non-foaming liquids and
75% or less for foaming liquids subject to acceptable entrainment and plate pressure
drop.

7.3.2. Sieve tray weeping

Weeping occurs at low vapor/gas flow rates. The upward vapor flow through the plate
perforationsprevents the liquid from leaking through the tray perforation. At low vapor
flow rates, liquid start to leak/rain through the perforation (called weeping). When none
of the liquid reaches the downcomer at extreme weeping condition at very low vapor
flow rate, it is called dumping. The weeping tendency increases with increasing fractional
hole area and liquid flow rates.
The vapor velocity at the weep point (where liquid leakage through holes starts) is the
minimum value for stable operation. For a chosen hole area,the minimum operating
vapor flow velocity (𝑈𝑚𝑖𝑛 ,𝑜𝑝 ) at minimum flow rate for stable operation should be above
weep point vapor velocity.
The minimum vapor velocity (𝑈min ) at the weep point ([3] page 569):
𝐾2 −0.9(25.4−𝑑 ℎ )
𝑈min = (7.3)
𝜌 𝑣 1/2
Where, 𝑑ℎ = hole diameter, mm,

𝜌𝑣 = vapor density, kg/m3 (maximum value of vapor density)
𝐾𝟐 = constant (𝐾2 ) of weep-point correlation depends on the depth of clear liquid
(weir crest + weir height) on the plate ([3] page 571).
Weir crest (ℎ𝑤𝑐 ) can be determined using the Francis’ weir correlation ([3] page 571):
2
𝐿𝑤𝑐 3
ℎ𝑤𝑐 = 750 [mm]
𝐿𝑊 𝜌 𝑙
(7.4)
𝐿𝑊𝐶 =weir length, m
𝐿𝑊 =liquid flow rate over the crest, kg/s
𝜌𝑙 = liquid density, kg/m3
minimum vapor flow rate
Actual operating minimum vapor velocity: 𝑈𝑚𝑖𝑛 ,𝑜𝑝 = [m/s] (7.5)
hole area
To avoid weeping: 𝑈𝑚𝑖𝑛 ,𝑜𝑝 > 𝑈min .

7.3.3. Liquid entrainment

Entrainment is the phenomena in which liquid droplets are carried by vapor/gas to the
tray above. Therefore, the less volatile liquid components from bottom tray are mixed
with liquid having relatively more volatile materials on the overhead tray. It counteracts
the desired mass transfer operation and the plate efficiency decreases. Entrainment
kg
increases with vapor velocity.The fractional entrainment (𝛹 = kg gross ) can
liquid flow
𝐿 𝜌 𝑣 0.5
predicted using Fair’s correlation in terms of the flow parameter [𝐹𝐿𝐺 = 𝑉 ] and
𝜌𝑙
actual flooding velocity ([4] page 14-28).
Effect of 𝛹 on Murphree plate efficiency can be estimated using Colburn equation ([4]
page 14-29):
𝐸𝑚𝑣
𝐸𝑎 = Ψ𝐸 𝑚𝑣
1+
1−Ψ
(7.6)
𝐸𝑚𝑣 =Murphree vapor efficiency
E𝑎 =Corrected Murphree vapor efficiency for liquid entrainment
7.4. Tray hydraulic parameters
Total plate pressure drop
All gas pressure drops (ℎ𝑡 ) are expressed as heads of the clear liquid and ℎ𝑡 is given by:
ℎ𝑡 = ℎ𝑑 + ℎ𝑤𝑐 + ℎ𝑤 + ℎ𝑟
(7.7)
Where, ℎ𝑑 =dry plate pressure drop, mm
ℎ𝑤𝑐 =height of liquid over weir (weir crest), mm
ℎ𝑤 =weir height, mm
ℎ𝑟 =residual head, mm
Dry plate pressure drop (𝒉𝒅 ):
Dry plate pressure drop occurs due to friction within dry short holes.ℎ𝑑 can be calculated
using following expression derivedfor flow through orifices ([3] page 575).
𝑈𝑚𝑎𝑥 2𝜌
𝑣
ℎ𝑑 = 51 [mm]
𝐶0 𝜌𝑙
(7.8)

