Article

Cite This: Ind. Eng. Chem. Res. 2017, 56, 14902−14913 pubs.acs.org/IECR

Kinetics of CO2 Absorption in Aqueous Hexamethylenediamine

Blended N‑Methyldiethanolamine
Bikash K. Mondal,† Syamalendu S. Bandyopadhyay,‡ and Amar N. Samanta*,†
†
Chemical Engineering Department, Indian Institute of Technology Kharagpur, Kharagpur 721302, India
‡
Cryogenic Engineering Centre, Indian Institute of Technology Kharagpur, Kharagpur 721302, India

ABSTRACT: Carbon dioxide (CO2) absorption kinetics in the

aqueous blend of hexamethylenediamine (HMDA) and N-
methyldiethanolamine (MDEA) is studied at the temperatures
varying from 303 to 333 K using the pressure decay technique in
a reaction calorimeter setup. For the experimental study, HMDA
concentration is varied in the range of 5−15 mass % keeping total
amine (HMDA + MDEA) concentration at 30 mass %. Overall
rate constant for this reaction system is estimated assuming
pseudo-first-order reaction condition. CO2 absorption kinetics are
significantly enhanced due to the presence of HMDA as
compared to single MDEA solvent. The individual rate
contribution of CO2−HMDA and CO2−MDEA reaction systems
are combined to represent the overall CO2 absorption rate in this
mixed amine solvent. The kinetic models (I and II) developed using the zwitterion and termolecular mechanism for HMDA−
CO2 reaction system and base catalyzed hydration mechanism for MDEA-CO2 reaction are able to predict the kinetic data with
good accuracy.

1. INTRODUCTION as well as regeneration energy. It is found from the recent

The major global challenge in the coming years is to reduce the literature that aqueous diamine has a high CO2 absorption rate
carbon dioxide emission in view of its apparent contribution to and capacity compared to the conventional amines. Some of the
the global warming. Fossil fuel-based power plants release major diamines which are found in the recent literature are piperazine
amounts of CO2 in the atmosphere; 40% of the world’s electricity (PZ)11 and its derivatives,12,13 2-((2-aminoethyl)amino)ethanol
is generated by burning coal which contributes about 44% of the (AEEA),14 hexamethylenediamine (HMDA),15−17 1,4-butanedi-
total CO2 emission worldwide.1 To mitigate the atmospheric amine (BDA), 1 8 ethylenediamine (EDA), 1 9 and 3-
CO2 emission, immediate implementation of the postcombus- (methylamino)propylamine (MAPA).20 These diamines can
tion capture and clean coal technology is essential. For the rapid potentially improve the solvent characteristics to a greater extent
capture of CO2 from large scale industrial gas streams, amine compared to the conventional MDEA blends because of their
scrubbing is the most developed and widely used technology. superior kinetic and loading capacity. Literatures are available on
Aqueous solutions of monoethanolamine (MEA) and diethanol- the PZ promoted MDEA solvent21−27 which show enhanced
amine (DEA) are the conventional amines used for the CO2 CO2 absorption kinetics and loading capacity of this activated
capture.2 But, the use of these amines for CO2 capture from solvent. CO2 absorption study in the aqueous (MDEA +
power plant flue gases leads to a substantial energy penalty due to MAPA)28 and (MDEA + AEEA)29 also indicate potential of the
high regeneration energy requirement to handle large volumetric diamine as better solvent activator. In our earlier works16,30 it is
flow rates of the flue gas stream having 10−15% (v/v) CO2 shown that HMDA has very high CO2 absorption rate (k2,HMDA:
content.3 So the CO2 capture research is mainly aimed at 59190 m3 kmol−1 s−1 at 313 K) and loading capacity (CO2
developing a novel solvent with higher CO2 absorption rate and loading: 1.143 mol/mol at 15 kPa CO2 partial pressure and 313 K
capacity and lower enthalpy of absorption. The higher loading temperature). These superior solvent properties of HMDA and
potential and low regeneration energy requirement of non- dearth of literature data motivated us to study the kinetic
carbamate forming amine N-methyldiethanolamine characteristics of aqueous (HMDA + MDEA) solvent. High CO2
(MDEA),4−7 has made it an attractive solvent for CO2 capture. loading potential31 and low absorption enthalpy32 of this blended
But the CO2 absorption rate with aqueous MDEA is very slow. solvent is already presented in our previous works. In this work,
So, this amine is mixed with another amine of higher CO2
absorption rate to improve the kinetic characteristics of the Received: July 5, 2017
solvent. Much literature is available on the MDEA blended Revised: November 20, 2017
conventional amine solvent such as (MDEA + MEA),8,9 (MDEA Accepted: November 21, 2017
+ DEA)10 to make the solvent more efficient in terms of kinetics Published: November 21, 2017

© 2017 American Chemical Society 14902 DOI: 10.1021/acs.iecr.7b02744

Ind. Eng. Chem. Res. 2017, 56, 14902−14913
Industrial & Engineering Chemistry Research Article

Figure 1. N2O solubility measurement setup: (1) Stirred-cell contactor; (2) buffer cell for N2O storage; (3) gas phase stirrer; (4) liquid phase stirrer
(magnetic bar); (5) magnetic stirrer speed controller; (6) pressure transducers; (7) circulator temperature controller; (8) N2O cylinder; (9)
temperature sensors; (10) liquid solution inlet; (11) external temperature sensor of circulator temperature controller; (12) thermostated water bath.

