SCHOOL OF CHEMICAL ENGINEERING

THE UNIVERSITY OF ADELAIDE

2017

CHEM ENG 3036: UNIT OPERATIONS LABORATORY

SEMESTER 2, 2017

SP2: GAS ABSORPTION

Group P1
Tien Dat Le a1730614

Project date: 24th August 2017

Due date: 7th September 2017
Abstract

The aim of this experiment is to investigate the dependence of mass transfer coefficient on gas
and liquid rate, which was conducted in counter-current packed column dumped with ceramic
intalox saddles. The application of packed column is used widely in such operation such as
distillation, stripping and gas absorption. In this experiment, CO 2 is contact with water through a
counter-current packed column. The relationship between the overall mass transfer coefficient
and liquid was observed in a form of power, which was compared to Sherwood and Holloway
correlation. The experimental results have a similar trend with theoretical one but some of data
did not exhibit expected behavior. These errors could be explained by the measurement
uncertainty 5.88%, which have some impacts on the accuracy of final results. There are some
recommendations and modification of the apparatus to reduce the errors in this experiment.
Table of Contents
ABSTRACT ........................................................................................................................................... 2
1. INTRODUCTION ............................................................................................................................. 5
1.1. AIM .............................................................................................................................................. 5
1.2. GAS ABSORPTION .......................................................................................................................... 5
1.3 PACKED COLUMN........................................................................................................................... 5
1.4 MASS TRANSFER ............................................................................................................................ 6
1.5 HENRY’S LAW................................................................................................................................ 8
1.6 THEORETICAL MASS TRANSFER COEFFICIENT ................................................................................. 8
2. EXPERIMENTAL METHOD ........................................................................................................ 11
2.1 EQUIPMENT ................................................................................................................................. 11
2.2 EXPERIMENTAL PROCEDURE ........................................................................................................ 12
2.2.1 Start up ................................................................................................................................ 12
2.2.2 Data collection ..................................................................................................................... 13
2.2.3 Shut-down ............................................................................................................................ 13
3. RESULTS ........................................................................................................................................ 14
4. DISCUSSIONS ................................................................................................................................ 17
4.1 Significant results .................................................................................................................... 17
4.2 Experimental and theoretical mass transfer coefficient ............................................................ 17
4.3 Flooding ................................................................................................................................. 17
4.4 Packing ................................................................................................................................... 17
4.5 Sources of errors ..................................................................................................................... 18
4.6 Suggested improvements .......................................................................................................... 18
5. CONCLUSIONS .............................................................................................................................. 19
6. NOMENCLATURE ........................................................................................................................ 20
7. REFERENCES ................................................................................................................................ 21
APPENDICES ..................................................................................................................................... 22
APPENDIX A. RAW DATA ........................................................................................................... 22
APPENDIX B. SAMPLE CALCULATION ................................................................................... 23
1. Convert CO2 flow rate from L/min to m3/s.................................................................................. 23
2. Calculate volumetric flow rate of CO2 ....................................................................................... 23
3. Calculate number of moles CO2 transferred............................................................................... 23
4. Calculate log mean driving force and Kx.a ................................................................................ 24
APPENDIX C. SMAMPLE CALCULATION OF KX.A ERROR.................................................. 25
1. Error in gas velocity.................................................................................................................. 25
2. Error in volumetric flow rate ..................................................................................................... 25
3. Error in number of moles CO2 transferred ................................................................................. 25
4. Error in log mean driving force ................................................................................................. 25
5. Error in mass transfer coefficient Kx.a ...................................................................................... 26
APPENDIX D. PROCESSED DATA OF KX.A ERROR ................................................................ 27
APPENDIX E. CALIBRATION CHART FOR WATER............................................................... 28
APPENDIX F. SHERWOOD CORRELATION ............................................................................ 29
TABLE OF CONTENTS - FIGURES
FIGURE 1. COMMON DUMPED PACKINGS .......................................................................................6
FIGURE 2. GAS PRESSURE DROP FOR COUNTER-CURRENT GAS AND LIQUID IN RANDOM PACKINGS ...9
FIGURE 3. EXPERIMENTAL EQUIPMENT ....................................................................................... 11
FIGURE 4. GRAPH OF KX.A VS LIQUID RATE ................................................................................ 15
FIGURE 5. GRAPH OF KX.A VS GAS RATE .................................................................................... 16
FIGURE 6. GRAPH OF KX.A (EXPERIMENTAL AND THEORETICAL) VS LIQUID RATE ....................... 16

TABLE OF CONTENTS -TABLES

TABLE 1.THE LIST OF EQUIPMENT AND DESCRIPTIONS ................................................................. 11
TABLE 2. OTHER EQUIPMENT...................................................................................................... 12
TABLE 3. RESULTS ..................................................................................................................... 14
TABLE 4. RAW DATA .................................................................................................................. 22
TABLE 5. PROCESSED DATA OF KX.A ERROR ............................................................................... 27
1. Introduction
1.1. Aim
The aim of this experiment is to investigate the relationship between the mass transfer coefficient
based on gas phase, Kx, and the flow rate of liquid and gas through a counter-current packed
column. It is expected that the number of moles of solute transferred to liquid phase will increase
as the flow rate of liquid increased for a set of gas flow rate.

