(23-5-6) NPTEL - Gas Separation

23
1
Earlier Lecture
• Earlier, we have studied the Temperature
composition diagrams, the Enthalpy composition
diagrams and their importance in Gas separation.
• The separation of a mixture is more effective

when the difference in the boiling points is more.
• In this column, Low and High Boiling components

are collected at top and bottom respectively.
• Murphree efficiency is the ratio of actual

change in mole fraction to the maximum possible
change that can occur.
2
Prof. M D Atrey, Department of Mechanical Engineering, IIT Bombay
Outline of the Lecture
Topic : Gas Separation (contd)
• Understanding of Rectification Column using an

Animation
• Theoretical Plate Calculations
3
Rectification Column
• Animation
4
• As seen earlier in a
p = 1atm rectification column, the
liquid moving down is
3'
Feed in enriched in high boiling
Temperature, K
4
2'
1 point component (O2).
3
2
1
2 3• On the other hand, the
2'
vapor moving up is
3'
4 enriched in low boiling
point component (N2).
O2 = 1 O2 = 0
N2 = 0 N2 = 1
5
• For getting 100% pure
p = 1atm products, infinite
number of rectification
3'
Feed in processes – plates,
Temperature, K
4
2'
1 would be required.
3
2
1
2 • But in reality, the size
3
2'
and the cost of the
3'
4 column limit the
number of rectification
processes and hence
O2 = 1 O2 = 0
N2 = 0 N2 = 1 the purity.
6
• In the past, researchers have
developed various mathematical
procedures to calculate the required
number of rectification processes –
plates, to obtain a desired purity.
• These procedures require the

following data.
• Number of components
• Phase diagrams of the mixtures
• Property data of mixture
• Heat transfer correlations
7
Theoretical Plate Calculations
• The methods of calculation
that are used for theoretical
plate calculations are
• Method of Ponchon and

Savarit.
• Method of McCabe and

Thiele.
• Numerical Methods.
8
• Ponchon – Savarit method
is an exact method for plate
calculations.
• It is applicable to any number

of components and this
method requires a detailed
data of enthalpy composition
diagram(s) of the mixture.
9
• McCabe – Thiele method
was proposed by two
American scientists, Warren
McCabe and Ernest Thiele, in
the year 1925.
• This method is less general

and is the simplest technique.
It is widely used for binary
mixtures at cryogenic
temperatures.
10
• Numerical methods are the
latest techniques, which are
tedious, time consuming and
computer intensive methods.
• For the sake of understanding

and simplicity, only McCabe –
Thiele method will be
explained in this topic.
11
McCabe – Thiele Method
• This method calculates liquid and vapor
fractions of each component at every
D plate and also the number of plates.
• For the sake of understanding, let the

n plates above the feed be denoted by
subscript n.
m • Similarly, the plates below the feed be

denoted by subscript m.
B
• Let the total mole flow rate of top and

bottom product be D and B respectively
12
j +1
Vj
Lj
j
• It is important to understand the indexing pattern

of the plate and its corresponding liquid and vapor.
• Let jth and (j+1)th plate be any intermediate plate

as shown in the figure.
• The liquid and vapor leaving from top of the jth

plate are Lj and Vj respectively.
13
j +1
L j +1
Vj
Lj
j
V j −1
• Similarly, the liquid coming to the jth plate is from

(j+1)th plate, therefore it is Lj+1.
• Also, the vapor coming to jth plate from bottom is

vapor leaving the (j-1)th plate. It is therefore, Vj-1.
• The vapor and liquid on any plate, Lj and Vj, are in

thermal equilibrium.
14
• Consider a control volume enclosing
the condenser and the top section of
D the nth plate as shown in the figure.
• As explained earlier, for this nth plate,

n the vapor leaving is Vn and the liquid
added is Ln+1.
m • Applying the mole balance across the

control volume per unit time, we have
B IN OUT
Vn Ln+1 =
Vn Ln +1 + D
D
15
• Multiplying the mole balance equation
with mole fraction of a particular
D component in a mixture, we get mole
balance for that component as
=
ynVn xn +1 Ln +1 + xD D
n • Where,
• yn, xn+1 and xD are mole fractions
of a particular component in vapor,
m liquid and top product respectively.
B
• It automatically means that xD
(mole fraction) is the desired purity
of the top product.
16
Q D • For control volume taking into account
QD (watts) as the heat rejected by the
D condenser, the enthalpy balance is
given by
H nV=
n hn +1 Ln +1 + hD D + Q D
n • Dividing the above equation by D, we
have
H nVn Ln +1 Q D
= hn +1 + hD +
m D D D
• Rearranging the total mole balance