the kinetics of CO2 absorption in aqueous (HMDA + MDEA) solvent to obtain dynamic viscosity. Each measurement is
solvent is studied in the temperature range of 303−333 K for its quintuplicated, and the standard relative uncertainty is calculated
potential application as CO2 capture solvent. to be within ±1.5%.
2.4. N2O Solubility and Liquid Phase Mass Transfer
2. EXPERIMENTAL SECTION Coefficient Measurement. Liquid phase mass transfer
2.1. Materials. Reagent grade HMDA (minimum purity 98% coefficient and physical solubility data are required to analyze
by mass) and MDEA (minimum purity 99% by mass) are the kinetic behavior of the CO2−amine system. Because of the
purchased from Sigma-Aldrich India. Nitrogen, nitrous oxide, reactive characteristics of CO2 with the amine solution, the
and carbon dioxide with minimum purity of 99.99% (by volume) physical solubility is measured using N2O which does not react
are supplied by Linde India Limited. Aqueous amine solutions with aqueous amine and also has a similar molecular and
are prepared on a mass percentage basis without further electronic configuration as CO2.33,34 First, equilibrium N2O
purification of the chemicals using a precision balance solubility in the aqueous amine solution is measured in the
(CITIZEN, CX-301 model, accuracy: ±0.001g). experimental temperature range and then it is converted to
2.2. Density. The density of aqueous (HMDA + MDEA) is physical CO2 solubility using the method of N2O-analogy.
required to estimate molar concentration of the solvent from the 2.4.1. Experimental Procedure. Experimental setup (Figure
mass % concentration. It is measured using a standard Gay- 1) and procedure for the solubility measurement is presented in
Lussac pycnometer (∼25 × 10−6 m3 at 298 K). Before the our previous work16 and it is described here briefly. N2O
experiment, the volume of the pycnometer is standardized using solubility is measured in a stirred cell contactor (500 × 10−6 m3)
double distilled water. The density of the solvent is measured connected to a buffer cell (650 × 10−6 m3) through a needle
dividing the mass of the solvent contained in the pycnometer by valve. The temperature of the cells are maintained using a
the standard volume of the pycnometer at constant temperature. thermostated water bath (JULABO F 32 HL, FRG, accuracy:
To maintain the experimental temperature, a thermostated water ±0.1 K). For each experimental run, 200 × 10−6 m3 amine
bath (JULABO F32 HL, FRG) with a precision of ±0.1 K is used. solution at the desired temperature is kept in the stirred cell
Each experiment is repeated for at least three times and relative under vacuum condition for at least 1 h to record the vapor
standard uncertainty is calculated to be ±0.1%. pressure of the solvent. Then N2O gas from the buffer vessel at
2.3. Measurement of Viscosity. Viscosity data is required the desired temperature is transferred to the stirred cell (PN2O ≈
to estimate diffusivity of CO2 in the solvent. It is also an
important property for solvent pumping cost estimation. A 100 kPa) and stirrers (both gas phase and liquid phase) are
Cannon-Fenske viscometer (size, 50; viscometer constant (VC), started. Pressure decrease in the stirred cell indicated by the
0.004 cSt·s−1) is used to measure the kinematic viscosity (KV) of pressure transmitter (model, Rosemount 3051TA; range, 0−50
the solvent. KV (cSt) is estimated by multiplying the efflux time psia; accuracy, ±0.04% of the range) is recorded continuously.
of a specific volume of the solvent through the capillary of the Equilibrium condition is indicated when there is no change of cell
viscometer with VC (0.004 cSt·s−1) at constant temperature. A pressure for at least 1 h.
temperature controlled water bath (JULABO F32 HL, FRG) 2.4.2. Estimation of N2O Solubility. N2O solubility in
with a precision of ±0.1 K is used to maintain experimental aqueous amine solvent is estimated using the following
conditions. Then KV is multiplied by the density (g·mL−1) of the expression.33
14903 DOI: 10.1021/acs.iecr.7b02744
Ind. Eng. Chem. Res. 2017, 56, 14902−14913
Industrial & Engineering Chemistry Research Article

RT (PNi 2O − PNe 2O) Vg So eq 3 simplifies to the following.

m N2O, m = = .
HN2O, m PNe 2O Vl PCO2
(1) NCO2 = mCO2EkL
RT (5)
Here mN2O,m is the dimensionless physical solubility defined as
the ratio of equilibrium N2O concentration in the liquid to the Again, the flux can also be expressed as
gas phase. HN2O,m is Henry constant in the aqueous mixed 1 dnCO2 V dPCO2
NCO2 = − =− G
solvent, PNi 2O is initial N2O partial pressure in the cell, PeN2O is the A dt ART dt (6)
equilibrium partial pressure of N2O in the cell, Vg and Vl are gas Here, A, VG, T, and R are the mass transfer area, gas phase
phase and liquid phase volume in the cell. volume, temperature of the system, and universal gas constant,
2.4.3. Estimation of Liquid Phase Mass Transfer Coefficient. respectively.
Liquid phase mass transfer coefficient (kL) (in absence of Then, combining and rearranging eq 5 and eq 6 gives
chemical reaction) is estimated using the expression35 given
below. dPCO2 mCO2EkLA
=− PCO2
m N2OAkL dt VG (7)
ln PN2O|t = t = − t + ln PN2O|t = 0
Vg (2) In this expression, E is a function of infinite enhancement factor
(Ei) and Hatta number (Ha). Ei is the enhancement factor
where Vg and A are the gas phase volume and interfacial area of corresponding to the infinitely fast reaction of CO2 in the liquid
gas−liquid interaction, respectively. PN2O and mN2O,m denote N2O film when the absorption rate is limited by the diffusion process
partial pressure in the stirred cell and dimensionless physical only. Ei for a single irreversible reaction is given by,
solubility parameter in the mixed solvent, respectively. From the
DCO2 DAm [Am]RT
slope of the “ln PN2O versus time” plot, kL can be calculated. Since Ei = +
the physical absorption process (without chemical reaction) is DAm DCO2 νCO2PCO2mCO2 (8)
slow, slope is calculated using data up to 100 s.
In this equation DCO2 and DAm are the diffusivities of CO2 and
2.5. Carbon Dioxide Absorption Rate Measurement.
amine in the aqueous solution, [Am] is amine solution
The rate of CO2 absorption in aqueous (HMDA + MDEA)
solvent is measured using the “pressure decay method”. concentration, νCO2 is stoichiometric coefficient of CO2 in the
Formulation of the pressure decay expression and kinetic reaction.
measurement technique are as follows. Ha is the ratio of maximum reactive conversion rate of CO2 in
2.5.1. Derivation of the Pressure Decay Expression. the film to the maximum diffusional transport through the film.35
Absorption of CO2 in aqueous amine solution is a complex According to the film theory of mass transfer, E and Ha are
phenomenon involving phase equilibria and reaction kinetics. related as follows.
During the absorption process CO2 is dissolved in the liquid Ha
phase physically and then reacts with amine components forming E=
tanh(Ha) (9)
different ions. Because of the chemical reaction, the rate in the
chemical absorption is much faster compared to the purely The terms involved in eq 7 can be converted to measurable
physical absorption process. This improved absorption rate is quantity by assuming pseudo-first-order (PFO) reaction
characterized by the enhancement factor (E) which is the ratio of conditions. Criteria for the PFO reaction conditions is as
the chemical absorption flux to the physical absorption flux under follows.36
same driving force. Assuming film theory, CO2 absorption flux
3 < Ha ≪ E i (10)
(NCO2) in an aqueous amine solution can be presented as follows.
* ) With this assumption, E becomes equal to Ha. Then E can be
(PCO2 − PCO expressed as follows.
NCO2 = K G(PCO2 * )=
− PCO 2
2 1 RT
kG
+ mCO2EkL k OVDCO2
(3)
E = Ha =
with kL (11)
1 1 RT Replacing the enhancement factor with Hatta number
= +
KG kG mCO2EkL (4) expression, eq 7 becomes as follows:

where PCO2 is the CO2 partial pressure in the system, P*CO2 is the dPCO2 mCO2A k OVDCO2
=− PCO2
equilibrium CO2 partial pressure corresponding to the CO2 dt VG (12)
concentration in the bulk liquid, KG is the overall mass transfer
coefficient and mCO2 is the physical solubility of CO2 in aqueous This is the differential form of pressure decay expression
When eq 12 is integrated, the integral form of the pressure
amine solvent. kG and kL are the gas phase and liquid phase mass decay expression is obtained as shown below.37
transfer coefficient, respectively. The terms involved in the above
equation can be simplified using the following assumption. mCO2A k OVDCO2
• negligible CO2 concentration in the bulk liquid (and hence ln PCO2|t = t = − t + ln PCO2|t = 0
VG (13)
* 2 = 0)
PCO
where VG and A are the gas phase volume and interfacial area of
• negligible gas phase resistant (and hence 1 = 0).
kG gas−liquid reaction, respectively. PCO2, mCO2, DCO2, and kOV
14904 DOI: 10.1021/acs.iecr.7b02744
Ind. Eng. Chem. Res. 2017, 56, 14902−14913
Industrial & Engineering Chemistry Research Article

Figure 2. Schematic of CO2 absorption rate measurement set up. (1) double-jacketed glass reactor (reaction calorimeter, model, RC1e); (2) anchor
type impeller; (3) reactor temperature sensor; (4) calibrated heater (25W); (5) Rosemount pressure transmitter (model, 3051TA); (6) magnetically
coupled stirrer; (7) solvent inlet; (8) (a) N2 cylinder, (b) CO2 cylinder; (9) (a) N2 gas regulator, (b) CO2 gas regulator; (10) thermostated water bath;
(11) temperature sensor to control water bath temperature; (12) CO2 feed coil.