Maximum volumetric vapor flow rate

Maximum vapor velocity: 𝑈𝑚𝑎𝑥 = 𝐴𝐻
(7.9)
𝐴𝐻 plate thickness
The orifice coefficient, 𝐶0 can be determined in terms of and ([3] page
𝐴𝑃 hole diameter
576).
Residual gas pressure head (𝒉𝒓 ):
The residual pressure drop results mainly from the surface tension as the gas releases
from a perforation. The following simple equation can be used to estimate ℎ𝑟 with
reasonable accuracy ([3] page 575).
12.5×10 3
ℎ𝑟 = 𝜌𝑙
(9.10)
Downcomer backup (𝒉𝒃 ) and downcomer residence time:
The liquid level and froth in the downcomer should be well below the top of the outlet
weir on the tray above to avoid flooding ([3] page 576).
ℎ𝑏 = ℎ𝑤𝑐 + ℎ𝑤 + ℎ𝑡 + ℎ𝑑𝑐
(7.11)
𝐿𝑤𝑑 2
Head loss in downcomer: ℎ𝑑𝑐 = 166 𝜌 𝑙 𝐴𝑚
(7.12)
𝐿𝑤𝑑 = Downcomer liquid flow rate, kg/s
𝐴𝑚 =Smaller of clearance area under the downcomer apron (𝐴𝑎𝑝 ) and downcomer
area(𝐴𝐷 )
1
The average density of aerated liquid in the dowmncomer can be assumed as of the
2
clear liquid density.Therefore, half of the sum of the plate spacing and weir height should
be greater than thedowncomer backup.
𝟏
𝐩𝐥𝐚𝐭𝐞 𝐬𝐩𝐚𝐜𝐢𝐧𝐠 + 𝐰𝐞𝐢𝐫 𝐡𝐞𝐢𝐠𝐡𝐭 > 𝒉𝒅 (7.13)
𝟐

Downcomer residence time (𝑡𝑑𝑟𝑡 ) should be sufficient for the disengagement of liquid
and vapor in the downcomer to minimize entrained vapor. The value of 𝑡𝑑𝑟𝑡 >3 s is
suggested. Downcomer residence timeis given by ([3] page 578):
𝐴𝐷 ℎ 𝑏𝑐 𝜌 𝑙
𝑡𝑑𝑟𝑡 = [s]
𝐿𝑤𝑑
(7.14)
ℎ𝑏𝑐 = clear liquid back up, mm

Lecture 3: Plate Design

7.5. Column sizing approximation
The column sizing is a trial anderror calculationprocedure,starting with a tentative tray
layout. The calculation is then revised until anacceptable design is obtained subject to
satisfyingthetray pressure drop, weeping, flooding and liquid entrainment limits. The
column sizing is carried at the tray where the anticipated column loading is the highest
and lowest for each section. However, the vapor flow rates have the highest impact on
tower diameter. For an example, the sizing calculation is performed on the top tray for
the above feed section and on the bottom tray for below feed section, for a single feed
distillation column with one top and one bottom product. The tray spacing determines the
column height. Lower tray spacing is desirable to minimize construction cost by
checking against the column performance criteria. The suggested tray spacing (𝑇𝑡 ) with
column diameter is appended below ([1] page 162). The detailed column sizing
calculations are discussed in the solved example.
Tower diameter, m Tray spacing, mm
1 or less 500 (150 mm is minimum)
1-3 600
3-4 750
4-8 900
7.6. Provisional plate design
7.6.1. Column diameter
The column diameter is determined from the flooding correlation for a chosen plate
spacing. The superficial vapor/gas velocity (𝑈𝑛𝑓 ) at flooding through the net area relates
to liquid and vapor densities according to Fair’s correlation (refer to section7.3.1).𝐶𝑠𝑏𝑓 is
an empirical constant, depends on tray spacing and can be estimated against the flow
parameter (𝐹𝐿𝐺 ) based on mass flow rate of liquid (𝐿) and vapor (𝑉) ([3] page 567, [4]
page 14-27).
Typically, the design velocity (𝑈𝑛 ) through the net area is about 80 to 85% of 𝑈𝑛𝑓 for
non-foaming liquids and 75% or less for foaming liquid depending on allowable plate
pressure drop and entrainment. It is a common practice to have uniform tower diameter in
all sections of the column even though the vapor/gas and liquid loadings are expected to

be different to minimize the cost of construction. The uniformity in tower diametermay

require selecting different plate spacing in different sections of the tower.
7.6.2. Hole diameter,hole pitchand plate thickness
The plate hole diameters (𝑑ℎ ) from 3 to 12 mm are commonly used. The bigger sizes are
susceptible to weeping. The holes may be drilled or punched and the plate is fabricated
from stainless steel and other alloys than carbon steel. The centre to centre distance
between two adjacent holes is called hole pitch (𝐼𝑃 ). Perforations can be arranged in
square or equilateral triangular arrays with respect to the vapor/gas flow direction. The
normal range of 𝐼𝑃 is from 2.5 to 5 times of 𝑑ℎ ([1] page 168).
𝐴𝐻 𝑑ℎ 2
For triangular pitch: = 0.907
𝐴𝑃 𝐼𝑃
(7.15)
Plate thickness (𝑡𝑡 ) typically varies from 0.2 to 1.2 times of the hole diameter and should
be verified by checking the allowable plate pressure drop ([3] page 576).
7.6.3. Weir heightand weir length
The depth of liquid on the tray is maintained by installing a vertical flat plate, called weir.
Higher weir height (ℎ𝑤 ) increases the plate efficiency. But it increases plate pressure
drop, entrainment rate and weeping tendency. Weir heights from 40 to 90 mm are
common in applications for the columns operating above the atmospheric pressure. For
vacuum operation, ℎ𝑤 =6 to 12 mm are recommended. The weir length (𝐿𝑤 ) determines
the downcomer area. A weir length of 60 to 80% of tower diameter is normally used
with segmental downcomers. The dependency of 𝐿𝑤 on downcomer area is calculated
𝐴𝐷
against the percentage value of ([3] page 572).
𝐴𝐴
7.6.4. Calming zones

Two blank areas called calming zone, are provided between the inlet downcomer or inlet
weir and the perforation area, and also between the outlet weir and perforation area. Inlet
calming zone helps in reducing excessive weeping in this area because of high vertical
velocity of the entering liquid in the downward direction. Outlet calming zone allows
disengagement of vapor before the liquid enters the downcomer area. A calming zone
between 50 to 100mm is suggested.