1.2. Gas absorption

Gas absorption is an operation in which a gas mixtures is contacted with liquid with the aim of
transferring one or more species from the gas phase into liquid phase. However, this process
does not change the gas properties. The reverse process is called desorption. Physical and
chemical absorption are two types that are used widely in absorption operation. Physical
absorption depends on solubility of a particular gas with respect to the liquid mixture and
volatility of the liquid while chemical absorption relies on chemical reaction between a particular
gas and the reagent in the liquid.

There are two methods for gas and liquid contacting in operation namely counter-current and co-
current operation. Regarding to the counter-current operation, gas come in from the bottom of
packed column while liquid enter on the top one. In contrast, co-current operation is a process
that gas and liquid enter on the packed column with the same direction. In this experiment, gas
(CO2) would be absorbed by liquid (pure water) in a counter-current packed column.

1.3 Packed column

The application of packed column is used widely in such operation such as distillation, stripping
and gas absorption. In general, using packed column allows two phases to get in contact with one
another and to encourage the mass transfer between immiscible fluids. This equipment consists
of a cylindrical column which have been filled with packing in packed bed, a gas and liquid inlet
at the bottom and top, respectively.

There are some characteristics of packing that should be considered in order to maximize the
transfer rate per unit volume of the column. Firstly, packing have to provide a large interfacial
surface between two phases. Secondly, it allows large volume of liquid go through the column
without loading and flooding. Good wetting characteristic, corrosion resistant and low pressure
drop for the gas are also worth to take into consideration.

Generally, packings are divided into three types: those with being randomly packed into the
column, those with being arranged in order or structured packings and those are known as using
an open-lattice structure. In the past, random packings are made from cheap materials such as
broken stone, clay or porcelain.
In this experiment, dumped packings are used for the purpose of this practical. It is important to
know that different packing materials could cause a negative change in mass transfer coefficient.
Hence, this characteristic should be selectivity based on particular gas and liquid. Ceramic
intalox saddle are used in this experiment and other common dumped packings are shown in
figure below.
 Figure 1. Common dumped packings

The packing size could have some influences on the cost of packing and the mass transfer
efficiency. Generally, an increase in the size of packing would reduce the cost of packing per
unit volume and the mass transfer efficiency. When mass transfer efficiency is reduced, a taller
of column is required. Hence, the overall column cost would not be reduced by increasing the
packing size. Furthermore, increasing packing size would increase the pressure drop. Therefore,
it is important to find out the optimum size of packing to reduce pressure drop, the cost and
increase the mass transfer efficiency. According to Coulson and Richardson, as the column is
dumped randomly, the packing size should not be over one-eighth of the column diameter.

1.4 Mass transfer

The transfer of solute across a differential area of gas-liquid phase could be described by
equation
dN  K y y  y * dA   K x x  x* dA (1)

where
 N: the number of moles of solute transferred (mole) (mole)
 Kx: the overall mass transfer coefficient for the gas (mol.m-2.s-1)
 Ky: the overall mass transfer coefficient for the liquid (mol.m-2.s-1)
 x: solute mole faction in liquid phase
 x*: solute mole faction in liquid phase at equilibrium stage
 y: solute mole faction in gas phase
 y*: solute mole faction in gas phase at equilibrium stage
 A: the interface area (m2)
The given total number of mole transferred can be considered by equation below
dN  d (Lx) (2)
where
 L: the flow rate of liquid phase (mol.s-1)

If the change in the liquid molar flow rate is not significant, equation (2) becomes
dN  Ldx (3)

Combining equation (1) and (3) could give

 K x ( x  x * )dA   Ldx (4)

The differential area could be defined

dA  S.a.dh (5)

Combining equation (4) and (5) could give

K x .( x*  x).S.a.dh  L.dx (6)

Integrating from 0 to Z (the packing height) and x1 to x2 lead to expression (7)

Z  x*  x 
K x  S  a   ln  * 2  (7)
L  x  x1 
where
 S: the cross-sectional area of the column (m2)
 a: the wetted area of packing (m2.m-3)

The total number of moles transferred may be calculated as

N  L( x1  x2 ) (8)

Rearranging L from equation (7) and substituting that into equation (8) gives
K y  S  a  x1  x 2 
N  K x  S  a  Z  x LM (9)
 x 2 
ln  
 x1 
Where
 xLM : log-mean driving force

It is important to know that column construction and the conditions of operating process have
some influences on the wetted area of packing. For absorption of pure gas, the resistance of gas
flow could be ignored. Therefore, kxA will equal to Kxa (kx is the liquid film coefficient)
𝐾𝑥 × 𝑎 = 𝑘𝑥 × 𝐴 (10)
Carbon dioxide and water are used in this experiment. Before entering the packed column, pure
carbon dioxide has to be flowed through a saturator to make sure that it did not pick up any
additional water (saturated). Then, measuring the time for bubble travel between two fixed point
could determine the gas inlet and outlet flow rate in bubble meter. Assuming that there is no
pressure drop across the column, the number of moles of carbon dioxide that transferred from
gas to water is calculated by following the equation of ideal gas law.
𝑃
𝑁 = 𝑅×𝑇 × (𝑄𝑖𝑛𝑙𝑒𝑡 − 𝑄𝑜𝑢𝑡𝑙𝑒𝑡 ) (11)