Figure 3. Pressure decay plot for CO2 absorption rate measurement (for 5% HMDA + 25%MDEA at 303 K).

denote the partial pressure of CO2, dimensionless physical calorimeter setup (Mettler Toledo; model, RC1e). The reactor
solubility, diffusivity of CO2, and overall reaction rate constant, of the calorimeter setup is a thermostated glass vessel (volume,
respectively. From the slope of the “ln PCO2 versus t” (eq 13), the 1.2 × 10−3m3) equipped with an impeller (anchor type), a digital
pressure transducer (model, Rosemount 3051TA; range, 0−15
overall rate constant for the CO2 absorption kinetics can be
psia; accuracy, ±0.04% of the pressure range), a calibrated heater
evaluated since all other parameters are known. (25 W) and a temperature sensor (Pt 100). The reactor is
2.5.2. Absorption Kinetics Measurement Setup and connected to a coil dipped in a temperature controlled water bath
Procedure. Details of the CO2 absorption rate measurement for feeding CO2 into the reactor at the desired temperature. The
setup (Figure 2) and experimental procedure are published in entire system is tested to make it leak-proof.
our earlier work16 which is described here briefly. CO 2 For rate measurement, 500 g of aqueous solution is heated to
absorption rate in this work is measured using a reaction the desired temperature under vacuumed condition for at least 1
14905 DOI: 10.1021/acs.iecr.7b02744
Ind. Eng. Chem. Res. 2017, 56, 14902−14913
Industrial & Engineering Chemistry Research Article

h and then vapor pressure (Pv) of the solvent is recorded. Once Here ηi, xi, and Gij denote dynamic viscosity of pure component i,
the reactor content attains thermal equilibrium, a small amount mole fraction of component ‘i’ in the mixture, and interaction
of CO2 heated to the desired temperature is transferred into the parameter between components i and j, respectively. For the
reactor. After feeding CO2 gas, the reactor content is stirred at ternary mixed solvent system viscosity is given by the following
150 rpm and real time data of the pressure decay inside the expression
reactor due to chemical absorption is logged continuously. This
pressure decay data is used to estimate the overall rate constant ln ηm = x1 ln η1 + x 2 ln η2 + x3 ln η3 + x1x 2G12 + x1x3G13
(kOV) by applying the pressure decay expression (eq 13). A
+ x 2x3G23 (21)
typical plot of “ln (PCO2/Pa) versus t/s” is shown in Figure 3.
From the initial slope of the plots, kOV is calculated. Initial CO2 3.3. Physical Solubility Model. The physical solubility of
partial pressures for all the experimental runs are kept low to N2O in aqueous (HMDA + MDEA) solution is modeled using
uphold the PFO condition. excess Henry’s coefficient model developed by Wang et al.40 The
excess Henry coefficient is defined as the excess quantity over the
3. MODEL sum of the product of the volume fraction and logarithm of
3.1. Density Model. For modeling the density data of individual Henry’s coefficient in the mixed solvent. Expression
aqueous (HMDA + MDEA), the excess molar volume expression for the excess Henry’s coefficient as defined by Wang is shown
proposed by Redlich and Kister38,39 is used. It is presented as below.
follows. n
R = ln HN2O, m − ∑ φi ln HN O,i
2
V E = Vm − ∑ xiVio i=2 (22)
i (14)
where HN2O,m, HN2O,i and φi denote the Henry coefficient of N2O
where Vm and Voi denote the molar volume of the mixed solvent in the mixed solvent, Henry coefficient of N2O in pure
and the molar volume of the pure component respectively at the component i ,and volume fraction of the component i in the
system temperature. Mixed solvent molar volume is calculated as mixed solvent, respectively. Volume fraction is calculated as given
follows. below.
∑i xiMi n
Vm = φi = xiV io/∑ xiV io
ρm (15) (23)
i=1

here, Mi and xi are molecular weight and mole fraction of Voi

xi and are the mole fraction and molar volume of component i
component i and ρm is the measured density of mixed solvent. in the mixture.
Excess molar volume for a binary mixed solvent system, can be Again, for a ternary mixed solvent system, the excess Henry
expressed as given below. coefficient is given by the following equation.
n
R = φφα
i j ij
+ φφ α + φφ
i k ik
α + φφφ
j k jk
α
i j k ijk (24)
E
V12 /m 3·kmol−1 = x1x 2 ∑ Ai (x1 − x 2)i
i=0 (16) Here φi, φj, and φk are the volume fraction of HMDA, water, and
Excess molar volume expression for the ternary mixed solvent MDEA in the mixed solvent, respectively. αijk is a three-body
system (HMDA + MDEA + H2O) is given by the following interaction parameter which is set to be constant for the ternary
equation. system studied. αij, αik, and αjk are two-body interaction
parameters which are assumed to be a function of temperature.
V E = V12
E E
+ V13 E
+ V23 (17) The temperature dependency of these parameters is expressed as
follows.
where Ai are interaction parameters. These are assumed to be the
function of temperature as given below. k2
αij = k1 +
a k3 + T (25)
Ai = i
T (18) Equation 22 and 24 are combined to form the model expression
3.2. Viscosity Model. The Grunberg and Nissan (1949) for the Henry’s coefficient (HN2O,m) which is fitted to the
model38,39 is used in this work to correlate the dynamic viscosity experimental data by regressing the coefficients of αij.
of the ternary (HMDA + MDEA + H2O) solvent with that of 3.4. Rate Model. The overall rate of CO2 absorption in the
pure component (HMDA, MDEA, and H2O) viscosity. They (HMDA + MDEA + H2O) mixed solvent is the combined rate
introduced a temperature-dependent interaction parameter to contribution of HMDA−CO2, MDEA−CO2, and water−CO2
account for the deviation in the viscosity due to nonideal mixing. reaction system. For the amine−CO2 reaction, there is three
The expression for this model is shown below. generally accepted mechanisms to analyze the kinetic behavior of
CO2 absorption. These are the zwitterion, termolecular, and base
ln ηm /mPa·s = ∑ xi ln ηi + ∑ ∑ xixjGij catalyzed hydration mechanisms. The rate contribution and
i i j (19) reaction mechanism of the individual reaction systems are
presented in the following section.
with 3.4.1. Reaction Rate Dependence on HMDA. In the HMDA
b2 molecule, two primary amine groups are connected through a six
Gij = b1 + carbon straight chain. It forms a carbamate ion upon reaction
(b3 + T ) (20) with CO2 in the aqueous medium. In this work, both zwitterion
14906 DOI: 10.1021/acs.iecr.7b02744
Ind. Eng. Chem. Res. 2017, 56, 14902−14913
Industrial & Engineering Chemistry Research Article

and termolecular mechanisms are used to analyze the kinetic rCO2 − HMDA = (k H2O[H 2O] + kMDEA[MDEA]
characteristics of the CO2−HMDA reaction.
Zwitterion Mechanism. The reaction steps using this + kHMDA[HMDA])[HMDA][CO2 ] (29)
mechanism are as follows.
3.4.2. Reaction Rate Dependence on MDEA. Since MDEA is
a tertiary amine, it undergoes base-catalyzed hydration reaction
with CO2. The mechanism for this reaction is shown below.