3. Stepwise design tray procedure

Iterative tray design approach ([3] page 566) is listed below. The design is performed
separately both above feed plate (top section) and below feed plate (bottom section) for
single feed two product distillation column.
Step #1: Determine the number of theoretical plate and vapor and liquid flow-rates
separately both in top and bottom sections.
Step #2: Obtain the physical properties of the system
Step #3: Select a trial plate spacing
Step #4: Estimate the column diameter based on flooding considerations
Step #5: Decide the liquid flow arrangement (reverse, single-pass, or multiple-pass). A
guideline is provided in Figure 11.28 ([3] page 568).
Step #6: Make a provisional tray layout including downcomer area, active area,
perforated area, hole area and size, weir height, weir length
Step #7: Check the weeping rate, if not satisfactory go back to step #6 and reselect tray
layout
Step #8: Check the plate pressure drop, if too high return to step #6
Step #9: Check downcomer back-up, if too high go back to step #6 or #3
Step #10: Decide plate layout including calming zones and unperforated areas and check
hole pitch, if unsatisfactory return to step #6
Step #11: Recalculate the percentage of flooding based upon selected tower diameter
Step #12: Check for entrainment, if too high then return to step #4
Step #13: Optimize design: repeat steps #3 to #9 to find smallest diameter and plate
spacing acceptable to get the lowest cost for the specified application
Step #14: Finalize design: draw up the plate specification and sketch the layout

Lecture 4: Hand on Design

4. Design problem
Design a continuous distillation column (plate) to recover acetone from a 50-50 mole %
acetone-water mixture available at 30°C. The feed stream flow rate is 25,000 kg/h. The
top product should contain at least 95 mole% acetone and the bottom product should
contain <1 % acetone by mole. Consider reboiler as equivalent to one stage. This column
is operated at atmospheric pressure (top tray). Column efficiency of 60% and pressure
drop per plate of 1.25 kPa may be assumed. You can take the minimum liquid flow as
70% of the maximum rate both above and below the feed plate. The vapor liquid
equilibrium (VLE) data for the acetone-water system at atmospheric pressure is provided
in Table 7.2.
Data given:
Latent heat of water= 41,360 J/mol; latent heat of acetone= 28,410 J/mol
Specific heat of water=75.3 J/mol°C (mean); Specific heat of acetone 128 J/mol°C
(mean)
Table 7.2: VLE data for the acetone-water system at 1 atm.
𝒙 0.0 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95
𝒚 0.0 0.6381 0.7301 0.77160.7916 0.8034 0.81240.82010.82690.83760.8387 0.84550.8532 0.86150.87120.8817 0.8950.9118 0.9335 0.9627
BP, 10074.8 68.53 65.26 0.63.5962.6 61.87 61.26 60.75 60.35 59.95 59.54 59.12 58.71 58.29 57.9 57.4957.08 56.68 56.3
°C
𝑥= Mole fraction of acetone in liquid; 𝑦= Mole fraction of acetone in vapor; BP: Bubble
point
Step #1: Mass balance and determination of number of theoretical stage

Feed and products compositions:
Component Feed mole Top product mole Bottom product mole
fraction fraction fraction
Acetone 0.50 0.95 0.01
Water 0.50 0.05 0.99
Bubble point of feed (from the data shown in table) = 59.95°C

Latent heat of the feed = 28,410×0.5 + 41,360×(1 - 0.5) = 34,885 J/mol
Specific heat of the feed = (128×0.5) + 75.3× (1 - 0.5)= 101.75 J/mol °C

Heat required to vaporize 1 mole of the given feed = (59.95 - 30) ×101.75 +
34,885=37933 J
Heat required to vaporize 1 mole of the given feed 37933
𝑞= = = 1.09
Latent heat of the feed 34885
𝑞 1.09
Slope of the q-line= 𝑞−1 = 1.09−1 = 12.44
Here, the top operating line just touches the equilibrium curve at the point of tangency of
the rectifying section operating line at which the minimum reflux takes place.
xD
From the Figure 7.8:  0.57 ; Rmin= 0.67 for xD=0.95
Rmin  1
Here, reflux ratio, R=2.5×Rmin =2.5×0.67= 1.675 is taken for this design.
Average molecular wt. of feed= 0.5×58 + 0.5×18=38
Molar feed flow (𝐹) rate=25,000/38=657.9 kmol/h
Acetone balance: 𝐷 × 0.95 = 657.9 × 0.5⇒𝐷 = 346.2 kmol/h
Vapor flow (𝑉) rate above feed plate, 𝑉 = 𝐷 1 + 𝑅 = 346.2 1 + 1.675 = 926.2
kmol/h
(Assuming constant molar overflow)
Top section liquid flow rate, 𝐿 = 𝑉 − 𝐷 = 580 kmol/h
Bottom product: 𝐵 = 𝐹 − 𝐷 = 657.9 − 346.2 = 311.7 kmol/h
Mass balance below feed plate: 𝐿′ = 𝑉 ′ + 𝐵
𝐿′
Slope of the bottom section operating line (Figure 7.8): 𝑉 ′ = 1.32
𝐿′ = Liquid flow rate below feed plate = 1285.7 kmol/h