Where
 Q: the gas flow rate (m3/s)
1.5 Henry’s law

It is necessary to get the basis understanding of Henry’s law that would help to quantify the
driving force of mass transfer. This experiment was conducted at room temperature and pressure.
Hence, the absorption of carbon dioxide into water could be followed by Henry’s law. By using
equation below, we can calculate the liquid mole fraction with the vapor phase for a system at
equilibrium.
𝑦𝑃 = 𝐻𝑥 ∗ (12)
Where
 y: carbon dioxide mole fraction in the gas phase
 P: pressure of carbon dioxide
 H: Henry’s constant
 x*: carbon dioxide mole fraction in the liquid phase at equilibrium system

Because Henry’s constant is a function of temperature, it is important to know the temperature of

the gas entering the packed column. By interpolation, the value of Henry’s constant for CO2 at
289K is 1268 atm/mole.frac

1.6 Theoretical mass transfer coefficient

From the experiment, the relationship between mass transfer coefficient of gas (CO 2) and water-
carbon dioxide system is observed in the form of power relationship. For the liquid phase,
Sherwood and Holloway correlation for packed two phase contactors is calculated by the
equation below
𝐿 0.5 𝑙𝑏
𝐻𝐿 = 𝑎𝐿 (𝜇 )𝑛 × 𝑁𝑆𝑐,𝐿 , 𝐿 = ℎ𝑟×𝑓𝑡 2 . (13)
𝐿

Sherwood and Holloway correlation was also developed for absorption counter-current in a
packed column dumped with random packings. The system involving dilute-liquid water is
related to correlation given below
𝐿0.7 ×𝜇 0.3
𝐾𝑥 × 𝑎 = (14)
𝑎𝐿 ×𝑆𝑐 0.5
where
  𝑎𝐿 is a constant value of 0.0067. Assuming that the ceramic intalox saddles has the same
construction and properties as ½ inch Berl saddles. Hence, this value might be used in
this investigation
 𝜇𝐿 is a viscosity of water, taking the value of 0.00112 Pa.s at standard temperature and
pressure
 𝑆𝑐 is Schmidt number with the value of 0.94 for CO2 at standard temperature and
pressure
With assumption that ceramic intalox saddles and ½ inch Berl saddles are similar in terms of
construction and properties, equation (14) can be simplified into:
𝐾𝑥 × 𝑎 = 1.434326 × 𝐿0.7 (15)

For most random packings, pressure drop and flooding point are influenced by the gas and liquid
flow rate. Sherwood, Carman/Kozeny-Darcy, and Ergun have made some significant
contributions to predict these behaviors in packed column. While it might go beyond the scope of
this experiment, it is valuable to know the basic understanding of packed column behaviors.

For a fixed gas velocity through the packed column, as the liquid flow rate increases, the
interface available between two immiscible fluid become smaller, leading to an increase of the
gas-pressure drop. In the region figure 2 below A, the liquid holdup is constant regardless of
changing the flow rate of gas. Concerning the region between A and B, a rapid increase of the
liquid holdup with the gas rate would result in the reduced free area for gas flow and the pressure
drop increase more quickly. This phenomenon is known as loading.

Figure 2. Gas pressure drop for counter-current gas and liquid in random packings
When continuing to increase the gas flow rate, the liquid holdup would increase within the
packed column. At certain level (depending on the method of packing and settling of the
packing), the free cross available for gas flow become so small that liquid could be considered as
continuous and gas as discontinuous. This is knowns as phase inversion. There would be an
appearance of bubble from the bottom of the column that flow upward through a continuous
liquid phase. The pressure drop then changes quickly. This phenomenon is referred to as
flooding. In this case, the gas rate should be reduced to re-established the steady and continuous
the gas and liquid flow.

It is therefore important to know the limit of fluid flow in packed column. A correlation chart
(Appendix E) based on laboratory measurements and a correlation equation are developed by
Sherwood, Carman/Kozeny-Darcy, and Ergun as given below:
𝑣𝐺 2 𝑎 𝑝 𝐿 𝑝 0.5
× 𝜙3 × 𝑝𝐺 vs 𝐺 × ( 𝑝𝐺 ) (13)
𝑔 𝐿 𝐿
Where
 g: gravitational constant with value 9.81 (m/s2)
 a: packing surface area per unit volume (m2 m-3)
 pG and pL: gas and liquid density, respectively (kg/m3)
 v: superficial velocity (m/s)
2. Experimental method

2.1 Equipment
Below are the diagram of experimental arrangement and the descriptions of equipment

Figure 3. Experimental equipment

Table 1.The list of equipment and descriptions

Equipment Descriptions
Supply tank (T1) -Supply water to the system
Centrifugal pump (PM1) -Pump water to header tank
Header tank (T2) -Supply a constant source of water to packed column
Valve (V1) -Control flow rate of water
-Manufactured by Comet Cigweld
Valve (V2) -Control flow rate of CO2
-Manufactured by Comet Cigweld
Rotameter (FM1) -Allows and measures variable water flow rate to system.
 -Manufactured by KDG flow meters
-Accuracy ±0.5mm
Rotameter (FM2) -Allows and measures variable gas flow rate to system
-Manufactured by GEC-Elliott Process Instruments Ltd
-Accuracy ±0.5L air/min @150C
Saturator (S1) -Saturate the gas before entering packed column
-Manufactured by QVF Process Systems, De Dietrich & Co
Bubble meter (FM3) -Measures the gas inlet flow rate used with stopwatch
- Attached with reservoir containing detergent
-Glass column with internal diameter: 26mm
Bubble meter (FM4) -Measure the gas outlet flow rate used with stopwatch
-Attached with reservoir containing detergent
-Glass column with internal diameter: 26mm
Packed column (PC) -Glass absorption column: absorb pure CO2 into water
-Diameter column: 38.1 mm and length: 1170mm
-Packing materials: 1/2inch ceramic Intalox saddles
-Packing weight: 1.181kg
Manometer (P2) -Measures the pressure drop across the column
-Accuracy ±1mm
Pressure regulating gauge (P1) -Regulate the pressure of CO2 before entering the system