It is reported by many authors37,41,42 that the reaction of CO2

According to this mechanism, the reaction rate can be written as with aqueous MDEA follows a pseudo-first-order reaction
given below. kinetics. Rate expression for this reaction is as follows.
k 2,HMDA[CO2][HMDA] rCO2 − MDEA = k 2,MDEA[MDEA][CO2 ]
rCO2 − HMDA = (31)
k−1
1+ ∑ k b[B]
This second-order rate equation is used to represent the rate of
=
[HMDA][CO2 ] the CO2−MDEA reaction system.
1 1
+ k2,HMDAk H O 3.4.3. Reaction Rate Dependence on Water. Water interacts
k 2,HMDA 2 [H O] + k2,HMDAk MDEA [MDEA] + k2,HMDAk HMDA [HMDA]
2
k−1 k−1 k−1 with the CO2 by the following reactions
(27)
CO2 + H 2O ↔ H 2CO3 (32)
where Σkb[B] is the contribution of all bases (HMDA, MDEA,
OH−, H2O) for the deprotonation reaction. k2,HMDA and k−1 are CO2 + OH ↔ −
HCO−3 (33)
the forward and reverse rate constants for the zwitterion
formation reaction. kHMDA, kMDEA, and kH2O are the rate constants Since the rate constant of the CO2 hydration reaction (eq 32) is
very small (kH2O = 0.026 s−1 at 298 K),43 the contribution of this
for the deprotonation reaction. In this reaction system H2O,
MDEA, and HMDA are taken as deprotonating bases. The reaction to the overall rate is generally neglected. The
contribution of OH− ion is neglected because of low contribution of the bicarbonate formation reaction (eq 33) is
concentration. also neglected without significant loss of the accuracy in the
Termolecular Mechanism. The reaction in the termolecular reaction rate due to the low concentration of hydroxyl ion.37,41,42
mechanism proceeds through the formation of a loosely bound 3.4.4. Overall Reaction Rate of CO2 with (HMDA + MDEA +
encounter complex in a single step as shown below. H2O). The overall CO2 absorption rate in the (HMDA + MDEA
+ H2O) solvent, is given below.
HMDA + CO2 + B ↔ [encounter complex] rOV = rCO2 − HMDA + rCO2 − MDEA (34)
→ (HMDA)COO− + BH+ (28)
Kinetic Model I. On the basis of the zwitterion mechanism for
The rate expression using this mechanism where H2O, MDEA, HMDA−CO2 reaction system and base catalyzed hydration
and HMDA are the dominating bases (B), can be presented as mechanism for MDEA−CO2 reaction system, the overall rate
follows. expression becomes,

[HMDA][CO2 ]
rOV = k 2,MDEA[MDEA][CO2 ] + 1 1
= k OV[CO2 ]
k 2,HMDA
+ k2,HMDAk H O
2 k2,HMDAk MDEA k2,HMDAk HMDA
k−1
[H 2O] + k−1
[MDEA] + k−1
[HMDA] (35)

where, kOV is the overall rate constant for (HMDA + MDEA +

H2O + CO2) reaction system and it is given by the following
expression. where, kOV, the overall rate constant for (HMDA + MDEA +
H2O + CO2) reaction system, is given by the following equation.
k OV = k 2,MDEA[MDEA]
[HMDA] k OV = k 2,MDEA[MDEA] + (k H2O[H 2O] + kMDEA[MDEA]
+ 1 1
+
k 2,HMDA k2,HMDAk H O
2 [H O] + k2,HMDAk MDEA [MDEA] + k2,HMDAk HMDA [HMDA]
2
+ kHMDA[HMDA])[HMDA] (38)
k−1 k−1 k−1

(36)
Kinetic Model II. Using a termolecular mechanism for the 4. RESULTS AND DISCUSSION
HMDA−CO2 reaction system and a base-catalyzed hydration Physico-chemical properties and CO2 absorption rate in the
mechanism for the MDEA−CO2 reaction system, the rate HMDA-blended MDEA solvent is studied in this work at the
expression takes the following form. temperatures 303, 313, 323 and 333 K. Solvent compositions
used for the physicochemical property study are (5% HMDA +
rOV = k 2,MDEA[MDEA][CO2 ] + (k H2O[H 2O] + kMDEA[MDEA] 25% MDEA + 70% H2O), (10% HMDA + 20% MDEA + 70%
+ kHMDA[HMDA])[HMDA][CO2 ] H2O), (15% HMDA + 15% MDEA + 70% H2O), and (20%
HMDA + 10% MDEA + 70% H2O), and for the rate
= k OV[CO2 ] (37) measurement HMDA composition varied in the concentration
14907 DOI: 10.1021/acs.iecr.7b02744
Ind. Eng. Chem. Res. 2017, 56, 14902−14913
Industrial & Engineering Chemistry Research Article

Table 1. Physico-Chemical Properties of Aqueous (HMDA + MDEA)

composition temperature density viscosity diffusivity ×109 dimensionless N2O solubility physical CO2 solubility
%HMDA + %MDEA K g/mL cP m2/s mol/mol mol/mol
5% + 25% 303 1.0150 2.89 0.836 0.490 0.675
313 1.0101 2.10 1.139 0.436 0.614
323 1.0043 1.59 1.508 0.393 0.569
333 0.9976 1.24 1.961 0.361 0.534
10% + 20% 303 1.0067 2.80 0.857 0.498 0.686
313 1.0013 2.02 1.173 0.442 0.625
323 0.9951 1.54 1.548 0.407 0.588
333 0.9884 1.20 1.999 0.374 0.554
15% + 15% 303 0.9983 2.73 0.872 0.507 0.699
313 0.9925 1.98 1.192 0.455 0.643
323 0.9862 1.51 1.571 0.417 0.604
333 0.9795 1.19 2.022 0.390 0.576
20% + 10% 303 0.9895 2.72 0.875 0.514 0.708
313 0.9837 1.98 1.193 0.463 0.654
323 0.9772 1.50 1.579 0.428 0.620
333 0.9700 1.18 2.030 0.405 0.599

Figure 4. Density of aqueous (HMDA + MDEA) solvent: Experimental data and model predictions (AAD, 0.05%).

