Packed Column
Packed Column
ABSTRACT
1. Objective
A Packed column was applied in mass transfer processes such as gas
absorption, distillation, and liquid extraction. These experiments were to investigate
the hydrodynamic characteristics and the effects of the gas and liquid velocities on the
operating performance of the packed column. In the packed column, the inlet liquid is
distributed over the top of the packed section and flows downward in the column,
while the inlet gas enters the distributing space below the packed section and flows
upward through the interstices in the packing countercurrent to the flow of the liquid.
Through the experiments, the students could determine:
1) The effects of gas and liquid velocities on the pressure drop of gas flow in the
packed column
2) The variation of the friction coefficient of dried – packing (fdry) as a function
Reynolds number (Re) of inlet gas flow rate and inferring the experimental relations
3) The change of “σ” factor, showing the relationship between the pressure drop of the
gas flow through the dried packing and that through the wetted packing in different
inlet liquid velocities
4) The diagram of limiting operation performance of packed column (loading and
flooding points)
2. Experiment method
Gas flow is supplied with a different flow rate to packed columns containing
ceramic chocks. Pressure drop is surveyed in turn when only gas flow goes into
column (dried column) and when gas flow goes into column combined with liquid
flow goes down from the top with different flow rates (wetted column)
3. Result data
L L
∆Pdr
(gal/ (kg/m ∆Pwet
y
min) 2/s) G%
L= L= L= L= L= L= L= L=
L=0 L=1
0.2 0.4 0.6 0.8 1.2 1.4 1.6 1.8
0 0
10 1 1 1 1 1 2 2 4 5 6
20 2 4 3 3 4 5 8 12 13 16
30 3 5 5 6 8 11 15 20
40 6 8 8 9 15 18 24
50 9 11 11 13 19 26 37
60 12 16 16 19 26
70 17 20 21 24 34
80 21 24 26 31
90 24 28 32
100 29 33 41
II. THEORY
1. Pressure drop of inlet gas flow
The Pressure drop (∆Pdry) of the gas flow through the packed column is
dependent on the gas velocity (G) in the condition of dried – packing (no
countercurrent-liquid flow). The pressure drop in the packed column increases as
raising the velocity of the inlet gas flowing through interstices in the packing
materials. The pressure drop increases with the 1.8 – 2.0 power of the velocity of the
gas.
When the liquid enters the column at the top of packing and flows downward, the
interstices in the packing are reduced, due to the occupation of the liquid in the free
volume of the packing material. Therefore, the displacement of the gas flow was
harder in the packed column. With an increment of gas flow rate, the experimental
effect of the liquid flow regularly increases. When the gas flow rate reaches a critical
value, the pressure drop rises gradually. If the gas flow rate exceeds this loading point,
the pressure drop increases more rapidly and does not follow the equation. In the local
regions of the column, the liquid becomes a continuous phase, and the column reaches
the flooding point. The entire column may fill with fluid at this flooding point.’
The graph of log(∆Pc/Z), describing the pressure drop per unit packing depth, is
shown in figure 1.
The pressure drop (∆Pdry) of gas flow through the packed column with no
countercurrent-liquid flow can be determined using Ergun equation:
Chilton and Colburn suggested a relation between pressure drop of gas flow
and gas velocity through the packed column
γh = 0.35
γw = 1
Friction factor fdry is a function of dimensionless Rec and is determined using the
following equation:
The correlation between the pressure drop for dried column ∆Pdry and that for wetted column
∆Pwet can be described as following
∆Pwet = σ∆Pdry
fwet = σ fdry
σ = 10ΩL
or log σ = ΩL
The value of σ depends on types, dimension, packing mode (random or regular mode) and
liquid velocity L.