Table 2. Other equipment

Equipment Descriptions
Stopwatch -Record the time travel of bubble in 2 bubble meter column
Thermocouple (A, B) -Measure the temperature of water and gas before entering the
packed column
Ruler -Measure the differences between two fixed point in bubble
meter column

2.2 Experimental procedure

2.2.1 Start up

1. Ensure that the CO2 control valve was closed

2. The CO2 delivery pressure was approximately 100kPa using regulator to control
4. An additional regulator is placed on the rear of the board to measure the pressure of CO2
entering the apparatus, which should be about 20kPa
5. Ensure that the supply tank was filled with pure water before doing experiment
6. Make sure that the timer for water pump was switched on all the time doing experiment in
order to supply water to header tank.
7. The inlet and outlet bubble-meters was required to wet before using.
2.2.2 Data collection

1. Open slowly the control valve water and using rotameter 1 to control the flow rate of water at
5cm.
2. The CO2 isolation valve was slowly open to allow gas to the system. Using CO2 delivery
pressure regulator to set the pressure about 10-20kPa.
3. Open slowly the control valve CO2 and using rotameter 2 to control the gas flow rate at
0.5L/min.
4. To reach the steady condition in packed column, allowing flow rate of water and CO2 to run
about 3-5 minutes.
5. Elevating the reservoir to gas junction to produce flat-evenly bubble in two inlet and outlet
bubble column.
6. Using digital stopwatch to record the time for a bubble to travel the length of the bubble
column.
7. The inlet temperature of CO2 and water was taken using the thermocouple
8. The pressure of CO2 leaving saturator was recorded by the U-tube manometer.
9. This experiment was repeated with several combinations of flow rate of wter and CO2,
following the same procedure from step 1 to step 8. For the 1 st set, water flow rate was fixed at
5cm and CO2 flow rate was increased from 0.5L/min to 2.0L/min with an increment of 0.3L/min.
Repeated experiment with the same set for gas flow rate and increase water flow rate with
increment of 5 cm.

2.2.3 Shut-down

1. Close the water and CO2 isolation valve

2. Ensure that the CO2 cylinder valve was closed
3. The water pump was switched off
4. Thermocouple was turned off
3. Results

In this experiment, there are some following assumptions in order to carry out this investigation:
 The atmosphere pressure in this practical was 1atm (101 325 Pa)
 At standard pressure and temperature, the compressibility factor (Z) for CO 2 is equal
to 1
 𝑎𝐿 constant value of ceramic intalox saddles is replaced by the value of ½ inch Berl
saddles
 Diffusion occurs instantaneously at the interface fluid
Temperature of CO2 before entering packed column and surrounding is at 16oC (289K).

The calculated of value overall mass transfer coefficient is shown in table below

Table 3. Results

Number of
Flow rate of Flow rate of gas moles Theoretical Kx.a
Kx.a (mol/m3.s)
liquid (m3/s) (m3/s) transferred (mol/m3.s)
(mol/s)
8.33E-06 1.53E-04 3.69 4.27
1.33E-05 1.55E-04 3.81 4.27
1.83E-05 1.59E-04 4.29 4.27
3.83E-06
2.33E-05 1.82E-04
2.83E-05 1.46E-04 3.10 4.27
3.33E-05 5.36E-05 0.60 4.27
8.33E-06 2.33E-04 4.67 7.05
1.33E-05 9.39E-05 1.07 7.05
6.33E-06 1.83E-05 2.38E-04 4.94 7.05
2.33E-05 2.38E-04 4.92 7.05
2.83E-05 2.40E-04 5.03 7.05
3.33E-05 2.62E-04 7.02 7.05
8.33E-06 3.32E-04 6.46 10.21
1.33E-05 3.48E-04 7.37 10.21
1.83E-05 3.62E-04 8.40 10.21
9.17E-06
2.33E-05 3.26E-04 6.17 10.21
2.83E-05 3.32E-04 6.44 10.21
3.33E-05 3.32E-04 6.42 10.21
8.33E-06 CO2 outlet flow insufficient for bubble formation
1.33E-05 4.22E-04 7.76 13.48
1.21E-05 1.83E-05 4.27E-04 7.95 13.48
2.33E-05 4.27E-04 7.97 13.48
2.83E-05 4.30E-04 8.12 13.48
3.33E-05 4.36E-04 8.39 13.48
The graph of overall mass transfer coefficient (Kx.a) against gas and water flow rate are given in
figure 4 and figure 5 below. A comparison between the experimental and theoretical values
against liquid rate is also given in figue 6 below.