range of 5−15% keeping the total amine concentration at 30%. Table 2. Binary Parameters of the Excess Molar Volume
Experimental data and modeling results are as follows. Model for [(HMDA (3) + MDEA (2) + H2O (1)] Solvent
4.1. Density. The density of unloaded aqueous (HMDA +
binary pair
MDEA) solvent measured at different temperatures and solvent
compositions is tabulated in Table 1 and shown in Figure 4. As parameter H2O−MDEA H2O−HMDA MDEA−HMDA
seen from the figure, the density of the solvent linearly decreases a0 −565.27 −698.35 −430.82
with the increase in temperature as well as with the increase in the a1 −878.16 −1037.48 −4.29
HMDA concentration in the solvent (total amine concentration a2 −1117.29 −1296.68 4.14
being 30 mass %). The solvent density data is correlated using the
Redlich−Kister excess molar volume model. For the estimation values are compared with the experimental density data in Figure
of excess molar volume of the aqueous ternary mixed solvent, 4. This figure reveals a good correlation between experimental
pure component density data is required. Molar volume data of data and model predicted results with an average absolute
pure water and MDEA are taken from the work of Hsu and Li44 deviation (AAD = (1/n)Σin= 1 |(MExp
i − MMod
i i |) of 0.05%.
)/MExp
and that for HMDA is taken from our previous work.16 Binary Density data obtained in this work is used to estimate the molar
parameters of the Redlich−Kister model are regressed to fit concentration of the solvent and dynamic viscosity data.
experimental molar volume data to that of the model expression 4.2. Viscosity. Dynamic viscosity of the unloaded solvent is
using the least-square method. Regressed model parameters of estimated from the measured kinematic viscosity and density
the aqueous (HMDA + MDEA) solvent system are given in data. Dynamic viscosity data obtained in this work is given in
Table 2. Using these parameters, the model predicted density Table 1 and shown in Figure 5. It is seen from the figure that
14908 DOI: 10.1021/acs.iecr.7b02744
Ind. Eng. Chem. Res. 2017, 56, 14902−14913
Industrial & Engineering Chemistry Research Article

Figure 5. Viscosity of aqueous (HMDA + MDEA) solvent: Experimental data and model predictions (AAD: 0.31%).

dynamic viscosity of the solvent decreases exponentially with 4.3. Physical Solubility. The N2O solubility (mN2O,m) data
temperature. It also decreases with the increase in HMDA (Table 1) measured in this work decreases exponentially with the
concentration (mass%) in the solvent keeping total amine increase in temperature and with the decrease in HMDA
concentration (30 mass%) constant. concentration (Figure 6) in the solvent having constant total
Grunberg-Nissan model is used to correlate the mixed solvent amine concentration (30 mass %). N2O solubility data in
dynamic viscosity data. For this model, the viscosity of pure water aqueous (HMDA + MDEA) is correlated using excess Henry’s
and MDEA are taken from the literature.39 The viscosity of pure coefficient model.40 For this model mN2O,m data is converted in
HMDA is taken from our previous work.16 Binary parameters of
the Grunberg−Nissan model are regressed to fit experimental the form of Henry’s coefficient (HN2O,m = RT/mN2O,m) using eq 1.
viscosity data to that of model predicted values using the least- Henry’s coefficient data for N2O in pure water,33 MDEA,40 and
square method and are presented in Table 3. The model HMDA16 are taken from the literature. The least-squares
optimization method is used to fit the experimental HN2O,m
Table 3. Interaction Parameters of the Viscosity Model for data with that of model predicted HN2O,m values by regressing
[(HMDA (1) + MDEA (2) + H2O (3)] Solvent the two body (αij, αik, αjk) and three body (αijk) interaction
interaction parameters parameters. Regressed model parameters are presented in Table
parameter G12 G13 G23
4. The model predicted HN2O,m values converted to mN2O,m are
b1 62.91 0.939 −6.88 compared with that of experimental mN2O,m values in Figure 6.
b2 −5800.83 1371.64 3122.34 This figure indicates good prediction capability of the model with
b3 −266.95 −232.14 −183.48 0.38% AAD.
Experimentally measured N2O solubility data is used to
estimate physical CO2 solubility (mCO2) in the studied solvent
using N2O analogy. In aqueous amine solvent, physical CO2
predicted dynamic viscosity data for aqueous (HMDA + MDEA) solubility is proportional to the N2O solubility at constant
are compared with the corresponding experimental data in temperature and the proportionality constant is the ratio of the
Figure 5. This work shows good model predictions with 0.31% solubility of CO2 and N2O in pure water. The expression for the
AAD between experimental data and model results. N2O analogy is given below.
The dynamic viscosity data obtained in this work is used to
estimate the diffusivity of the CO2 in an aqueous mixed amine ⎛ mCO ⎞ ⎛ mCO ⎞
solvent using the modified Stokes−Einstein relation15,45 as given ⎜⎜ 2
⎟⎟ = ⎜⎜ 2
⎟⎟
below. ⎝ m N2O ⎠ ⎝ m N2O ⎠ (41)
aminesolution water
(DCO2, Lη0.74)|water = (DCO2, Lη0.74)|Am (39)
Physical CO2 and N2O solubility in pure water are taken from the
Diffusivity of CO2 in water is taken from the work of Versteeg literature.33 Estimated mCO2 data is given in Table 1 which is used
and van Swaaij.33 for the evaluation of the rate constant.
⎛ −2119 ⎞ 4.4. Overall Rate Constant. Rate of CO2 absorption in
DCO2,water = 2.35 × 10−6 exp⎜ ⎟
aqueous (HMDA + MDEA) is measured using the “pressure
⎝ T ⎠ (40)
decay technique”. Overall rate constant (kOV) for different
Estimated diffusivity values are given in Table 1. solvent composition and temperature is estimated using eq 13.
14909 DOI: 10.1021/acs.iecr.7b02744
Ind. Eng. Chem. Res. 2017, 56, 14902−14913
Industrial & Engineering Chemistry Research Article

Figure 6. Dimensionless N2O solubility (mN2O,m) in aqueous (HMDA + MDEA) solvent: Experimental data and model predictions (AAD: 0.38%).