For example, with the Raschig rings for 12.7 mm, random packing, the porosity ε of 0.568,
the liquid velocity L from 0.39 to 11.7 kg/m2s, the column is operated below the loading
point: Ω = 0.084
∆𝑝𝑤𝑒𝑡
From some previous studies, the relation between the ratio of ∆𝑝𝑑𝑟𝑦
and coefficient of fluid
For A < 0.3 and ceramic packing material with d < 30 mm:
4. The flooding point of packed column
When the packed column is flooded, the liquid seizes the packed section entirely, and
the flows are disordered drastically. This phenomenon is atrocious for the operation of the
packed column. The GL which is responsible for this flooding situation is called GL*.
Zhavoronkov pointed out that the flooding situation appears when the two numbers of
π1 and π2 for the packed column have the relationship.
And
v: the flow rate of gas before feeding to the packed column, m/s
µtđ: relative viscosity of liquid compared to water, if the liquid is water µtđ = 1
Therefore, the correlation of π1 and π2 as shown in the plot of log π1 – log π2
determined the flooding graph of the packed column. The limiting operation of the packed
column is under this plot.
Packing material which is Raschig rings is arranged randomly, with the nominal
diameter “d” = 12.7 mm, surface area “a” = 370 – 380 m2/m3, porosity “ε” = 0.586.
3. Experimental method
a. Preparation
- Close all of the liquid valves
- Turn on the air valve and switch the air blower in 5 min for removing all moisture in
the column. Then, shut down the blower
- Open valve “1” and valve “3”, and pump water into the bottom part of the column.
Using valve “2” to adjust the water level equal to the height of the level device (the
height = 3/4 of the bottom part of the column). Then, turn off the pump and close
valve “3”.
b. Measurement of pressure drop for dried column
- Close all of the liquid valves. Open air valve “6” when the air valve “5” is closed.
Turn on the air blower and open valve “5” slowly for adjusting the gas flow into the
column
- With a fixed gas flow rate, determining of Pdry on the pressure gauge for the mmH2O
unit. After that, turn off the blower and waiting for 5 min
c. Measurement for pressure drop of the wetted column
- Switch on the air blower and adjust gas flow rate into the column in range of 15 –
20% for maximum flow rate
- Open the liquid valve “1” and turn on the pump. Using valve “VL” from the liquid
flowmeter to adjust the water flow rate at the top of the column. If the float in
flowmeter does not pop up as the valve “VL” is opened at full volume, close the valve
“1” closely to raise the liquid flow rate
- At a specific liquid flow rate, altering the air flow rate and measuring the Pwet until
the operation of the column reaches the flooding point and accomplishing the
experiment
- Raising the liquid flow rate and conducting the experiment as above
Note:
- Adjust the valve “4” to maintain the height of water level equal to 3⁄4 the height of the
bottom part of the column through the measurement of the pressure drop
- Using the valve “2” to reduce water level from the column, if needed
- Turn off the water pump 𝐵𝐿 before opening the valve “4” at full volume. Finally,
- If the water infiltrates into the gas feeding pipe, opening the blue valve behind the
panel to eliminate water.