Graph of Kx.a vs Liquid rate ( at different gas rate)

14.00

12.00 y = 6E+11x2.1799

y = 94975x0.8264
Kx.a (mol/m3.s)

10.00

y = 9969.5x0.6244
8.00 y = 4499x0.5479

y = 9552.2x0.6321 y = 242617x0.932
6.00

4.00

2.00

0.00
0.00E+00 2.00E-06 4.00E-06 6.00E-06 8.00E-06 1.00E-05 1.20E-05 1.40E-05

Liquid rate (m3/s)

gas rate= 0.5 L/min gas rate = 0.8L/min gas rate = 1.1L/min

gas rate= 1.4L/min gas rate= 1.7L/min gas rate= 2.0L/min

Power (gas rate= 0.5 L/min) Power (gas rate = 0.8L/min) Power (gas rate = 1.1L/min)

Power (gas rate= 1.4L/min) Power (gas rate= 1.7L/min) Power (gas rate= 2.0L/min)

Figure 4. Graph of Kx.a vs Liquid rate

 Graph of Kx.a vs gas rate (at different liquid rate)
9.00
8.00
7.00
Kx.a (mol/m3.s)

6.00
5.00
4.00
3.00
2.00
1.00
0.00
0.00E+00 5.00E-06 1.00E-05 1.50E-05 2.00E-05 2.50E-05 3.00E-05 3.50E-05
Gas rate (m3/s)

liquid rate= 5cm liquid rate= 10cm liquid rate= 15cm

liquid rate= 20cm Power (liquid rate= 5cm) Power (liquid rate= 10cm)
Power (liquid rate= 15cm) Power (liquid rate= 20cm )

Figure 5. Graph of Kx.a vs Gas rate

Graph of Kx.a (Experimental and theoretical) vs Liquid rate

16.00

14.00
Kx.a(mol/m3.s)

12.00

10.00

8.00

6.00

4.00

2.00

0.00
0.00E+00 2.00E-06 4.00E-06 6.00E-06 8.00E-06 1.00E-05 1.20E-05 1.40E-05

Liquid rate (m3/s)

Series1 Series3 Series4
Series5 Series6 theoretical line
Power (Series1) Power (Series2) Power (Series3)
Power (Series4) Power (Series5) Power (Series6)
Power (theoretical line)
Figure 6. Graph of Kx.a (Experimental and Theoretical) vs Liquid rate
4. Discussions

The data show that there are quite different between experimental mass transfer coefficient and
theoretical values themselves. However, the experimental values seem to show the same trend as
developed by Sherwood and Holloway with liquid rate. According to Sherwood and Holloway
correlation, the overall mass transfer coefficient and liquid rate can exhibit power relationship
but there is no relationship between Kx.a and gas flow rate. There are some derived data that
exhibit unexpected behavior, leading to a fail in producing desired relationship when plotted
figure 4. The cause of this could be due to human malpractice or equipment, which is discussed
further in 4.5 sources of errors.

4.1 Significant results

It can be clearly seen in figure 4 that an increase of liquid flow rate for each set of gas flow rate
result in a higher Kx.a value. It also means that the total absorption of gas will increase when the
gas flow rate is increased. However, there are some points that did not show this trend due to
errors in this experiment which would be discussed later.

At constant liquid flow rate with different gas flow rate, the number of moles of solute absorbed
by liquid should be constant. This could be explained by the limit of saturation of liquid.
However, the figure 5 states differently. The graph of Kx.a against gas flow rate show the
exponential trend instead of linear (R2 value is small). This could be due to errors during doing
experiment.

4.2 Experimental and theoretical mass transfer coefficient

According to the figure 6, which is a graph of the value of overall mass transfer coefficient and
theoretical prediction against liquid flow rate, it might be said that the experimental values do no
match with the line predicted by Sherwood and Holloway correlation. However, these values are
quite close to theoretical one, except for value at liquid flow rate 20 cm. The relative uncertainty
in Kx.a t that is calculated in Appendix C is 0.1187.

4.3 Flooding

Flooding was occurred at the base of the column when setting the lower rate for gas and liquid.
To avoid this problem, the flow rate of liquid should be adjusted to zero before doing the next
experiment. It might be beyond the scope of this investigation when compared flooding point
with the values predicted by Sherwood correlation. It is noticeable that the flooding point
occurred at the bottom of the column instead of the top one. This unusual featured could be
explained by the way of packing.

4.4 Packing

Sherwood and Holloway correlation was using constant aL to calculate the mass transfer
coefficient. However, due to the presence of constant aL of ½ inch ceramic intalox saddles, it was
necessary to use the constant value of ½ inch Berl saddles instead. With assumption that ceramic
intalox saddles and ½ inch Berl saddles are similar in terms of construction, properties, surface
area per unit volume, calculation of mass transfer coefficient was enabled to be performed. It is
suggested that using accurate constant value of aL would minimize the error and increase the
credibility of experimental results.

4.5 Sources of errors

a) It is important to reach equilibrium mass transfer of the system before changing the flow
rate of the liquid. In this experiment, data was taken directly after another flow rate. This
could contribute to the inaccuracy in measuring the time of travel bubble and gas velocity
bubble in the bubble meter.
b) It is hard to control water flow rate and gas flow rate by using rotameter. The values are
not constant and fluctuated while readings are taken.
c) The fluctuation was also seen on monometer. This could contribute to error in measuring
the number of moles of CO2 transferred to liquid phase and the overall mass transfer
coefficient.
d) Bubble meter might be one of the major sources of errors in this experiment. Regarding
to the reservoir, when it was raised to reach to the gas junction to produce the flat-evenly
bubble, reservoir should be kept at stable. Otherwise, it could affect the velocity of
bubble travelling upwards. In addition, there were more than one formation of bubble in
the bubble meter. Hence, it could block and decrease the speed of bubble up as well.
Furthermore, dust inside the column might contribute to the inaccuracy in determining
the bubble velocity. Concerning to the flow rate of gas, when it was set lower than
1L/min, there was no formation of bubble inside the column. Therefore, there were no
data recorded, leading to a lack of data to analyze.
e) This experiment required human to set up equipment and operate during the practical. So,
errors made by human were unavoidable. Stopwatch was used to measure the time of
bubble traveled between two fixed point in the bubble meter. It is impossible for human
to start and stop stopwatch consistently and read consistently two marked point in bubble
column.
f) The explanation for the flooding at the base of column might be that the packing could
have broken into small pieces over the time. According to Sherwood, flooding Would
occur in regions of lower porosity. If it happened, the void space at the bottom of the
column would be reduced, leading to flooding.