Table 4. Interaction Parameter of Excess Henry Coefficient higher than that in 30 mass % MDEA alone. At a higher HMDA
Model for [(HMDA (1) + MDEA (2) + H2O (3)] Solvent concentration the enhancement of the rate is more pronounced.
Kinetic Model I. In the kinetic model based on the zwitterion
three-body interaction
two-body interaction parameter parameter mechanism (for HMDA−CO2) and base catalyzed hydration
mechanism (for MDEA−CO2), kOV is a function of kinetic
parameter α12 α13 α23 α123
constants k2,HMDA, k2,MDEA, k2,HMDA kH2O/k−1, k2,HMDA kHMDA/k−1,
k1 11.49 −2.59 −4.89 −66.57
and k2,HMDA kMDEA/k−1. Using the values of k2,MDEA taken from
k2 8344.62 371.03 2162.02
the literature4 as shown in eq 42, the values of k2,HMDA, k2,HMDA
k3 −77.38 −218.51 20.95
kH2O/k−1, k2,HMDA kHMDA/k−1, and k2,HMDA kMDEA/k−1 are
For the estimation of the kOV value, required diffusivity and regressed in the form of Arrhenius expression by fitting the
physical CO2 solubility data are evaluated in the previous experimental kOV data to the kinetic model.
sections. Estimated kOV values are given in Table 5 and shown in
⎛ 5400 ⎞
k 2,MDEA /(m 3·kmol−1·s−1) = 4.01 × 108 exp⎜ − ⎟
Table 5. Kinetic Data for the Absorption of CO2 in Aqueous ⎝ T /K ⎠
(HMDA + MDEA) (42)
concentration T kL kOV To evaluate the kinetic constants, the following objective
mass% mass% 105
function is minimized using the generalized reduced gradient
HMDA MDEA K ms −1
s−1 Ha Ei optimization technique.
5 25 303 1.37 1489.9 81 1345
313 1.95 3093.2 96 1520 N exp mod
323 3.01 5865.9 99 1677 1 k OV, i − k OV, i
F= ∑ exp
333 3.90 8738.1 106 1815 N i=1
k OV, i (43)
10 20 303 1.67 5465.7 130 1304
313 2.28 7706.2 132 1472 where, N is the total number of data points, kExp Mod
OV,i and kOV,i are the
323 3.28 12423.1 134 1596 experimental and model predicted values of the overall rate
333 4.13 17204.9 142 1725 constant, respectively.
15 15 303 1.99 10771.5 153 1264 The temperature dependent form of the regressed kinetic
313 2.62 13928.2 155 1409 constants obtained in this work are given below.
323 3.55 19806.6 157 1534
333 4.31 23911.2 162 1633 ⎛ 478 ⎞
k 2,HMDA /(m 3·kmol−1·s−1) = 8.42 × 104 exp⎜ − ⎟
⎝ T /K ⎠
Figures 7 and 8. It can be seen from the plot that kOV increases (44)
with the increase in temperature as well as HMDA concentration
k 2,HMDAk H2O
in the solution. In this figure, the overall rate constant for CO2 /(m 6·kmol−2·s−1)
absorption in 30 mass % (2.5 M) aqueous MDEA4 is also k −1
compared with that of aqueous (HMDA + MDEA) mixed amine ⎛ 11032 ⎞
solvent. At 313 K, the overall rate constant for CO2 absorption in = 7.66 × 1013 exp⎜ − ⎟
aqueous (5%HMDA + 25%MDEA) is found to be 41 times ⎝ T /K ⎠ (45)

14910 DOI: 10.1021/acs.iecr.7b02744

Ind. Eng. Chem. Res. 2017, 56, 14902−14913
Industrial & Engineering Chemistry Research Article

Figure 7. Comparison of experimental kOV data with model prediction based on Kinetic Model I.

Figure 8. Comparison of experimental kOV data with model prediction based on Kinetic Model II.

k 2,HMDAkHMDA to the kinetic model eq 38 developed based on the termolecular

/(m 6·kmol−2·s−1) mechanism for HMDA−CO2 reaction system and base-catalyzed
k −1
hydration mechanism for (MDEA−CO2) reaction system using
⎛ 10589 ⎞
= 1.36 × 1019 exp⎜ − ⎟ the generalized reduced gradient nonlinear optimization
⎝ T /K ⎠ (46) technique using the same objective function as shown in eq 43.
But no meaningful results are obtained when all the bases (H2O,
k 2,HMDAkMDEA MDEA, HMDA) are considered in the termolecular expression.
/(m 6·kmol−2·s−1)
k −1 So, to fit the kinetic model with the experimental data, the
⎛ 12668 ⎞ contribution of MDEA in the termolecular expression of
= 2.28 × 1018 exp⎜ − ⎟ HMDA−CO2 reaction is neglected. This type of approach is
⎝ T /K ⎠ (47) also reported by Ramachandran et al.46 for the aqueous (MDEA
Using these kinetic constants, predicted kOV values are + MEA) system. In view of this, the modified rate expression
compared with the experimental data in Figure 7. The AAD which is fitted to the experimental data, is shown below.
between the model predicted and experimental kOV values is
estimated to be 6.5%. k OV = k 2,MDEA[MDEA] + (k H2O[H 2O]
Kinetic Model II. In this model, kOV is a function of kinetic
constants kHMDA, kMDEA, and kH2O. Experimental kOV is also fitted + kHMDA[HMDA])[HMDA] (48)

14911 DOI: 10.1021/acs.iecr.7b02744

Ind. Eng. Chem. Res. 2017, 56, 14902−14913
Industrial & Engineering Chemistry Research Article

The temperature dependent kinetic constants kHMDA and kH2O

regressed is this work are shown below.
■ AUTHOR INFORMATION
Corresponding Author
⎛ 417 ⎞ *E-mail: amar@che.iitkgp.ernet.in. Tel.: + 91-3222-283948.
kHMDA /(m 3·kmol−1·s−1) = 1.61 × 104 exp⎜ − ⎟ ORCID
⎝ T /K ⎠
Amar N. Samanta: 0000-0002-0290-7461
(49)
Notes
⎛ 6511 ⎞ The authors declare no competing financial interest.