𝑉 ′ = Vapor flow rate below feed plate = 974 kmol/h
The construction of operating lines and number of theoretical stages are shown in this
Figure 7.8.
Total number of tray= 6 (above feed) +3 (below feed) =9
9−1
Total number of real stages= ≈ 14 (60% column efficiency; reboiler was
0.6
considered as equivalent to one theoretical tray)

1.0
Equilibrium curve 6th stage
Top operating line for Rmin
0.9
0.8
y, mole fraction of Acetone

6
0.7
q-line
0.6
0.5
9th stage Top operating line for R=2.5Rmin

0.4
0.3
0.2
xD/(Rmin+1)
Bottom operating line
0.1
0.0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
x, mole fraction of Acetone
Figure 7.8: McCabe-Thiele construction.
Step #2: Estimation of physical properties

Column top pressure= 101325 Pa (1 atm)
Column pressure drop=1.25×103×14=16800 kPa
Pressure drop of 1.25kPa per tray is specified
Top section: Bottom section:
Column top pressure= 101325 Pa (1.0147 Column bottom pressure=101325 +16800
bar) and temperature= 56.3 °C = 118825 Pa (=1.19 bar)
𝑃𝑀 101325 ×56.5 Boiling point of water at 118825 Pa (1.19

𝜌𝑣 = = 329 .3×8.314 ×10 3 = 2.08 kg/m3
𝑅𝑇
bar)= 105 °C (bottom contains 99 mole %
water)
𝜌𝑙 =744 kg/m3 (density of the mixture)
From the steam table at 118825 Pa and 105

(water density= 985 and acetone density=
3 °C: 𝜌𝑣 = 0.693; 𝜌𝑙 = 955 kg/m3
735 kg/m at 56.3 °C)
Average molecular weight of vapor:

Average molecular weight of vapor:
M=40.58
M=56.5
Average molecular weight of liquid:
Average molecular weight of liquid: M=56
M=18.4
Surface tension, 𝜍 =20×10-3 N/m

𝜍 =58×10-3 N/m

Lecture 5: Provisional Plate Design

Step #3: Plate spacing
Plate spacingof 600 mm is considered for the first trial to calculate capacity parameter
(𝐶𝑠𝑏𝑓 ) for the estimation of maximum allowable vapor velocity through the net plate area
([3] page 567, [4] page 14-27). The suggested plate spacing is 600 mm for column
diameter>1.5 m.
Step #4: Column diameter (refer to sections 7.3.1 &7.6.1; Eqs. 7.1 & 7.2)
1st trial is started with the following considerations:
Design is performed for 80% flooding at maximum gas flow rate.
Total downcomer top and bottom seal area is 10% of the net area.
Top section: Bottom section:
Flow parameter (𝐹𝐿𝐺 ) based on mass flow rate, 𝐿′ 𝜌 𝑣 0.5
𝐹𝐿𝐺 =𝑉 ′ 𝜌𝑙
𝐿 𝜌 𝑣 0.5 580 ×56 2.08 0.5
= 926.2×56.5 =0.033 1285 .7×18.4 0.693 0.5
𝑉 𝜌𝑙 744
= =0.016
974 ×40.58 955
Capacity parameter (𝐶𝑠𝑏𝑓 ) = 0.12 m/s

𝐶𝑠𝑏𝑓 = 0.11 m/s
Gas velocity through the net area at flooding:

𝜍 0.2 𝜌𝑙 − 𝜌𝑣 0.5
𝜍 0.2 𝜌 𝑙 −𝜌 𝑣 0.5 𝑈𝑛𝑓 = 𝐶𝑠𝑏𝑓
𝑈𝑛𝑓 = 𝐶𝑠𝑏𝑓 = 0.12 × 20 𝜌𝑣
20 𝜌𝑣
0.2 0.5
20 0.2 744 −2.08 0.5 58 955 − 0.693
= 2.26 m/s = 0.11 ×
20 2.08 20 0.693
[𝜍 = liquid surface tension, mN/m] = 5.05 m/s
The linear design gas velocity (𝑈𝑛 ) based on net

area (80% flooding): 𝑈𝑛 = 0.8 × 5.05 = 4.04 m/s
𝑈𝑛 = 0.8 × 2.26 = 1.8 m/s
The maximum volumetric vapor flow rate

