Optimum Column Design

for Capacity and Efficiency

(Presented at: [100e]-Distillation Honors: Dale Nutter)

G. X. Chen
Fractionation Research, Inc.
P.O. Box 2108, Stillwater, OK 74076
GXChen@fri.org

Prepared for presentation at the AIChE Spring National Meeting

New Orleans, LA
Monday, April 26, 2004

AIChE shall not be responsible for statements or opinions contained in papers or

printed in its publications.

gxc/FRI AIChE 2004 [100e] 1

1. Capacity and Efficiency Performance
2. Capacity
p y and Efficiency
y Models and
Predictions
3 Examples
3.

gxc/FRI AIChE 2004 [100e] 2

Tray Capacity and Efficiency Performance
– Flooding
Fl di Capacity
C i
• System limit
• Jet flooding
• Froth flooding
• DC choke flooding
• DC backup flooding
– Trayy eefficiency
c e cy
• Point efficiency
• Tray liquid mixing pools
• Murphree tray efficiency
• Dry overall tray/column efficiency
• Overall tray/column efficiency
gxc/FRI AIChE 2004 [100e] 3
Packing Capacity and Efficiency Performance
– Flooding Capacity
• System limit
• Fl di capacity
Flooding i
– Packing efficiency or HETP
• HTUOG, HTUG, HTUL
• Mass transfer driving forces, non-ideal flow
• HETP hump for structured packing
• Effect of bed length
• HETP

gxc/FRI AIChE 2004 [100e] 4

System Limit or Ultimate Capacity
Definition:
– Capacity factor is defined as vapor velocity
multiplied
l i li d the
h square root off the
h vapor density
d i
divided by the difference between liquid and vapor
densities.
Vload ρV
Cs = = uV
As ρ L − ρV

– A limiting combination of vapor and liquid loads

which is a function of system properties only. If
exceeded,
d d massive
i entrainment
i occur. Conventional
C i l
tray/packing capacity cannot exceed system limit.

gxc/FRI AIChE 2004 [100e] 5

System Limit or Ultimate Capacity
S
System Limit
i i Model:
d l
– FRI developed system limit model based on data
obtained
b i d for
f trays andd packings.
ki
– System limit is only a function of system physical
properties
ti andd is
i independent
i d d t off hardware.
h d
– The system limit is useful in determining the maximum
capacit a column
capacity col mn can possibly
possibl have.
ha e

gxc/FRI AIChE 2004 [100e] 6

Simulated System Limit for Common Systems
System Limits for FRI Used Systems
0.700
IC4/NC4, 300 psia
IC4/NC4, 400 psia
0.600 IC4/NC4, 165 p
psia
C6/C7, 24 psia
O/P Xylene, 100 mmHg
0.500

0.400
Cs, ft/s

0.300

0.200

0.100

0.000
0.0 20.0 40.0 60.0 80.0 100.0
Liquid rate, gpm/ft2
gxc/FRI AIChE 2004 [100e] 7
System Limit for the O/P Xylene System at 240 mmHg
3 2
Liquid Rate, m /h-m
0 20 40 60 80 100 120 140
0.70
ctor Cs , ft//s

Capacity Facctor Cs, m/s

0 20
0.20
0.60
0.50 0.15
0.40
apacity Fac

0.30 0.10

0.20
0.05
0.10
Ca

C
0.00 0.00
0 10 20 30 40 50 60
2
Liquid
q Rate,, gpm/ft
gp

Calculated System Limit 29% Dualflow Tray, 4'&6' TS

gxc/FRI AIChE 2004 [100e] 8

System Limit for the IC4/NC4 System at 400 psia
3 2
Liquid Rate
Rate, m /h-m
/h m
0 20 40 60 80 100 120 140
0.40 0.12
apacity Factor Cs, ft//s

Capacity Factor Cs, m/s

0.10
0.30
0.08

0 20
0.20 0 06
0.06

0.04
0.10
0.02
Ca

C
0.00 0.00
0 10 20 30 40 50 60
2
Liquid Rate,
Rate gpm/ft

TR133 Sys. Limit 59% Baffle, 12"gap, 2'TS 8% Sieve, 30/7% DC,3'TS
Mellapak 250.Y No.2.5 Nutter Ring

gxc/FRI AIChE 2004 [100e] 9

Trayy Jet Flooding
g Capacity
p y
– Jet flooding is due to massive entrainment of liquid to
the tray above.
– FRI has developed empirical models for jet flood
capacity.