■
k H2O/(m 3·kmol−1·s−1) = 1.38 × 1011 exp⎜ − ⎟
⎝ T /K ⎠ (50)
REFERENCES
Experimental data and predicted results based on this model (1) Mills, S. Coal-Fired CCS Demonstration Plants; IEA Clean Coal
are compared in Figure 8 which shows low average absolute Centre, 2012.
deviation of 8.24%. (2) Kohl, A. L.; Nielsen, R. B. Gas Purification, 5th ed.; Gulf Publishing
The infinite enhancement factor (Ei) and Hatta number (Ha) Company, TX, 1997.
for this reaction system are calculated to check the PFO reaction (3) D’Alessandro, D. M.; Smit, B.; Long, J. R. Carbon Dioxide Capture:
criteria. Estimated Ei and Ha values are listed in Table 5. For all Prospects for New Materials. Angew. Chem., Int. Ed. 2010, 49, 6058.
the runs, Ha values are higher than 3 and Ei values are much (4) Ko, J.; Li, M. Kinetics of Absorption of Carbon Dioxide into
higher than the Ha values. This indicates that the criteria for the Solutions of N-Methyldiethanolamine + water. Chem. Eng. Sci. 2000, 55,
pseudo-first-order reaction is satisfied and eq 13 is valid for the 4139.
(5) Shen, K. P.; Li, M. H. Solubility of Carbon Dioxide in Aqueous
interpretation of CO2 absorption kinetics in aqueous (HMDA + Mixtures of Monoethanolamine with Methyldiethanolamine. J. Chem.
MDEA). Eng. Data 1992, 37, 96.
(6) Kamps, Á . P.-S.; Xia, J.; Maurer, G. Solubility of CO2 in (H2O +
5. CONCLUSION piperazine) and in (H2O + MDEA + piperazine). AIChE J. 2003, 49,
Kinetics of CO2 absorption in the aqueous blend of HMDA and 2662.
MDEA is studied at 303, 313, 323, and 333 K in a reaction (7) Böttger, A.; Ermatchkov, V.; Maurer, G. Solubility of Carbon
calorimeter setup using the pressure decay technique. Density, Dioxide in Aqueous Solutions of N -Methyldiethanolamine and
Piperazine in the High Gas Loading Region. J. Chem. Eng. Data 2009,
viscosity, and physical solubility of the unloaded solvent are also 54, 1905.
measured in this work. Diffusivity and physical solubility of CO2 (8) Mandal, B. P.; Guha, M.; Biswas, A. K.; Bandyopadhyay, S. S.
in the solvent are evaluated using the modified Stokes−Einstein Removal of Carbon Dioxide by Absorption in Mixed Amines: Modelling
relation and N2O-analogy, respectively. The overall rate constant of Absorption in Aqueous MDEA/MEA and AMP/MEA Solutions.
for the CO2 absorption in aqueous (HMDA + MDEA) is Chem. Eng. Sci. 2001, 56, 6217.
measured in the PFO reaction regime. Kinetic data shows (9) Liao, C. H.; Li, M. H. Kinetics of Absorption of Carbon Dioxide
significant enhancement of the CO2 absorption rate in aqueous into Aqueous Solutions of monoethanolamine + N-Methyldiethanol-
(HMDA + MDEA) solvent due to the addition of a small amount amine. Chem. Eng. Sci. 2002, 57, 4569.
of HMDA in aqueous MDEA. At 313 K, overall CO2 absorption (10) Rinker, E. B.; Ashour, S. S.; Sandall, O. C. Absorption of Carbon
rate constant in aqueous (5%HMDA + 25%MDEA) is 41 times Dioxide into Aqueous Blends of Diethanolamine and Methyldiethanol-
amine. Ind. Eng. Chem. Res. 2000, 39, 4346.
higher than that of aqueous 30% MDEA. Again, absorption rate (11) Derks, P. W. J.; Kleingeld, T.; van Aken, C.; Hogendoorn, J. A.;
of this solvent (19.7 × 10−6 kmol·m−2·s−1 at 313 K) is 14% higher Versteeg, G. F. Kinetics of Absorption of Carbon Dioxide in Aqueous
than that of aqueous (5%PZ + 25%MDEA) (16.9 × 10−6 kmol· Piperazine Solutions. Chem. Eng. Sci. 2006, 61, 6837.
m−2·s−1 at 313 K).21 It is also indicated in our previous work that (12) Chen, X.; Rochelle, G. T. Aqueous Piperazine Derivatives for CO2
at 15 kPa CO2 partial pressure, loading capacity of this solvent Capture: Accurate Screening by a Wetted Wall Column. Chem. Eng. Res.
(αCO2 = 0.584 mol·mol−1 at 313 K) is 24.5% and 16.7% higher Des. 2011, 89, 1693.
(13) Ume, C. S.; Ozturk, M. C.; Alper, E. Kinetics of CO2 Absorption
than that of aqueous 30% MDEA (αCO2 = 0.0.441 mol·mol−1 at by a Blended Aqueous Amine Solution. Chem. Eng. Technol. 2012, 35,
313 K)47 and (5%PZ + 25%MDEA) (αCO2 = 0.486 mol·mol −1 at 464.
313 K)48 respectively. Low CO2 absorption enthalpy of this (14) Ma’mun, S.; Dindore, V. Y.; Svendsen, H. F. Kinetics of the
solvent is also reported in our previous work.32 At 313 K and 0.5 Reaction of Carbon Dioxide with Aqueous Solutions of 2-((2-
Aminoethyl)amino)ethanol. Ind. Eng. Chem. Res. 2007, 46, 385.
mol·mol −1 CO2 loading, absorption enthalpy of aqueous (5% (15) Singh, P.; van Swaaij, W. P. M.; Brilman, D. W. F. Kinetics Study
HMDA + 25%MDEA) is 67 kJ·mol −1 CO2. All these properties of Carbon Dioxide Absorption in Aqueous Solutions of 1,6-
indicate the superior solvent characteristics of aqueous (HMDA Hexamethyldiamine (HMDA) and 1,6-Hexamethyldiamine, N,N′ Di-
+ MDEA) for CO2 capture. Methyl (HMDA, N,N′). Chem. Eng. Sci. 2011, 66, 4521.
To evaluate the kinetic characteristics of this mixed amine (16) Mondal, B. K.; Bandyopadhyay, S. S.; Samanta, A. N. Kinetics of
solvent, the overall rate of CO2 absorption is assumed to be the CO2 Absorption in Aqueous Hexamethylenediamine. Int. J. Greenhouse
combined effect of rate contributions of CO2−HMDA and Gas Control 2017, 56, 116.
CO2−MDEA reaction system. The rate contribution of CO2− (17) Dinda, S.; Goud, V. V.; Patwardhan, A. V.; Pradhan, N. C. Kinetics
HMDA reaction system is analyzed using both zwitterion of Reactive Absorption of Carbon Dioxide with Solutions of 1,6-
(Kinetic Model I) and termolecular (Kinetic Model II) Hexamethylenediamine in Polar Protic Solvents. Sep. Purif. Technol.
2010, 75, 1.
mechanisms, whereas the CO2−MDEA reaction system is (18) Xu, Z.; Wang, S.; Chen, C. Kinetics Study on CO2 Absorption
analyzed by the base catalyzed hydration mechanism. Both with Aqueous Solutions of 1,4-Butanediamine, 2-(Diethylamino)-
kinetic models developed in this work are able to predict the Ethanol, and Their Mixtures. Ind. Eng. Chem. Res. 2013, 52, 9790.
kinetic data with good accuracy. Kinetic Model I is found to (19) Li, J.; Henni, A.; Tontiwachwuthikul, P. Reaction Kinetics of CO2
represent the experimental data better with an AAD of 6.5% in Aqueous Ethylenediamine, Ethyl Ethanolamine, and Diethyl
compared to the Kinetic Model II (AAD: 8.3%). Monoethanolamine Solutions in the Temperature Range of 298−313

14912 DOI: 10.1021/acs.iecr.7b02744

Ind. Eng. Chem. Res. 2017, 56, 14902−14913
Industrial & Engineering Chemistry Research Article