(𝑄𝑚𝑎𝑥 ): 𝑉 × 𝑀 974 × 40.58
𝑄𝑚𝑎𝑥 = =
𝑉 × 𝑀 926.2 × 56.5 𝜌𝑣 0.693
𝑄𝑚𝑎𝑥 = =
𝜌𝑣 2.08 = 57034.5 m3/h=15.84 m3/s
=25158.8 m3/h=6.98 m3/s

Net area required: 𝑄𝑚𝑎𝑥 15.84

= =3.92 m2
𝑈𝑛 4.04
𝑄𝑚𝑎𝑥 6.98
= =3.88 m2
𝑈𝑛 1.8
Totals tower cross-section area:

3.88 3.92
=4.31 m2 = =4.36 m2
0.9 0.9
(Total downcomer top and bottom seal area is

10% of the net area)
Colum (tower) diameter: Colum (tower) diameter:

4.31 4.36
=2.34 m =2.36 m
0.785 0.785
Use the highervalue of the tower diameter for the uniformity between sections, if the
difference is not greater than 20%. In this case, the bottom diameter is used both in
top and bottomsections. Higher area than the design area (here top section) can be
taken care by reducing the perforated area.
The nearest recommended shell (nominal diameter 2400 mm) fabricated from
carbon steel or stainless steel sheet in IS 2844-1964: ID 2403 mm with minimum
wall thickness: 8 mm for carbon steel) and 6 mm for stainless steel.
Step #5: Selection of liquid-flow arrangement
580 ×56
Liquid volumetric flow rate in the top section = 3600 ×744 ≈ 0.012m3/s
1285 .7×18.4
Liquid volumetric flow rate in the top section = ≈ 0.007m3/s
3600 ×955
Therefore, single pass cross-flow sieve plate is chosen for this service ([3] page 568).
Step #6: Provisional plate design (refer to sections 7.6.2 & 7.6.3)
Column (tower) diameter (ID): 𝐷𝑇 ≈2.4 m
Column cross-section area: 𝐴 𝑇 =0.785×𝐷𝑇 2 =4.52 m2
Downcomer area: 𝐴𝐷 = 0.1𝐴 𝑇 = 0.452 m2
Net area: 𝐴𝑁 = 𝐴 𝑇 − 𝐴𝐷 = 4.068 m2

Weir Length (𝑙𝑊 ) = 0.73×𝐷𝑇 = 1.752 m ([3] page 573)

Weir height, ℎ𝑤 = 40 mm is considered.
Active area:𝐴𝐴 = 𝐴 𝑇 − 2 × 𝐴𝐷 =3.616 m2
1
For the first trial, consider hole diameter: 𝑑ℎ =12 mm (2inch). The plate thickness=hole
diameter is selected for the first trial.

Step #7: Checking for weepage (refer to section 7.3.2; Eqs. 7.3 & 7.4)
Top section Bottom section
580 ×56 1285 .7×18.4
Maximum liquid flow rate (𝑚𝑚𝑎𝑥 )= =9.02 𝑚𝑚𝑎𝑥 = 3600 =6.571 kg/s
3600
kg/s
𝑚𝑚𝑖𝑛 =0.7×6.571=4.6 kg/s
Minimum liquid flow rate (𝑚𝑚𝑖𝑛 ) (70% of
𝑚𝑚𝑎𝑥 )=0.7×9.02=6.32 kg/s
2
6.571 3
2 ℎ𝑤𝑐 = 750 =18.7
𝐿𝑤𝑐 3 1.752 ×955
Maximum weir crest, ℎ𝑤𝑐 = 750 = mm liquid height
𝐿𝑊 𝜌 𝑙
2
9.02 3
750 =27.2 mm liquid height
1.752 ×744
2
4.6 3
2 ℎ𝑤𝑐 = 750 =14.7
𝐿𝑤𝑐 3 1.752 ×955
Minimum weir crest, ℎ𝑤𝑐 = 750 = mm liquid height
𝐿𝑊 𝜌 𝑙
2
6.32 3
750 =21.5 mm liquid height
1.752 ×744
𝐾2 = 30.2at ℎ𝑤𝑐 + ℎ𝑤 =54.7
The constant (𝐾2 ) of weep-point correlation= 30.3 at
ℎ𝑤𝑐 + ℎ𝑤 =40+21.5=61.5 mm using minimum liquid
flow rate ([3] page 571).
The minimum vapor velocity (𝑈min ) at the weep 𝑈 min at the weep point:
point: 𝑈 =
30.2−0.9(25.4−12)
=21.8 m/s
min (0.693)1/2
𝐾2 −0.9(25.4−𝑑 ℎ ) 30.3−0.9(25.4−12)
𝑈min = 1/2 = =12.6 m/s
𝜌𝑣 (2.08)1/2
Actual minimum vapor velocity at minimum vapor

flow rate: 0.7×15.84
= 0.489 =22.7 m/s
Actual vapor flow rate 70 % of 𝑄𝑚𝑎𝑥 0.7×6.98
= = = 0.489 =10
𝐴
𝐻 𝐴 𝐻
m/s