4.6 Suggested improvements

a) A pressure gauge should be used instead of U-tube manometer in order to avoid the
fluctuation while taking the reading data.
b) Bubble meter should be definitely replaced by mechanical devices to avoid human errors,
which is discussed in sources of errors.
c) It is important to check the condition of apparatus before doing experiment to define
potential errors.
5. Conclusions

a) There was a consistency that Kx.a value will increase if the flow rate of liquid is
increased with a set of gas flow rate.
b) There is no relationship between Kx.a and gas flow rate as suggested by Sherwood and
Holloway.
c) The average uncertainty is around 5.88%
d) There are some recommendations and modification of the apparatus to reduce the errors
in this experiment.
6. Nomenclature
7. References
APPENDICES

APPENDIX A. RAW DATA

Table 4. Raw data

Liquid Gas flow rate

Pressure difference Time inlet Time outlet
flow rate (L/min@150C
(mmH20) (s) (s)
(cm) 760mmHg)
0.5 18 24.91 45.42
0.8 26 16.62 25.26
1.1 37 12.2 17.1
5
1.4 48 10.21 14.17
1.7 60 8.53 11.02
2.0 78 7.54 8.9
0.5 18 25.65 71.46
0.8 26 16.76 22.5
1.1 36 12.2 19.4
10 1.4 46 10.27 15.34
1.7 58 8.53 12.11
2.0 74 7.57 10.66
0.5 20 26.28 221.2
0.8 28 16.43 42.96
1.1 36 12.2 24.57
15
1.4 50 10.19 17.2
1.7 64 8.5 13.35
2.0 74 7.46 11.18
0.5 CO2 outlet flow insufficient for bubble formation
0.8 32 16.35 58.5
1.1 38 13.79 37.37
20
1.4 50 9.9 19.27
1.7 62 8.57 15.27
2.0 76 7.32 12.1
APPENDIX B. SAMPLE CALCULATION
The table below would be used to show the calculation of processed data.

Gas flow rate Pressure

Liquid flow rate Time inlet Time outlet
(L/min@150C difference
(cm) (s) (s)
760mmHg (mmH20)
5 0.5 18 24.91 45.42

1. Convert CO2 flow rate from L/min to m3/s

𝐿 10−3 𝑚3
𝑄𝐶𝑂2 = 0.5 𝑚𝑖𝑛 = 0.5 × × =8.33E-06 (m3/s)
60 𝑠

2. Calculate volumetric flow rate of CO2

𝐷2 0.0262
𝜋× ×𝐿𝑖𝑛𝑙𝑒𝑡 𝜋× ×0.45
𝑄𝐶𝑂2(𝑖𝑛𝑙𝑒𝑡) = 4
= 4
= 9.59E-06 (m3/s)
𝑡𝑖𝑛𝑙𝑒𝑡 24.91
𝐷2 0.0262
𝜋× ×𝐿𝑜𝑢𝑡𝑙𝑒𝑡 𝜋× ×0.51
𝑄𝐶𝑂2(𝑜𝑢𝑡𝑙𝑒𝑡) = 4
= 4
= 5.96E-06 (m3/s)
𝑡𝑜𝑢𝑡𝑙𝑒𝑡 45.42

Where
 D: internal diameter of bubble column (0.026m)
 Linlet and Loutlet: the length of 2 point in inlet and outlet bubble column (0.45m and 0.51m)
 tinlet and toutlet : time of travel of bubble in inlet and outlet bubble column (24.91s and
45.42s)
 QCO2inlet and QCO2outlet: volumetric flow rate of CO2 (m3/s)

3. Calculate number of moles CO2 transferred

𝑘𝑔 𝑚 18
𝑃𝑎𝑏𝑠 = 𝑃𝑎𝑡𝑚 + (𝜌𝐻2𝑂@298𝐾 × 𝑔 × 𝐻) = 101325 + 998.9 × 9.81 × 𝑚
𝑚3 𝑠 2 1000
= 101501.4(𝑃𝑎)
Where
 𝜌: density of H2O at 298K (998.9 kg/m3)
 𝐻: pressure differences (18mmH20)
Number of moles CO2 transferred:
𝑃 × ∆𝑄 𝑃𝑎𝑏𝑠 × (𝑄𝑐𝑜2(𝑖𝑛𝑙𝑒𝑡) − 𝑄𝐶𝑂2(𝑜𝑢𝑡𝑙𝑒𝑡) )
𝑁= =
𝑅×𝑇 𝑅×𝑇

=(101501.4Pa x ( 9.59E-06 5.96E-06 ))/(8.31 x 289)=1.53E-04 (mol/s)

Where
 R =8.31 (Pa.m3/K.g.mol)
 G=9.81 (m/s2)
4. Calculate log mean driving force and Kx.a