K, Using the Stopped-Flow Technique. Ind. Eng. Chem. Res. 2007, 46, (2-Amino-2-Methyl-1-Propanol + Monoethanolamine), and (2-Amino-
4426. 2-Methyl-1-Propanol + Diethanolamine). J. Chem. Eng. Data 2003, 48,
(20) Monteiro, J. G. M.-S.; Hussain, S.; Majeed, H.; Mba, E. O.; 703.
Hartono, A.; Knuutila, H.; Svendsen, H. F. Kinetics of CO2 Absorption (40) Wang, Y. W.; Xu, S.; Otto, F. D.; Mather, A. E. Solubility of N2O in
by Aqueous 3-(Methylamino)propylamine Solutions: Experimental Alkanolamines and in Mixed Solvents. Chem. Eng. J. 1992, 48, 31.
Results and Modeling. AIChE J. 2014, 60, 3792. (41) Versteeg, G. F.; van Swaaij, W. P. M. On the Kinetics between
(21) Samanta, A.; Bandyopadhyay, S. S. Absorption of Carbon Dioxide CO2 and Alkanolamines Both in Aqueous and Non-Aqueous solutions-
into Piperazine Activated Aqueous N-Methyldiethanolamine. Chem. I. Primary and Secondary Amines. Chem. Eng. Sci. 1988, 43, 573.
Eng. J. 2011, 171, 734. (42) Tomcej, R. a; Otto, F. D. Absorption of CO2 and N2O into
(22) Dubois, L.; Thomas, D. CO2 Absorption into Aqueous Solutions Aqueous Solutions of Methyldiethanolamine. AIChE J. 1989, 35, 861.
of Monoethanolamine, Methyldiethanolamine, Piperazine and Their (43) Pinsent, B. R. W.; Pearson, L.; Roughton, F. J. W. The Kinetics of
Blends. Chem. Eng. Technol. 2009, 32, 710. Combination of Carbon Dioxide with Hydroxide Ions. Trans. Faraday
(23) Tan, C. S.; Chen, J. E. Absorption of Carbon Dioxide with Soc. 1956, 52, 1512.
Piperazine and Its Mixtures in a Rotating Packed Bed. Sep. Purif. Technol. (44) Hsu, C.; Li, M.-H. Densities of Aqueous Blended Amines. J. Chem.
2006, 49, 174. Eng. Data 1997, 42, 502.
(24) Bishnoi, S.; Rochelle, G. T. Absorption of Carbon Dioxide in (45) van Holst, J.; Versteeg, G. F.; Brilman, D. W. F.; Hogendoorn, J. A.
Aqueous Piperazine/ methyldiethanolamine. AIChE J. 2002, 48, 2788. Kinetic Study of CO2 with Various Amino Acid Salts in Aqueous
(25) Zhang, X.; Zhang, C. F.; Qin, S. J.; Zheng, Z.-S. A Kinetics Study Solution. Chem. Eng. Sci. 2009, 64, 59.
on the Absorption of Carbon Dioxide into a Mixed Aqueous Solution of (46) Ramachandran, N.; Aboudheir, A.; Idem, R.; Tontiwachwuthikul,
Methyldiethanolamine and Piperazine. Ind. Eng. Chem. Res. 2001, 40, P. Kinetics of the Absorption of CO2 into Mixed Aqueous Loaded
3785. Solutions of Monoethanolamine and Methyldiethanolamine. Ind. Eng.
(26) Seo, D. J.; Hong, W. H. N-Methyldiethanolamine Piperazine Chem. Res. 2006, 45, 2608.
Effect of Piperazine on the Reaction Rate Constant of Carbon Dioxide (47) Kundu, M.; Bandyopadhyay, S. S. Modelling Vapour − Liquid
into Aqueous N-Methyldiethanolamine Solutions. J. Korean Inst. Chem. Equilibrium of CO2 in Aqueous N-Methyldiethanolamine through the
Eng. 1999, 37, 593. Simulated Annealing Algorithm. Can. J. Chem. Eng. 2005, 83, 344.
(27) Xu, G.; Zhang, C.; Qin, S. J.; Wang, Y. Kinetics Study on (48) Dash, S. K.; Bandyopadhyay, S. S. Studies on the Effect of
Absorption of Carbon Dioxide into Solutions of Activated Methyl- Addition of Piperazine and Sulfolane into Aqueous Solution of N-
diethanolamine. Ind. Eng. Chem. Res. 1992, 31, 921. Methyldiethanolamine for CO2 Capture and VLE Modelling Using
(28) Choi, S. Y.; Nam, S. C.; Yoon, Y.; Park, K. T.; Park, S. J. Carbon eNRTL Equation. Int. J. Greenhouse Gas Control 2016, 44, 227.
Dioxide Absorption into Aqueous Blends of Methyldiethanolamine
(MDEA) and Alkyl Amines Containing Multiple Amino Groups. Ind.
Eng. Chem. Res. 2014, 53, 14451.
(29) Bonenfant, D.; Mimeault, M.; Hausler, R. Comparative Analysis
of the Carbon Dioxide Absorption and Recuperation Capacities in
Aqueous 2-(2-Aminoethylamino)ethanol (AEE) and Blends of
Aqueous AEE and N -Methyldiethanolamine Solutions. Ind. Eng.
Chem. Res. 2005, 44, 3720.
(30) Mondal, B. K.; Bandyopadhyay, S. S.; Samanta, A. N. Vapor-
Liquid Equilibrium Measurement and ENRTL Modeling of CO2
Absorption in Aqueous Hexamethylenediamine. Fluid Phase Equilib.
2015, 402, 102.
(31) Mondal, B. K.; Bandyopadhyay, S. S.; Samanta, A. N. Equilibrium
Solubility Measurement and Kent-Eisenberg Modeling of CO 2
Absorption in Aqueous Mixture of N-Methyldiethanolamine and
Hexamethylenediamine. Greenhouse Gases: Sci. Technol. 2017, 7, 202.
(32) Mondal, B. K.; Bandyopadhyay, S. S.; Samanta, A. N.
Measurement of CO2 Absorption Enthalpy and Heat Capacity of
Aqueous Hexamethylenediamine and Its Aqueous Mixture with N-
Methyldiethanolamine. J. Chem. Thermodyn. 2017, 113, 276.
(33) Versteeg, G. F.; Van Swaaij, W. Solubility and Diffusivity of Acid
Gases (Carbon Dioxide, Nitrous Oxide) in Aqueous Alkanolamine
Solutions. J. Chem. Eng. Data 1988, 33, 29.
(34) Laddha, S. S.; Diaz, J. M.; Danckwerts, P. V. The N2O Analogy:
The Solubilities of CO2 and N2O in Aqueous Solutions of Organic
Compounds. Chem. Eng. Sci. 1981, 36, 228.
(35) Ying, J.; Eimer, D. A. Determination and Measurements of Mass
Transfer Kinetics of CO2 in Concentrated Aqueous Monoethanolamine
Solutions by a Stirred Cell. Ind. Eng. Chem. Res. 2013, 52, 2548.
(36) Danckwerts, P. V. Gas-Liquid Reactions; McGraw-Hill Book Co:
New York, 1970.
(37) Blauwhoff, P. M. M.; Versteeg, G. F.; Van Swaaij, W. P. M. A Study
on the Reaction between CO2 and Alkanolamines in Aqueous Solutions.
Chem. Eng. Sci. 1983, 38, 1411.
(38) Prausnitz, J. M.; Lichtenthaler, R. N.; Azevedo, E. G. de. Molecular
Thermodynamics of Fluid-Phase Equilibria, 3rd.; Prentice Hall Inc: NJ,
1999.
(39) Mandal, B. P.; Kundu, M.; Bandyopadhyay, S. S. Density and
Viscosity of Aqueous Solutions of (N -Methyldiethanolamine +
Monoethanolamine), (N -Methyldiethanolamine + Diethanolamine),

14913 DOI: 10.1021/acs.iecr.7b02744

Ind. Eng. Chem. Res. 2017, 56, 14902−14913