Therefore, the minimum operating velocity both in top and bottom sections is of above
the weep point velocity.
Step #8: Plate pressure drop (refer to section 7.4; Eqs. 7.7, 7.8 & 7.10)
𝑄 𝑚𝑎𝑥 6.98 𝑄 15.84
Maximum vapor velocity: 𝑈 𝑚𝑎𝑥 = 𝐴 = 0.489=14.3 𝑈 𝑚𝑎𝑥 = 𝐴𝑚𝑎𝑥 = 0.489=32.4 m/s
𝐻 𝐻
m/s
32.4 2 0.693
ℎ 𝑑 = 51 = 50.2 mm
Maximum dry plate pressure drop: ℎ𝑑 = 0.88) 955
𝑈 𝑚𝑎𝑥 2𝜌
𝑣 14.3 2 2.08 liquid
51 = 51 = 37.5 mm liquid
𝐶0 𝜌𝑙 0.88) 744
[The orifice coefficient, 𝐶 0 =0.88 at 𝐴 𝐻 /𝐴 𝑃 = 15%
plate thickness
and =1 ([3] page 576).
hole diameter 12.5×103 12.5×103
ℎ𝑟 = 𝜌𝑙
= 955
=13 mm
3 3
12.5×10 12.5×10
Residual head: ℎ 𝑟 = = =16.8 mm liquid
𝜌𝑙 744
liquid
ℎ 𝑡 = 50.2 + (40+18.7)+13≈122
mm liquid
Total plate pressure drop: ℎ 𝑡 = ℎ 𝑑 + ℎ 𝑤𝑐 +
ℎ 𝑤 + ℎ 𝑟 = 37.5 + (40+27.2)+16.8≈122 mm liquid
The plate pressure drop of 1.25 kPa (=127 mm of water and 161 mm of acetone
pressure) was assumed. The estimated value in the first trial is therefore acceptable.
Step # 9: Downcomer backup liquid and downcomer residence time: (refer to

section 7.4; Eqs. 7.11-7.13)
Downcomer back: ℎ 𝑏 = ℎ 𝑤𝑐 + ℎ 𝑤 + ℎ 𝑡 + ℎ 𝑑𝑐
𝐿 𝑤𝑑 2
Head loss in downcomer: ℎ 𝑑𝑐 = 166
𝜌 𝑙 𝐴𝑚
Downcomer liquid flow rate (𝐿 𝑤𝑑 ) =maximum liquid flow rate is taken
𝐴 𝑚 is smaller of 𝐴 𝑎𝑝 and 𝐴 𝐷 .
𝐴 𝑎𝑝 = ℎ 𝑎𝑝 𝑙 𝑤 = 30 × 10−3 × 1.752 =0.0525 m2 (typically ℎ 𝑎𝑝 = ℎ 𝑤 − 10)
Here, 𝐴 𝑎𝑝 < 𝐴 𝐷 = 0.452 m2
9.02 2 6.571 2
ℎ 𝑑𝑐 = 166 744×0.0525
=8.8 mm ℎ 𝑑𝑐 = 166 955×0.0525
=2.9 mm
ℎ 𝑏 = 27.2 + 40 + 122 + 8.8 ≈198 mm ℎ 𝑏 = 18.7 + 40 + 122 + 2.9 ≈184 mm

Downcomer residence time: 0.452×(184×10−3 )×955

𝑡 𝑑𝑟𝑡 = 6.571
≈ 12 s >3 s.
𝐴 𝐷 ℎ 𝑏𝑐 𝜌 𝑙 0.452× 184×10−3 ×955
𝑡 𝑑𝑟𝑡 =
𝐿 𝑤𝑑
=
6.571
≈ 12 s
>3 s.
ℎ 𝑏𝑐 = clear liquid back up
(ℎ 𝑏𝑐 ≈ ℎ 𝑏 can be approximated)
1 1
plate spacing + weir height = 2 600 + 40 = 320 mm
2
1
plate spacing + weir height > ℎ 𝑑
2
Therefore, the plate spacing and downcomer residence time in both the sections meet
the design required design criteria.

Lecture 6: Provisional Plate Design

Step #10: Calming zones and hole pitch (refer to sections7.6.2 &7.6.4; Eq. 7.15)
Perforated area (𝐴 𝑃 ): AP  AA  ACZ  AES
Where, 𝐴 𝐶𝑍 = calming zone area (Figure 7.3).