B
equation, we have
Ln +1 Vn
Ln += V − D = −1
1 n D D
17
• Eliminating Ln+1/D from the earlier
equations, we get
H nVn  Vn  Q D
= hn +1  − 1 + hD +
D D  D
Vn Q D
( H n − hn+1 ) = + hD − hn +1
D D
• Rearranging as a ratio of D and Vn,
we have
D H n − hn +1
= 
Vn QD
+ hD − hn +1
D
18
N2 – O2 at 1 atm
Enthalpy, h – kJ/mol
90
88
85
84
83 • The enthalpy composition
82 diagram for a mixture of N2
81 and O2 is as shown.
80
78
• If we neglect the enthalpy
variation with the mole
fraction, the bubble and dew
Mole fraction N2 lines can be taken as
horizontal.
19
D H n − hn +1 Ln +1 D
= = 1−
Vn Q D Vn Vn
+ hD − hn +1
D
• These arguments lead to the fact that liquid (h)
and vapor (H) enthalpies are constant. Hence,
D/Vn and Ln+1/Vn are constant.
• Rearranging the molar balance for a component as

 Ln +1  D
=yn   xn +1 +   xD
 Vn   Vn 
• The above equation represents a straight line and
is called as Operating Line for stripping section.
20
 Ln +1  D
=yn   xn +1 +   xD
 Vn   Vn 
• For the top or upper most
plate near the condenser,
xn+1=xD.
 Ln +1  D  Ln +1 D 
= yn   xD +   xD =
yn  +  xD
• Substituting,  Vn   Vn  1  n
V Vn 
• For y – intercept, xn+1=0. yn = xD
D Two Points

yn =   xD yn=xD @ xn+1=xD
 Vn  yn=(D/Vn)xD @ xn+1=0
21
• A plot of vapor versus
liquid mole fractions
for a particular
component, say A, is
yA
• Let 45o diagonal or
y=x line be as shown.
• The desired purity of

xA xD this component A, in
the top product is xD
22
 Ln +1  D
=yn   xn +1 +   xD
OP line for  Vn   Vn 
Enriching Section
• The y – intercept of
the straight line is
yA (D/Vn)xD.
xD D
Vn • Similarly, the slope of
the operating line is
given by Ln+1/Vn, as
xA xD shown in the above
equation.
23
Q D • Similarly, for the analysis of mth plate
and boiler in the lower part, we have
D the following equations.
• Mole Balance: Lm +=
1 Vm + B
n
xm +1=
Lm +1 ymVm + xB B
• Energy Balance: hm +1 Lm +1 + Q B= H mVm + hB B

m
B • where, B and QB are mole flow rate

out at the bottom and heat input to
the boiler respectively.
Q B
24
Q D • Rearranging the above equations, we
have the following.
D
B H −h Lm +1 B
=  m m +1 = 1+
Vm Qm
− hB + hm +1 Vm Vm
B
n
• Applying the assumption, we have Hm
and hm+1 as constant, implies B/Vm
and Lm+1/Vm are constant. The
m operating line for stripping section is
B  Lm +1   B
=ym   xm +1 −   xB
 Vm   Vm 
Q B
25
 Lm +1   B
=ym   xm +1 −   xB
 Vm   Vm 
• For the bottom or lower most
plate near the boiler,
xm+1=xB.
 Lm +1   B  Lm +1 B 
= ym   xB −   xB =
ym  −  xB
• Substituting,  Vm   Vm 
1  Vm Vm 
• For y – intercept, xm+1=0. ym = xB
 B Two Points
ym = −   xB ym=xB @ xm+1=xB
 Vm  ym=-(B/Vm)xB @ xm+1=0
26
• The plot of vapor
versus liquid mole
OP line for fractions for a
Enriching Section
component A with
operating line and 45o
diagonal be as shown.
yA
xD D
Vn • The purity of
component A in the
bottom product is xB
xB xA xD as shown in the figure.
27
 Lm +1   B
=ym   xm +1 −   xB
OP line for  Vm   Vm 
Enriching Section
OP line for
Stripping Section • The y – intercept of
the straight line is
yA (-(B/Vm)xB).
xD D
Vn • The slope of the
operating line is given
by Lm+1/Vm as shown
xB xA xD in the above equation.
xB B
Vm
28
Q D • The mixture that is to be separated is
called as Feed. It is introduced into
D the column through an opening called
as Feed inlet as shown in the figure.
n • Consider a control volume enclosing

