Mass transfer study using an electrochemical method

Proceedings of European Congress of Chemical Engineering (ECCE-6)

Copenhagen, 16-20 September 2007

Mass transfer study using an electrochemical method

M.A. Cancela*, R. Maceiras, E. lvarez, X.R. Nvoa
University of Vigo, Chemical Engineering Department, ETSEI; Rua Maxwell s/n, 36310 Vigo, Spain.

Abstract
The volumetric mass transfer coefficients at different axial positions in a bubble column
were measured using a conductivity method. The absorption process of CO2 in
monoethanolamine solutions was carried out in a prismatic column 0.06 m side and 1.03 m
height. Measurements were conducted at various gas flow rates and amine concentration. The
local mass transfer was found to vary significantly in the axial direction.
Keywords: Local mass transfer coefficient, electrochemical method, carbon dioxide,
monoethanolamine, bubble column.

1. Introduction
Many processes in the field of chemical engineering are related to mass transfer
operations between different phases, generally gas and liquid phases. One of the most
important is the removal of acid gases, such as H2S and CO2 from natural and industrial
exhauts using alkanolamine solutions. Among the industrially important alkanolamines, the
primary amines such as the monoethanolamine (MEA) are the most employed.
More stringent regulations governing the emission of toxic gases from industrial sources
have evoked an increasing demand for new and efficient abatement techniques. In the last
years, different electrochemical methods have been used to study the mass transfer, measuring
the mass transfer coefficient in different equipments (Kleifges et al., 1997; Vilar and Coeuret,
1999; Oduoza and Wragg, 2000; Coueret, 2001; Gostick et al., 2003; Verma and Rai, 2003;
Arzutug et al., 2005; Dang-Vu et al., 2006a,b; Han and Al-Dahhan, 2007).
In this work, the application of an electrochemical method to directly measure local mass
transfer coefficients during the removal of carbon dioxide using monoethanolamine aqueous
solutions is discussed. This method overcomes some of the shortcomings inherent to other
techniques since it permits mass transfer coefficients to be directly determined from the

A. Cancela et al.

measurement of local conductivity changes. Two requirements must be considered in order to

analyze the absorption process by means of conductivity measurements. On the one hand the
thorough contact between the gas and the absorbent liquid (to transfer the reactant from the
gas phase to the liquid phase); and on the order hand the thorough contact between the liquid
phase and the metallic electrode (which drives the necessary electric current through the
system). Considering these requirements, a bubble column with electrodes distributed along
the reactor walls was designed. This approach can be easily extended in future work to
investigate the effect of other factors.

2. Theory
Monoethanolamine (MEA) solution is an important solvent in the CO2 removal process
because it reacts quickly with carbon dioxide for its primary amine characteristics. The
monoethanolamine, H2N-CH2-CH2-OH, is a primary amine in which the nitrogen atom is
|

bonded to an ethanol group, and both the amine group ( N H ) and the alcohol group (OH)
can react with CO2 to produce carbamic acid (Eq. 1) and carbonic acid (Eq. 2) (Astarita et al.,
1964; Clarke, 1964; Hart, 1976; Hikita et al., 1977a,b; lvarez-Fuster et al., 1980; Deckwer,
1980; Camacho et al., 2000; Dhotre et al., 2005; Verma, 2005):
NH 2 + CO 2
NHCOO H +
OH + CO 2
OCOO H +

(1)
(2)

The carbonic acid formation may take place in alkaline solution at a pH11.
Nevertheless, for pH<10, even in a carbonate solution, the formation of the carbonic-acid
derivative can be considered negligible (Astarita et al., 1964). Measurements indicate that, in
our experiments, the pH was less than 11 from the beginning of the absorption process and
decreased gradually during the process. On the basis of the results obtained, the main reaction
can be considered to be the formation of a carbamic-acid derivative (Eq. 1).
On the other hand, the reaction regime between CO2 and MEA solutions is instantaneous
(Clarke, 1964; Hikita et al., 1977a,b; lvarez-Fuster et al., 1980). In this case, the rate of
absorption of carbon dioxide per unit volume, RA, is computed from the following expression:

D B C Bo
R A = k L a C *A 1 +
*
z DA CA

Therefore, the volumetric mass transfer coefficient, kLa, can be obtained as follows:

(3)

A. Cancela et al.

k La =

RA

D B C Bo
C *A 1 +
*
z DA CA

(4)

where CA* is the gas solubility (kmolm-3), z is the stoichiometric coefficient, DA and DB are
the diffusitives of CO2 and MEA in aqueous monoethanolamine solutions, respectively,
(m2s1) and CBo is the bulk initial concentration of the amine (kmolm3).