From calibration chart for water at 5cm, flow rate of water is approximately about 0.23 L/min
Convert water flow rate from L/min to m3/s
10−3 𝑚 3
𝑄𝑤𝑎𝑡𝑒𝑟 = 0.23 × = 3.83E-06 (m3/s)
60 𝑠

Next, calculate L which is water flow rate (mol/s)

𝑚3 𝐾𝑔 𝑔
𝑄𝑤𝑎𝑡𝑒𝑟 ×𝜌𝐻2𝑂@289𝐾 3 ×1000 3.83×10−6 ×998.9×1000
𝑠 𝑚 𝑘𝑔
𝐿= 𝑔 = = 0.213 (mol/s)
𝑀𝑊𝐻20 18.016
𝑚𝑜𝑙

As number of moles CO2 transferred and water flow rate are known, using equation (8) to
calculate the difference of solute mole fraction in liquid phase from the bottom to top of tower:
𝑁 1.53×10−4
𝑥1 − 𝑥2 = = =7.183E-04
𝐿 0.213

Because water from the top of the tower is pure water, so x2=0. Hence
𝑥1 − 𝑥2 = 𝑥1 =7.183E-04

At equilibrium, carbon dioxide mole fraction in the liquid phase is calculated by using Henry’s
law:
𝑦×𝑃
𝑦𝑃 = 𝐻𝑥 ∗ ⟺ 𝑥 ∗ =
𝐻
Because CO2 entering the system is pure, so y=1
Henry constant is a function of temperature. Hence, we can calculate H@16 oC B by interpolation
10oC 1040 atm/mol.frac
16oC x
o
20 C 1420 atm/mol.frac

From interpolation, H289K= 1268 atm/mol.frac

𝑃 1
𝑥∗ = = = 8 × 10−4
𝐻 1420

Then, calculating log mean driving force

𝑥1−𝑥2 𝑥1 7.183×10−4
∆𝑥𝐿𝑀 = 𝑥∗ −𝑥 = 𝑥∗ = 8×10−4
= 3.12 × 10−4
ln( ∗ 2 ) 𝑙𝑛 ∗ 𝑙𝑛
𝑥 −𝑥1 𝑥 −𝑥1 8×10−4−7.183×10−4

Finally, calculating Kx.a

𝑚𝑜𝑙
𝑁 1.53×10−4
𝐾𝑥 × 𝑎 = 𝑆 (𝑚2)×𝑍(𝑚)×∆𝑥 𝑠
= 0.03812
= 3.69 (mole/m3.s)
𝐿𝑀 𝜋× ×1.17×3.12×10−4
4
APPENDIX C. SMAMPLE CALCULATION OF Kx.a ERROR

The table below would be used to show the calculation of error.

Liqui Distance Distance

Gas flow rate Pressure
d flow Time inlet Time outlet for inlet for outlet
(L/min@150 difference
rate (s) (s) bubble bubble
C 760mmHg (mmH20)
(cm) column column
Uncertainty Uncertainty Uncertainty
Uncertainty Uncertainty
: ±2 : ±0.5 mm : ±0.5 mm
: ±0.1 s : ±0.1 s
mmH20
5 0.5 18 24.91 45.42 450 mm 510 mm
1. Error in gas velocity
For the error in inlet gas velocity
𝛿𝑣𝑖𝑛𝑙𝑒𝑡 𝛿𝑆𝑖𝑛𝑙𝑒𝑡 𝛿𝑡𝑖𝑛𝑙𝑒𝑡 0.5 𝑚𝑚 0.1 𝑠
= + = + = 5.27 × 10−3
𝑣𝑖𝑛𝑙𝑒𝑡 𝑆𝑖𝑛𝑙𝑒𝑡 𝑡𝑖𝑛𝑙𝑒𝑡 450 𝑚𝑚 24 𝑠

For the error in outlet gas velocity

𝛿𝑣𝑜𝑢𝑙𝑒𝑡 𝛿𝑆𝑜𝑢𝑡𝑙𝑒𝑡 𝛿𝑡𝑜𝑢𝑡𝑙𝑒𝑡 0.5 𝑚𝑚 0.1 𝑠

= + = + = 3.18 × 10−3
𝑣𝑜𝑢𝑡𝑙𝑒𝑡 𝑆𝑜𝑢𝑡𝑙𝑒𝑡 𝑡𝑜𝑢𝑡𝑙𝑒𝑡 510 𝑚𝑚 45.42 𝑠

2. Error in volumetric flow rate

𝛿𝑄𝑖𝑛𝑙𝑒𝑡 𝛿𝑣𝑖𝑛𝑙𝑒𝑡 𝛿𝐴 𝛿𝑣𝑖𝑛𝑙𝑒𝑡
= + = = 5.27 × 10−3
𝑄𝑖𝑛𝑙𝑒𝑡 𝑣𝑖𝑛𝑙𝑒𝑡 𝐴 𝑣𝑖𝑛𝑙𝑒𝑡

𝛿𝑄𝑜𝑢𝑡𝑙𝑒𝑡 𝛿𝑣𝑜𝑢𝑡𝑙𝑒𝑡 𝛿𝐴 𝛿𝑣𝑜𝑢𝑡𝑙𝑒𝑡

= + = = 3.18 × 10−3
𝑄𝑜𝑢𝑡𝑙𝑒𝑡 𝑣𝑜𝑢𝑡𝑙𝑒𝑡 𝐴 𝑣𝑜𝑢𝑡𝑙𝑒𝑡

∆𝑄 𝛿𝑄𝑖𝑛𝑙𝑒𝑡 𝛿𝑄𝑜𝑢𝑡𝑙𝑒𝑡
= − = 5.27 × 10−3 − 3.18 × 10−3 = 2.09 × 10−3
𝑄 𝑄𝑖𝑛𝑙𝑒𝑡 𝑄𝑜𝑢𝑡𝑙𝑒𝑡