F the nth and mth plates and feed inlet
as shown.
m
• Let F be the total number of moles in
B the Feed.
Q B
29
Q D Ln +1 Vn
n
F
D
Vm Lm +1
m
• For the above control volume, we have

n
IN OUT
F
F
m Vm Vn
Ln+1 Lm+1
B • Applying the molar balance
F = Vn − Vm + Lm +1 − Ln +1
Q B
30
Ln +1 Vn
n
F
Vm Lm +1
m
• We define a parameter q as the

ratio of liquid moles in the feed to
the total number moles in the feed.
q=
( Lm +1 − Ln +1 )
• Mathematically,
F
• That is for q=0, feed is totally vapor

and for q=1, it is totally liquid.
31
• From the earlier slides,
we know the equations
OP line for for both the sections.
Enriching Section
OP line for
Stripping Section
• The locus of intersection

yA of these operating lines
xD D
denotes the feed
Vn condition.
• The condition of the

xB xA xD feed is vital to
xB B
determine the number
Vm
of plates.
32
• Based on feed equation and q definition, we have
F = Vn − Vm + Lm +1 − Ln +1 q=
( Lm +1 − Ln +1 )
Vn − Vm =(1 − q ) F
F
• Again, from the operating lines of upper and lower

sections, we can rearrange to give Vn and Vm as
 Ln +1  D  Lm +1   B 
=Vn   xn +1 +  =  xD Vm   xm +1 −   xB
 yn   yn   ym   ym 
• It is important to note that Vn-Vm is the vapor

content in the feed.
33
• In the calculation of point of intersection of
operating lines, we choose a common point to both
these lines as (x,y).
• Hence, xn+1, xm+1, ym and yn are replaced with

this point as shown in the following equation.
Vn − Vm =
( Ln +1 − Lm +1 ) x ( xD D + xB B )
+ =(1 − q ) F
y y
• The locus of this point of intersection is the feed

line or q line and is calculated as explained in the
next slide.
34
Q D • For a column as a whole, using the
mass balance, we can write
D
x= xDD+ xBB q=
( Lm +1 − Ln +1 )
FF
F
n
• Rearranging the following equations,
we have
( Ln+1 − Lm+1 ) x + ( xD D + xB B ) =
m (1 − q ) F
y y
B
x xF F
−qF + (1 q ) F
=−
y y
Q B
35
• Rearranging,
 q  xF
=y  x+
 q −1  1− q
• The above equation represents a
straight line with q/(q-1) and
xF/(1-q) as slope and y –
intercept respectively.
• More importantly, it is the locus of

point of intersection of operating
lines. This line is called as Feed
line or q line.
36
Q D • It is clear that the value of parameter
q is yet to be determined.
D
• Applying energy balance to the

control volume as shown in figure, we
n have
hF F =Vn H n − Vm H m + Lm +1hm +1 − Ln +1hn +1
m
• Mathematically, McCabe – Thiele
assumption is
B
=
H n H=
m H , hm=
+1 h=
n +1 h
Q B
37
• Upon substitution, we have
hF F =Vn H n − Vm H m + Lm +1hm +1 − Ln +1hn +1
hF F =(Vn − Vm ) H + ( Lm+1 − Ln+1 ) h

• Also, we have the following equations.
Vn − Vm =(1 − q ) F q=
( Lm +1 − Ln +1 )
F
• Combining the above equations and
rearranging, we have
H − hF
q=
H −h
38
H − hF  q 
N2 – O2 at 1 atm q= slp =  
H −h  q − 1 
• Depending on the feed
Enthalpy, h – kJ/mol
90
88
condition, q can take any
85
84
83 value.
82
81
80
78
Condition q Slp
Sat. Vap. (hF=H) q=0 0
Sat. Liq. (hF=h) q=1 ∞
2 ph. (H<hF<h) 0<q<1 -ve
Mole fraction N2 Sub. Liq. (hF<h) q>1 +ve
Sup. Vap. (hF>h) q<0 +ve
39
• The equation of feed
line is
OP line for
Enriching Section  q  xF
OP line for =y   x +
Stripping Section
 q − 1  1− q
yA
xD D
Vn Condition q Slp
xB xA 2 ph. (H<hF<h) 0<q<1 -ve
xB B Sub. Liq. (hF<h) q>1 +ve
Vm Sup. Vap. (hF>h) q<0 +ve
40
• The point of
intersection of feed line
OP line for or q line and y=x gives
Enriching Section
OP line for the content of the
Stripping Section
component A in feed,
xF .
yA
xD D
Vn • It is calculated by
substituting y=x in the
feed line as shown.
xB xA xF
 q  xF
xB B =x   x + x = xF
Vm  q − 1  1− q
41
• Graphically, it is easier
to draw a line through
OP line for two given points rather
Enriching Section
OP line for than using a given
Stripping Section
slope and a point.
yA
xD D
• This intersection point
Vn is used to draw the
feed line as shown in
the figure.
xB xF x A
xB B
Vm
42
Summary
• Plate calculation procedures require the data like
number of components, phase diagrams, property
data of the mixtures, heat transfer correlations.
• McCabe – Thiele method is less general and is