IV. RESULTS
CHART
Figure 13: Chart of effect of G (gas flow rate) on ΔP/Z (pressure drop per height unit)
and relative comparison among different flow rate values L
Figure 14: Chart of effect of LogG on Log(ΔP/Z) (pressure drop per height unit) and
relative comparison among different flow rate values L
Figure 15: Chart of the effect of Re on log fck and logfcö and relative comparison
among the different values L
Figure 16: Chart of the relation between Logσ and L at positions under loading point
(choosing position G = 10, 20, 30%)
y = 863,05x -124,57
ΔPck/Z followed G at L = 0
y = 960,38x - 110,56
ΔPcö/Z followed G at L = 0.2
y = 1155,1x - 171,29
ΔPcö/Z followed G at L = 0.4
y = 1140,8x - 138,47
ΔPcö/Z followed G at L = 0.6
y = 1472,2x - 156,83
ΔPcö/Z followed G at L = 0.8
y = 1632,8x - 137,81
ΔPcö/Z followed G at L = 1,0
y = 1632,8x - 137,81
ΔPcö/Z followed G at L = 1,2
y = 2141,4x - 93,429
ΔPcö/Z followed G at L = 1,4
y = 2141,4x - 93,429
ΔPcö/Z followed G at L = 1.6
y = 2676,7x - 93,429
ΔPcö/Z followed G at L = 1,8
y = -0,0011x + 1,7287
Logfck followed Re at L = 0
y = -0,0011x + 1,7287
Logfck followed Re at L = 0.2
y = 286,2x - 409,52
Logfck followed Re at L = 0.4
y = 286,2x - 409,52
Logfck followed Re at L = 0.6
y = -0,0011x + 1,7287
Logfck followed Re at L = 0.8
y = 286,2x - 409,52
Logfck followed Re at L = 1,0
y = -0,0011x + 1,7287
Logfck followed Re at L = 1,2
y = -0,0011x + 1,7287
Logfck followed Re at L = 1,4
y = -0,0011x + 1,7287
Logfck followed Re at L = 1.6
y = -0,0011x + 1,7287
Logfck followed Re at L = 1,8
V. DISCUSSION
After the loading point, ΔP rises rapidly. Operating the packed column in
bubble mode, the optimal condition, becomes challenging in this region. To avoid
flooding, experiments are usually stopped before reaching this point.
The chart relates the friction factor (f) to the Reynolds number (Re), expressing
the influence of fluid flow rate on the resistance force. As the flow rate increases, f
also increases. The chart helps determine suitable flow rates to minimize resistance
and maximize mass transfer efficiency without causing flooding. However,
experimental observations suggest a contrary trend, likely due to experimental errors.
Given either f or Re, the chart can be used to determine the other value by
drawing lines and finding intersections.
3. Does the relation between the analyzed objectives follow the prediction? If not,
explain?
The observed relationship between the objectives is fairly close to the predicted
one:
- Fluctuations in liquid and gas flow rates due to pump and fan instability.
- Insufficient water level in the column's bottom section, causing flooding in the
differential pressure tube and affecting results.
- Errors in reading measurements and experimental setup.
- Variations in experimental conditions between measurements.
Note:
VI. APPENDIX
2
∆𝑃𝑐𝑘.ε .ρ𝐾.𝐷𝑒
1. Calculation 𝑓 : 𝑓 = 2
𝑐𝑘 𝑐𝑘 2.𝐺 .𝑍
2. Calculation 𝑓 = σ. 𝑓𝑐𝑘
𝑐𝑜
𝐺𝐷𝑒 4𝐺
3. Calculation 𝑅𝑒 = εµ
= 𝑎µ
𝑐𝑘
4. Calculation σ: ∆𝑃 = σ. ∆𝑃𝑐𝑘
𝑐𝑜
3
V = 0.286 𝑚 /𝑝ℎ
2 2
π𝑑 3,14.0.09 2
F – across area of the packed column = 4
= 4
= 0. 0064 (𝑚 )
( ) 𝑓𝑐𝑘.𝑎 ρ𝐺
2
𝑣 0.2
∏ = 3 × 2𝑔
× ρ𝐿
× µ𝑡𝑑
ε
1
𝐿 ρ𝐺
∏ = 𝐺
× ρ𝐿
2
Where:
𝐺
𝑣= 𝐿
– Long velocity of the gas stream before entering the column, m/s
µ𝑙
µ𝑡𝑑 = µ𝑛𝑢𝑜𝑐
– Relative viscosity of liquid compared to water
VII. REFERENCES
1. Treybal, R.E., “Mass Transfer Operation”, Third Edition, pages 187 – 201,
McGraw-Hill Book Company, New York, 1980.
2. Bennett, C.O., and J.E. Myers, “Momentum, Heat and Mass Transfer”, Third
Edition, pages 187 – 201, McGraw-Hill Book Company, New York, 1982.