3. Experimental

3.1. Materials and methods

The experimental setup is shown in Fig. 1. A bubble column was used in this work for
the absorption of CO2 into aqueous monoethanolamine solutions. Absorption measurements
were performed at room temperature, operating in batches with respect to the liquid phase.
The bubble column is made of methacrylate, 1.03 m height, and has a square cross-section
(side length 6 cm). For the injection and uniform distribution of the gas phase, a gas sparger,
i.e., a porous plate of 4 mm in diameter is installed at the centre of the bottom plate. This plate
has another orifice for liquid outlet. There are also three orifices at the top plate: liquid inlet,
gas outlet and a thermometer.
Each experimental run was started by filling the column with the liquid phase up to 100
cm above the sparger. Carbon dioxide, saturated with water vapour at ambient temperature
(22 2 C), was fed through the bottom of the bubble column. The gas flow, before entering
the bubble column, was metered by a flow meter and controlled with a flow controller
(Brooks model 0154). The gas pressure was measured in the inlet using a digital manometer
(Testo model 512). The gas flow in the outlet was measured with a soap meter. The gas
absorption rate was calculated as the difference between the flow rates gas into and out of the
bubble column.
The electrical circuit consists of a function generator working at 1 kHz (sine wave)
connected through a switching box in series with a standard resistor and each of the electrode
pair in the column. The 1 kHz frequency was chosen from previous calibration in order to
overcome the electrodes polarization resistance. A high impedance AC voltmeter measures
the current flux through the standard resistor, which is inversely proportional to the ohmic
drop for each electrode pair. The geometry of the electrodes is chosen so that voltmeter
readings correspond directly to electrolyte resistivity (in cm).
Local mass transfer coefficients were determined under different conditions at 5 axial

A. Cancela et al.

levels in the column. The axial levels were 22 cm apart, starting 10 cm above the gas
distributor. At each axial level, two AISI316 stainless steel sheets were arranged in two
opposite faces to be used as the sensing electrodes. The lateral faces of the steel sheets were
coated with cataphoretic paint so that only the front part of the electrode contacts the
electrolyte. The steel sheets were fixed to the column by steel screws also insulated with the
cataphoretic paint.
Aqueous monoethanolamine (MEA) solutions of different concentrations were employed
as liquid phase, while the gas phase was carbon dioxide. Different gas flow rates for each run
were selected. The amine concentrations were varied between 0 and 1.0 M. Gas flow rates
were varied between 10 and 25 L/h to study the effect of the gas flow rate on mass transfer
coefficients. In all experiments the volume of liquid has been constant and equal to 3.6 L.

Figure 1. Schematic figure of the bubble column reactor: (1) pressurized CO2 source, (2) humidifier, (3) flowmeter / regulator, (4) digital manometer, (5) bubble column, (6) stainless steel (SS316) electrodes, (7) electronic
and electrical equipment for local resistance measurement, (8) thermometer, (9) output flow-meter.

3.2. Physical Properties

The interpretation and correlation of the experimental mass-transfer data require
knowledge of the physical properties of the liquid phases as well as the gas solubilities and
diffusivities in these phases.
The densities, , and viscosities, , were measured at 20, 25 and 30 C using a Anton

A. Cancela et al.

Paar DSA 5000 densimeter, with a precision of 105 gcm3, and a Shott-Gratte AVS 350
automatic viscometer, with a precision of0.01s, respectively. The experimental values were
correlated simultaneously with the amine concentration. The values are reported in Maceiras
et al. (in press). The solubilities and diffusivities of the CO2 and MEA were calculated using
the correlation equations found in the literature (Wilke and Chang, 1955; Danckwerts, 1970;
Sada et al., 1978; Versteeg et al., 1987; Versteeg and Van Swaaij, 1988).

4. Results and Discussion

Mass transfer coefficients can be obtained trough conductivity measurements because

CO2 reacts with the MEA producing ionic species. All the details about the data acquisition
are reported in Maceiras et al. (in press).
The plot in Fig. 2 shows the variation of the absorption rate (obtained trough
conductivity measurements) with the time, at different axial positions. It can be observed that
in the beginning the absorption process occurs initially at the bottom of the column while in
the upper part absorption begins only about 20 minutes later. This fact, evident from Fig. 2
data is due the reaction regime: when CO2 enters the column, the reaction occurs
instantaneously and the amount of carbon dioxide that rises in the column lies out of the
detection limit.
18
H = 0.10 m
H = 0.32 m
H = 0.54 m
H = 0.76 m
H = 0.98 m

16

RA10 (mol / Ls)

14
12
10
8
6
4
2
0
0

1000

2000

3000

4000

5000

6000

t (s)

Figure 2. Variation of the absorption rate with the time for MEA 0.3 M and 25 Lh1.