widely used for binary mixtures at cryogenic
temperatures.
• The major assumption in this method is that the

liquid and vapor enthalpies are independent of
mole fraction.
43
Summary
• The equations of operating lines for striping and
enriching sections are
 Ln +1  D  Lm +1   B
=yn   xn +1 +  =  xD ym   xm +1 −   xB
 Vn   Vn   Vm   Vm 
• The locus of intersection of these operating lines

denotes the feed condition. It is given as
 q  xF
= y  x+
 q −1  1− q
• The point of intersection of feed line or q line and
y=x gives the content of a component in the
feed, xF.
44
Summary
• The slope of q line is
given by
OP line for
Enriching Section
 q 
OP line for  
Stripping Section  q − 1 
yA
xD D
Vn Condition q Slp
xB xA 2 ph. (H<hF<h) 0<q<1 -ve
xB B Sub. Liq. (hF<h) q>1 +ve
Vm Sup. Vap. (hF>h) q<0 +ve
45
• A self assessment exercise is given after
this slide.
• Kindly asses yourself for this lecture.
46
Self Assessment
1. McCabe – Thiele method calculates ______ &
______ of each component at every plate.
2. For a jth plate, the liquid an vapor leaving from
top are denoted by ____ and ____ respectively.
3. The vapor and liquid on any plate are assumed to
be in ______ equilibrium.
4. In McCabe – Thiele method, liquid and vapor
enthalpies are assumed to be ________.
5. The slope of operating line for stripping section is
given by _______.
6. The y – intercept of operating line for enriching
section is given by _______.
7. Mixture that is to be separated is called as ____.
47
Self Assessment
8. q=0 when the feed is totally ______.
9. _______ and ______ are the slope and the y –
intercept of q line respectively.
10. Fill the following table.
Condition q Slp
Sat. Vap. (hF=H) q=0
Sat. Liq. (hF=h) ∞
2 ph. (H<hF<h) 0<q<1 -ve
Sub. Liq. (hF<h) +ve
q<0 +ve
48
Answers
1. Vapor fraction, liquid fraction
2. Lj and Vj
3. Thermal
Condition q Slp
4. Constant
5. Ln+1/Vn Sat. Liq. (hF=h) q=1 ∞
6. (-(B/Vm)xB) 2 ph. (H<hF<h) 0<q<1 -ve
Sub. Liq. (hF<h) q>1 +ve
7. Feed Sup. Vap. (hF>h) q<0 +ve
8. Vapor
9. q/(q-1) and xF/(1-q)
49
Thank You!
50

(23-5-6) NPTEL - Gas Separation

Hochgeladen von

Dokumentinformationen

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

(23-5-6) NPTEL - Gas Separation

Hochgeladen von

Copyright:

Verfügbare Formate

23

• The separation of a mixture is more effective

• In this column, Low and High Boiling components

• Murphree efficiency is the ratio of actual

• Understanding of Rectification Column using an

• Theoretical Plate Calculations

• These procedures require the

• Method of Ponchon and

• Method of McCabe and

• It is applicable to any number

• This method is less general

• For the sake of understanding

• For the sake of understanding, let the

m • Similarly, the plates below the feed be

• Let the total mole flow rate of top and

• It is important to understand the indexing pattern

• Let jth and (j+1)th plate be any intermediate plate

• The liquid and vapor leaving from top of the jth

• Similarly, the liquid coming to the jth plate is from

• Also, the vapor coming to jth plate from bottom is

• The vapor and liquid on any plate, Lj and Vj, are in

• As explained earlier, for this nth plate,

m • Applying the mole balance across the

• Rearranging the total mole balance

• Rearranging the molar balance for a component as

• For y – intercept, xn+1=0. yn = xD

D Two Points

• The desired purity of

• Energy Balance: hm +1 Lm +1 + Q B= H mVm + hB B

B • where, B and QB are mole flow rate

• For y – intercept, xm+1=0. ym = xB

n • Consider a control volume enclosing

• For the above control volume, we have

• We define a parameter q as the

• That is for q=0, feed is totally vapor

• The locus of intersection

• The condition of the

• Again, from the operating lines of upper and lower

• It is important to note that Vn-Vm is the vapor

• Hence, xn+1, xm+1, ym and yn are replaced with

• The locus of this point of intersection is the feed

• More importantly, it is the locus of

• Applying energy balance to the

hF F =(Vn − Vm ) H + ( Lm+1 − Ln+1 ) h

• McCabe – Thiele method is less general and is

• The major assumption in this method is that the

• The locus of intersection of these operating lines

• Kindly asses yourself for this lecture.

Das könnte Ihnen auch gefallen