𝐴 𝐸𝑆 = area occupied by edge strip (Figure 7.9)
𝑙 𝑊
=0.73; now, 𝜃 𝑐 = 95° ([3] page 573)
𝐷𝑇
Angle subtended by the chord (edge plate), 𝛼 = 180°-95° = 85°

The unperforated edge strip (edge plate) mean length from the geometry:
𝛼 ×𝜋 85×𝜋
𝑙 𝐸𝑆 =(𝐷 𝑇 -50×10-3)× 180 =(2.4-50×10-3)× 180
=3.49 m
𝐴 𝐸𝑆 = 50×10-3×𝑙 𝑀𝑆 =0.175 m2
Use 50 mm wide calming zones. The approximate mean length of zones:
𝑙 𝐶𝑍 =Weir length (𝑙 𝑊) + Width of unperfortaed edge strip
=1.752+50×10-3=1.802 m
𝐴 𝐶𝑍 = 2(50×10-3×𝑙 𝐶𝑍 )= 0.18 m2
Therefore, perforation area per tray (𝐴 𝑃 ) = 𝐴 𝐴 − 𝐴 𝐶𝑍 − 𝐴 𝐸𝑆 =3.616- 0.18-0.175=
3.26 m2
Take total hole area 𝐴 𝐻 = 0.15𝐴 𝐴 = 0.489 m2
𝐴 𝐻 = 0.785 × 𝑑 ℎ 2 × 𝑛 ℎ = 0.489 m2 [hole diameter 12 mm]
Number of holes (𝑛 ℎ )= 4326
𝐴 𝑑ℎ 2
𝐴 𝐻 /𝐴 𝑃 = 0.15. For equilateral triangular pitch: 𝐴 𝐻 = 0.907 𝐼𝑃
𝑃
This corresponds to hole-pitch to hole diameter ratio of (𝐼 𝑃 /𝑑 ℎ ) = 2.46. This is very

close to the normal range of 2.5 to 4.0 times of hole diameter.
The estimated hole pitch (𝐼 𝑃 )=is 29.5 mm.

Figure 7.9. Angle subtended by the chord ([3] page 583).

Steps # 11 and 12: Entrainment checking (refer to section 7.3.3; Eq. 7.6)
Actual vapor velocity (𝑈 𝑣 ) based on net area (𝐴 𝑁 ) 𝑈 𝑣 = 𝑄 𝑚𝑎𝑥 = 15.84 =3.9 m/s
𝐴𝑁 4.068
selected provisionally: 𝑈 3.9
𝑄 𝑚𝑎𝑥 6.98
% flooding=𝑈 𝑣 × 100 = 5.05 ×
𝑈 𝑣 = 𝐴 = 4.068 =1.7 m/s 𝑛𝑓
𝑁
100 =77%
𝑈𝑣 1.7
% flooding=𝑈 × 100 = 2.26 × 100 =76 %
𝑛𝑓
Ψ=0.18 at 𝐹 𝐿𝐺 = 0.016 and actual

The fractional entrainment, Ψ=0.09 at 𝐹 𝐿𝐺 = flooding velocity of 77 % ([4] page
𝐿 𝜌 𝑣 0.5 14-28).
= 0.033 and actual flooding velocity of
𝑉 𝜌𝑙
76 % ([4] page 14-28).
Effect of Ψ on Murphree plate efficiency can be

estimated from ([4] page 14-29):
𝐸 𝑎 = 0.53
𝐸 𝑚𝑣
𝐸𝑎 = Ψ𝐸 𝑚𝑣 = 0.57
1+ 1−Ψ
𝐸 𝑚𝑣 = 0.6 (Murphree vapor efficiency 60%)

E𝑎 =Murphree vapor efficiency, corrected for
liquid entrainment
The actual flooding is below the design flooding value of 80%. Usually, Ψ<0.1 is
desirable. However, the optimum design value may be above this.

Design problem: Absorption column

An industrial gas stream is available @ 2 kg/sfrom a cracking operation of
NH3containing 72% H2, 24% N2 and 4% NH3by mole, at 202.65 kPa and 35°C.You have
been asked to design a multistage countercurrent bubble cap absorber to remove NH3
from the above stream with water as the scrubbing liquid. The liquid mass flow rate is
limited to be 2 to 3.5 times of gas mass rate.NH3concentration should not be greater than
0.003 mg per m3 of the exit gas.
Assumptions/ design considerations:
 Lean water-NH3 system follows Henry’s law and the corresponding equilibrium
relation: 𝑦 ∗ = 0.85𝑥 @30°C
 Isothermal gas absorption at room temperature (~30°C)
 Optimum adsorption factor (𝐴 )= 1.2 to 2
 Overall column efficiency=70%
 Pressure drop per plate= 1 kPa
 Minimum liquid loading=70% of expected maximum loading
Hints:The flow rates of liquid and gas entering and leaving the absorber is almost
constant thought out the column if a small amount of the solute gas is absorbed. This is a
typical case, also common in practice when the solute gas concentration in the feed
stream is low (dilute gas absorption). For such operations, the variation of temperature
between column top and bottom trays is insignificant (~isothermal operation). The
pressure drop, if the column is not too tall, has minor effect on the physical properties of
process fluids that could influence the column design. The section-wise determination of
number trays and design are performed if the gas stream and/ or the solvent liquid
(usually makeup solvent) are introduced at any intermediate point of the column. The
number of theoretical trays can be estimated using either Kremser equation or graphical
technique for multistage counter current lean gas absorption ([1] page 290). The vapor
loading is the highest at the bottom tray even though its variation is not appreciable. For
the safe side, the design is usually performed at the bottom tray.