Once the absorption rate has been obtained, the local volumetric mass transfer coefficients
were calculated using the Eq. (4) for all the amine concentrations and gas flow rates. The
effect of the gas flow rate is presented in Fig. 3, were the Reynolds number is employed to
quantify flow rate. Thus, the effect of the electrode-to-sparger distance (H/d) is explored. As
expected (from Fig. 2 data), the mass transfer coefficients decrease considerably with

A. Cancela et al.

increasing (H/d) and kLa values increase with the Reynolds number at a given axial position.
6.0
Re= 37.19
Re= 55.78
Re= 74.37
Re= 92.97

5.0

-1

kLa10 (s )

4.0
3.0
2.0
1.0
0.0
0

10

15

20

(H/d)

Figure 3. Local mass transfer coefficient for different Re and amine concentration of 1 M.

5. Conclusions

In this work, an electrochemical technique has been used to measure the local mass
transfer coefficients. The results obtained suggest that the assumption of a constant mass
transfer coefficient throughout the bubble column is erroneous and may lead to important
errors in designing and scaling up bubble columns, since the local mass transfer coefficient
varies with the axial position.

References

lvarez-Fuster, C., Midoux, N., Laurent, A., Charpentier, J.C., (1980) Chemical Engineering
Science, 35, 1717-1723.
Arzutug, M. E., Yapici, S., Muhtar Kocakerim, M., (2005) International Communications in
Heat and Mass Transfer, 32, 842854.
Astarita, G., Marrucci, G., Gioia, F., (1964) Chemical Engineering Science, 19, 95-103.
Camacho, F., Snchez, S., Pacheco, R., (2000) Chemical Engineering & Technology, 23,
1073-1080.
Clarke, J.K.A., (1964) Industrial & Engineering Chemistry Fundamentals, 3, 239-245.
Coeuret, F., (2001) Journal of Applied Electrochemistry, 31, 193-199.
Danckwerts, P.V., Gas Liquid Reactions. McGraw-Hill, London (1970).
Dang-Vu, T., Doan, H.D., Lohi, A., (2006a) Industrial & Engineering Chemistry Research,

A. Cancela et al.

45, 1097-1104.
Dang-Vu, T., Doan, H.D., Lohi, A., Zhu, Y., (2006b) Chemical Engineering Journal, 123,
8191.
Deckwer, W.D., (1980) Chemical Engineering Science, 35, 1341-1346.
Dhotre, M.T., Vitankar, V.S., Joshi, J.B., (2005) Chemical Engineering Journal, 108, 117125.
Gostick, J., Doan, H. D., Lohi, A., Pritzker, M. D. (2003) Industrial & Engineering Chemistry
Research, 42, 3626-3634.
Han, L. M., Al-Dahhan, H., (2007) Chemical Engineering Science, 62, 131-139.
Hart, W.F., (1976) Industrial & Engineering Chemistry Process Design and Development, 15,
109-114.
Hikita, H., Asai, S., Ishikawa. H., Honda, M., (1977a) Chemical Engineering Journal, 13, 712.
Hikita, H., Asai, S., Ishikawa. H., Honda, M., (1977b) Chemical Engineering Journal, 14, 2730.
Kleifges, K-H., Kreysa, G., Juttner, K., (1997) Journal f Applied Electrochemistry, 27, 10121020.
Maceiras, R., Cancela, M.A., lvarez, E., (in press) Chemical Engineering Journal.
Maceiras, R., Nvoa, X.R., lvarez, E., Cancela, M.A., (in press) Chemical Engineering &
Processing.
Oduoza, C.F., Wragg, A.A., (2000) Journal of Applied Electrochemistry, 30, 1439-1444.
Sada, E., Kumazawa, H., Butt, M.A., (1978) Journal of Chemical & Engineering Data, 23,
161-163.
Verma, A.K., (2005) Industrial & Engineering Chemistry Research, 41, 882-884.
Verma, A.K., Rai, S., (2003) Chemical Engineering Journal, 94, 67-72.
Versteeg, G.F., Blauwhoff, P.M.M, van Swaaij, W.P.M., (1987) Chemical Engineering
Science, 42, 1103-1109.
Versteeg, G.F., Van Swaaij, W.P.M., (1988) Journal of Chemical & Engineering Data, 33,
29-34.
Vilar, E.O., Coeuret, F., (1999) The Canadian Journal of Chemical Engineering, 77, 855-862.
Wilke, C.R., Chang, P., (1955) AIChE Journal, 1, 264-270.