3. Error in number of moles CO2 transferred

𝛿𝑁 𝛿𝑃 𝛿∆𝑄 𝛿𝑅 𝛿𝑇 𝛿𝑃 𝛿∆𝑄 2
= + + + = + = + 2.09 × 10−3 = 0.1132
𝑁 𝑃 ∆𝑄 𝑅 𝑇 𝑃 ∆𝑄 18

4. Error in log mean driving force

𝛿(𝑥1 − 𝑥2 ) 𝛿𝑁 𝛿𝐿
= + = 0.1132
𝑥1 − 𝑥2 𝑁 𝐿

As x2=0, this can be simplified into

𝛿𝑥1
= 0.1132
𝑥1
∆𝑥2 1 ∆𝑥2 1 𝛿𝑥 ∗ 𝛿𝑥 ∗ 𝛿𝑥1 1 𝛿𝑥 ∗ 𝛿𝑥1
𝛿𝑙𝑛 [ ]= ×𝛿[ ]= × [ ∗ + ∗ + ]= × [2 ∗ + ]
∆𝑥1 ∆𝑥 ∆𝑥1 ∆𝑥 𝑥 𝑥 𝑥1 𝑥∗ 𝑥 𝑥1
[ 2] [ 2] [ ∗
𝑥 − 𝑥1
]
∆𝑥1 ∆𝑥1

Given x1=7.183E-04 and x*=8.0E-04

∆𝑥
𝛿𝑙𝑛 [∆𝑥2 ] =0.0126
1

5. Error in mass transfer coefficient Kx.a

∆𝑥2 ∆𝑥2
𝛿 (𝐾𝑥 𝑎) 𝛿𝑁 𝛿𝑆 𝛿𝑍 𝛿𝑥1 𝛿𝑙𝑛 [∆𝑥1 ] 𝛿𝑁 𝛿𝑙𝑛 [∆𝑥1 ]
= + + + + = + = 0.1132 + 0.0055 = 0.1187
𝐾𝑥 𝑎 𝑁 𝑆 𝑍 𝑥1 ∆𝑥2 𝑁 ∆𝑥2
𝑙𝑛 [∆𝑥 ] 𝑙𝑛 [∆𝑥 ]
1 1
Absolute certainty in Kx.a is calculated below:
𝛿𝐾𝑥 𝑎 = 0.1187 × (𝐾𝑥 𝑎) = 0.1187 × 3.69 = 0.438

Percentage error=0.438/(0.438+3.69) x 100 = 10.6%

APPENDIX D. PROCESSED DATA OF Kx.a ERROR

Table 5. Processed data of Kx.a error

Relative Absolute Percentage

Flow rate Flow rate Theoretical
Kx.a Uncertainty Uncertainty error
of liquid of gas Kx.a
(mol/m3.s) in Kx.a in Kx.a (%)
(m3/s) (m3/s) (mol/m3.s)
(mol/m3.s) (mol/m3.s)
𝛿(𝐾𝑥 𝑎) 𝛿(𝐾𝑥 𝑎)
𝐾𝑥 𝑎
8.33E-06 3.69 4.27 0.1186 0.437 10.60
1.33E-05 3.81 4.27 0.0846 0.322 7.80
1.83E-05 4.29 4.27 0.0620 0.266 5.84
3.83E-06
2.33E-05
2.83E-05 3.10 4.27 0.0416 0.129 4.00
3.33E-05 0.60 4.27 0.0333 0.020 3.22
8.33E-06 4.67 7.05 0.1192 0.557 10.65
1.33E-05 1.07 7.05 0.0841 0.090 7.76
6.33E-06 1.83E-05 4.94 7.05 0.0642 0.317 6.04
2.33E-05 4.92 7.05 0.0523 0.257 4.97
2.83E-05 5.03 7.05 0.0436 0.219 4.18
3.33E-05 7.02 7.05 0.0365 0.256 3.52
8.33E-06 6.46 10.21 0.1090 0.704 9.83
1.33E-05 7.37 10.21 0.0808 0.596 7.48
1.83E-05 8.40 10.21 0.0653 0.549 6.13
9.17E-06
2.33E-05 6.17 10.21 0.0496 0.306 4.73
2.83E-05 6.44 10.21 0.0412 0.265 3.95
3.33E-05 6.42 10.21 0.0371 0.238 3.58
8.33E-06 CO2 outlet flow insufficient for bubble formation
1.33E-05 7.76 13.48 0.0725 0.563 6.76
1.21E-05 1.83E-05 7.95 13.48 0.0628 0.499 5.91
2.33E-05 7.97 13.48 0.0505 0.403 4.81
2.83E-05 8.12 13.48 0.0430 0.349 4.12
3.33E-05 8.39 13.48 0.0373 0.313 3.60
APPENDIX E. CALIBRATION CHART FOR WATER
APPENDIX F. SHERWOOD CORRELATION