The mole fractions (𝑥 ,𝑦 ) between the phases are plotted in McCabe Thiele method of
𝑥 𝑦
distillation calculation. In case of absorption, the mole ratios ( 𝑋 = 1−𝑥 , 𝑌 = 1−𝑦 ) are
used for the determination of number of trays in graphical method instead of mole
fractions.
References
[1]. Robert E. Treybal, Mass Transfer Operations, McGraw-Hill, Inc., 3rd ed. 1981.
[2]. Perry’s Chemical Engineers’ Handbook, McGraw-Hill, Inc., 8th ed. 1997.
[3]. R. K. Sinnott, Coulson & Richardson’s Chemical Engineering: Chemical
Engineering Design (vol. 6), Butterworth-Heinemann, 3rd ed. 1999.
[4]. Perry’s Chemical Engineers’ Handbook, McGraw-Hill Companies, 7th ed. 1997.
[5]. Henry Z. Kister, Distillation Design, McGraw-Hill, Inc., 1st ed. 1992.

Pdfcoffee.com Distillationpdf 3 PDF Free

Uploaded by

Copyright:

Available Formats

Pdfcoffee.com Distillationpdf 3 PDF Free

Uploaded by

Document Information

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

Pdfcoffee.com Distillationpdf 3 PDF Free

Uploaded by

Copyright:

Available Formats

NPTEL – Chemical – Mass Transfer Operation 1

5.1.1. Vapor Pressure

Joint initiative of IITs and IISc – Funded by MHRD Page 1 of 9

5.1.2. Phase Diagram

5.1.3. Constant Pressure Phase Diagram

Joint initiative of IITs and IISc – Funded by MHRD Page 2 of 9

concentrations of each component in the liquid, many boiling point temperatures

Figure 5.1: Phase diagram of binary system at constant pressure

5.1.4. Constant temperature (isothermal) phase diagram

Joint initiative of IITs and IISc – Funded by MHRD Page 3 of 9

Figure 5.2: Binary Phase diagram at constant temperature

5.1.5. Relative volatility

The Equation (5.4) is a non-linear relationships between x and y. This Equation

Joint initiative of IITs and IISc – Funded by MHRD Page 4 of 9

More volatile component Less boiling component Optimal Relative

5.1.6. Vapor-Liquid Equilibrium (VLE)

Joint initiative of IITs and IISc – Funded by MHRD Page 5 of 9

expresses the bubble-point and the dew-point of a binary mixture at constant

y = 0.671x3 - 1.883x2 + 2.211x + 0.003

45o diagonal line

Mole fraction of benzene in liquid , x

Figure 5.3: Equilibrium diagram for a benzene-toluene mixture at 1 atmosphere

Joint initiative of IITs and IISc – Funded by MHRD Page 6 of 9

Figure 5.4: Vapor-liquid equilibrium curve for non-ideal systems

The most challenging VLE curves are generated by azeotropic systems. An

Joint initiative of IITs and IISc – Funded by MHRD Page 7 of 9

Although most distillations are carried out at atmospheric or near atmospheric

Joint initiative of IITs and IISc – Funded by MHRD Page 8 of 9

shown in Figure 5.6. Separability becomes less at higher pressures.

Figure 5.6: Variation of equilibrium curve with pressure

Joint initiative of IITs and IISc – Funded by MHRD Page 9 of 9

5.2. Distillation columns and their process calculations

5.2.1. Batch or differential distillation columns and their process

Joint initiative of IITs and IISc – Funded by MHRD Page 1 of 4

originally in still, L2 = final moles of liquid remained in still, x1 = initial liquid

Figure 5.7: Simple batch or differential distillation process

Initial amount in still = Amount left in still + Amount vaporized

xL  ( x  dx)( L  dL)  ydL (5.6)

Joint initiative of IITs and IISc – Funded by MHRD Page 2 of 4

xL  xL  xdL  Ldx  dxdL  ydL (5.7)

Ldx  ydL  xdL (5.8)

Example problem 5.1

Joint initiative of IITs and IISc – Funded by MHRD Page 3 of 4

The value of x2 = 0.067

Joint initiative of IITs and IISc – Funded by MHRD Page 4 of 4

5.2.2. Continuous distillation columns

Types of Continuous Columns

Joint initiative of IITs and IISc – Funded by MHRD Page 1 of 5

4.2.2.1. A single-stage continuous distillation (Flash distillation):

A single-stage continuous operation occurs where a liquid mixture is partially

Joint initiative of IITs and IISc – Funded by MHRD Page 2 of 5

Figure 5.8: Flash distillation process

The Equation (5.11) on rearranging becomes:

Joint initiative of IITs and IISc – Funded by MHRD Page 3 of 5

Figure 5.9: Graphical presentation of flash veporization

5.2.2.2. Multi-stage Continuous distillation-Binary system

Joint initiative of IITs and IISc – Funded by MHRD Page 4 of 5

Figure 5.10: Multi-stage binary distillation column

Joint initiative of IITs and IISc – Funded by MHRD Page 5 of 5

5.2.2.3. Analysis of binary distillation in tray towers: McCabe-