gxc/FRI AIChE 2004 [100e] 10

Tray Jet Flooding Model Functional Analysis:
– Jet capacity increases as bubbling and free area
increases.
– Jet capacity increases as tray spacing increases.
– Jet capacity increases as open area increases.
– Jet capacity increases as hole size decreases.
– Jet flooding often limits column capacity for low
pressure and low liquid rate systems.

gxc/FRI AIChE 2004 [100e] 11

Jet capacity increases as tray spacing increases.
Simulated Sieve Tray Jet Capacity

0.600

0.500

0 400
0.400
Cb, ft/s

0.300

0.200 Ts 24 inch
Ts 18 inch
Ts 30 inch
0.100

0.000
0 2 4 6 8 10 12 14 16
Liquid rate, gpm/inch
gxc/FRI AIChE 2004 [100e] 12
Tray DC Flooding Capacity
Definition:
fi i i
– DC choke flood: flooding occurs when the downward
velocity
l i off froth
f h isi so high
hi h that
h little
li l or no vapor
disengagement occurs. It is determined by DC size.
– DC bbackup
k fl flood:
d Downcomer
D bbackup
k exceedsd th
the
tray spacing plus weir height, so that the froth backs
up on the tray above
above. It is determined by DC depth.
depth

gxc/FRI AIChE 2004 [100e] 13

DC Choke Flooding Model Functional Analysis:
– Mainly a function of physical properties
– Depend on the DC top area
– Depend on the bubbling area
– Same model for all types of trays (sieve, valve,
fixed valve) with downcomer
– Capacity advantage with sloped and truncated
downcomer
– Not much depend on other tray design parameters
such as tray spacing, hole size, open area, etc.

gxc/FRI AIChE 2004 [100e] 14

DC Backup Flooding Model Functional Analysis:
– Mainly a function of pressure drop
– Reduce pressure drop or increase tray spacing to
increase DC backup flooding capacity.
– Same model for all trays with downcomer, but use
diff
different tray pressure ddrop and
d li
liquid
id holdup
h ld

gxc/FRI AIChE 2004 [100e] 15

Effect of DC Area on DC Capacity

Downcomer Flood Capacity

0.20
0.18
0.16
0.14
0 12
0.12
Cb, m/s

0.10
0.08
0.06 Case 1, DC flood, 8% DC Area
0.04 Case 2, DC flood, 16% DC Area

0.02
0.00
0.0 5.0 10.0 15.0 20.0 25.0
Liquid
q rate, kg/s
g

gxc/FRI AIChE 2004 [100e] 16

Effect of DC Depth on DC Capacity

Downcomer Flood Capacity

0.20
0.18
0.16
0.14
0 12
0.12
Cb, m/s

0.10
0.08
0.06
0.04 Case 1, DC Backup Flood, 0.4 m DC
Case 2, DC Backup Flood, 0.6 m DC
0.02
0.00
0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0
Liquid
q rate, kg/s
g

gxc/FRI AIChE 2004 [100e] 17

Effect of DC area on Column Capacity

0.35
Jet Flood
DC Flood
0.30

0.25
d, ft/s

0.20
Cs at flood

0.15

0 10
0.10

0.05

0.00
0 5 10 15 20 25 30 35 40
DC area as % of column area, %

gxc/FRI AIChE 2004 [100e] 18

Column capacity is limited by jet flood
0.45

0.40

0 35
0.35

0.30
od, ft/s

0.25
Cs at floo

0.20
Jet Flood
0.15 DC Choke Flood

0.10

0.05

0.00
0 5 10 15 20 25 30 35 40
DC area as % of column area, %

gxc/FRI AIChE 2004 [100e] 19

Column capacity is limited by DC capacity
0.40

Jet Flood
0.35 DC Choke Flood

0.30

0.25
od, ft/s
Cs at floo

0.20

0.15

0.10

0.05

0.00
0 5 10 15 20 25 30 35 40
DC area as % of column area, %

gxc/FRI AIChE 2004 [100e] 20

Tray Efficiency Models

1. Penetration theory
2. Two-film theory
3. Point efficiency
4. Mixing pools
5
5. Tray efficiency

gxc/FRI AIChE 2004 [100e] 21

Tray Efficiency Model Functional Analysis:

• Physical properties, m,
(MV/L)
• Flow path length, tray liquid
mi ing
mixing
• Open area
• Hole size
• Heir height

gxc/FRI AIChE 2004 [100e] 22

Calculated effect of open area on efficiency

100

90

80
efficiency, %

70

60
Overall tray e

50

40
15% open
30
10% open
O

20 5% open
10

0
0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14
Cb, m/s

gxc/FRI AIChE 2004 [100e] 23

Effect of hole size on efficiency

100

90

80
efficiency, %

70

60
Overall tray e

50

40
24 mm hole
30
12 mm hole
O

20 6 mm hole
10

0
0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14
Cb, m/s

gxc/FRI AIChE 2004 [100e] 24

Packing
g Floodingg Capacity
p y
– FRI has obtained semi-empirical models for packing
capacity.
– Mainly a function of packing specific area
– Crimpp angle
g also affects capacity
p y

gxc/FRI AIChE 2004 [100e] 25

Effect of packing area on packing capacity
Calculated Effect of Packing Specific Area on Capacity
Void Fraction=0.98, for a Given System
3 2
Liquid rate, m /h-m
0 20 40 60 80 100 120 140
0.60 0 18
0.18

0.16
0.50
0.14

dicted Csf, m/s

s
dicted Csf, ft/s
s

0.40 0.12

0.10
0.30
0.08
Pred

Pred
0.20 20 1/ft 0.06
40 1/ft 0.04
0.10 80 1/ft
160 1/ft 0 02
0.02

0.00 320 1/ft 0.00

0 10 20 30 40 50 60
2
Liquid rate, gpm/ft

gxc/FRI AIChE 2004 [100e] 26

Effect of physical properties on packing capacity
Calc lated Effect of Ph
Calculated Physical
sical Properties on Capacit
Capacity

3 2
Liquid rate, m /h-m
0 20 40 60 80 100 120 140
0 60
0.60 0.18
Air/Water, atm
0.16
0.50 C6/C7, 5 psia
IC4/NC4, 100 psia 0.14
IC4/NC4 165 psia
IC4/NC4,

dicted C sf, m/s

s
edicted C sf, ft/s
s

0.40 0.12
IC4/NC4, 300 psia
IC4/NC4, 400 psia 0.1
0.30
0.08

Pred
Pre

0.20 0.06

0.04
0.10
0.02

0.00 0
0 10 20 30 40 50 60
2
Liquid rate, gpm/ft

gxc/FRI AIChE 2004 [100e] 27

Packing
g Efficiencyy or HETP
– FRI has obtained semi-empirical models for packing
HETP
– Mainly a function of packing specific area
– Crimpp angle
g also affects efficiencyy
– Other factors: bed length, physical properties, m,
(mV/L), hump, etc.

gxc/FRI AIChE 2004 [100e] 28

Effect of packing area on HETP for a low pressure system
CAPACITY FACTOR
FACTOR, m/s
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18
8 2.4
Specific Area
2 3 2 3
7 ft /ft / m /m 2.1
15/50
6 1.8

5 15
1.5

HETP , m
HETP , ft

30/100
4 1.2

3 0.9
60/200
2 0.6
90/300
1 0.3
182/600
0 0.0
0 0.1 0.2 0.3 0.4 0.5 0.6
CAPACITY FACTOR, ft/s

gxc/FRI AIChE 2004 [100e] 29

Effect of hump on HETP for a system at 300 psia
4 00
4.00

3.50 Predicted

Measured
3.00

2.50
HETP, fft

2.00

1.50

1.00

0.50

0.00
0.000 0.050 0.100 0.150 0.200 0.250 0.300

Capacity Factor (Cs),

(Cs)

gxc/FRI AIChE 2004 [100e] 30

Optimization Examples:

• Optimization results by using FRI models:

– Vacuum: O/P Xylene, 2 psia
– Low pressure: C6/C7, 24 psia
– Medium pressure: IC4/NC4, 165 psia
– High pressure: IC4/NC4, 300 psia
• Designed
g at 80%
% of flood
• Focus on capacity and efficiency only
• Tray
Tra spacing specified at 24 inches

gxc/FRI AIChE 2004 [100e] 31

• Capacity:
C i
– Minimize the column diameter (D) for a given
rate
• Efficiency:
– Minimize
Minimi e the HETP
• Both Capacity and Efficiency:
– Minimize
Mi i i the
th column
l volume
l (V) for
f a given
i rate
t
V = 3.14*HETP*D*D/4

gxc/FRI AIChE 2004 [100e] 32

O/P Xylene, 2 psia, L/V=1
Fl rates:
Flow t L=V=95
L V 95 klb/hr
klb/h
D, in Eo, % Hetp,in Dp, in Volume
Sieve 121 74 32 2.6/tray 6.0
Sieve, 2p 128 65 37 2.3/tray 7.84
Valve 122 76 31 3.4/tray 6.0
2” Pall
2 120 26 0 4/ft
0.4/ft 4 85
4.85
M250.Y 113 15 0.38/ft 2.47

Structured packing has great advantage in terms of HETP, pressure,

capacity for vacuum service.

gxc/FRI AIChE 2004 [100e] 33

C6/C7, 24 psia
Flow
l rates: V=L=370 klb/hr,
klb/h 1 pass tray

D, iin Eo,
D E % Hetp,
H t in
i Dp, iin
D Volume
V l
Sieve 157 100 24 2.8/tray 7.67
Valve
l 158 98 24 3.8/tray
/ 7.88
2” Pall 146 22 0.6/stage 6.32
M250.Y 136 13.4 0.5/stage 3.20
N4T 125 28 1.1/stage 5.67

gxc/FRI AIChE 2004 [100e] 34

C6/C7, 24 psia, Just Sieve Tray
Flow
l rates: V=L=370 klb/hr
klb/h

Passes D, in Eo, % Hetp, in Dp, in Volume

1 157 100 24 2.8/tray 7.67
2 152 97 24 3 2/tray
3.2/tray 7 42
7.42
3 144 91 26 3.8/tray 7.09
4 152 87 28 3 7/stage
3.7/stage 8 23
8.23

A 3-pass tray gives smallest column size and volume.

gxc/FRI AIChE 2004 [100e] 35

IC4/NC4, 165 psia
Flow
l rates: V=L=700 klb/hr,
klb/h 3 pass trays

D, iin Eo,
D E % Hetp,
H t in
i Dp, iin
D Volume
V l
Sieve, 3p 155 111 22 4.5/tray 6.75
Valve,
l 3p 160 112 22 4.5/tray
/ 7.16
2” Pall 153 19 0.4/stage 6.1
M250.Y 148 18, hump 0.6/stage 5.14
N4T 133 24 5.51

gxc/FRI AIChE 2004 [100e] 36

IC4/NC4, 300 psia
Flow
l rates: V=L=420 klb/hr
klb/h
D, in Eo, Hetp, in Dp, in Volume
%
Sieve, 1p, ST 135 118 20 4.0/tray 4.83
Sieve 2p,
Sieve, 2p ST 128 113 21 4 0/tray
4.0/tray 4 51
4.51
Sieve, 3p, ST 121 105 23 6.2/tray 4.32
Sieve 4p,
Sieve, 4p ST 110 103 23 6 2/stage
6.2/stage 3 65
3.65
2” Pall 120 20 0.2/stage 3.74
M250.Y 113 19, hump 0.3/stage 3.25
N4T 104 24, hump 0.35/stage 3.35

gxc/FRI AIChE 2004 [100e] 37

Conclusions
– Optimum column designs are possible if good
models are available.
– Structured packing has great advantage in vacuum
service, but uncertain for high pressure system due
t HUMP.
to HUMP
– Tray and random packing have similar capacity for
high pressure system
system.
– Other factors not considered in this presentation also
affect the choice of internals and column designs.
designs

gxc/FRI AIChE 2004 [100e] 38