CO2 Absorption Simulation Paper
CO2 Absorption Simulation Paper
CO2 Absorption Simulation Paper
pubs.acs.org/IECR
ABSTRACT: A pilot plant at laboratory scale for absorption of CO2 has been constructed and operated to test the CO2 removal
by aqueous ammonia solutions. The design of the pilot plant is based on a standard absorption and desorption flow sheet and
partial or complete separation of gas mixtures. Pilot plant data for CO2 removal efficiencies, effects of CO2 loading, and
temperature profiles are obtained. A rate-based model, RateFrac in Aspen Plus simulator, is used to simulate the CO2 absorption
of the pilot plant. The simulation results of the CO2 capture predicted by the rate-based model are in good agreement with the
experimental data of the pilot plant. Further, the optimization covering operational parameters is carried out using the rate-based
model.
© 2012 American Chemical Society 5309 dx.doi.org/10.1021/ie2030536 | Ind. Eng. Chem. Res. 2012, 51, 5309−5319
Industrial & Engineering Chemistry Research Article
consumption, and practical experience, is still scarce in the mixtures. This is an integrated pilot plant operating with
public domain. Therefore, it is necessary to study intensively absorber, stripper, separation system, and connected unit
the CO2 absorption and desorption processes. operations. All of columns and reactors are made of 304
Process simulation also provides the means to interpret stainless steel. The packing is metal wire gauze structured
laboratory experiments and quantify the process performance packing (Tianjin Septech Co., Ltd. DZ-II-750Y). Its correlation
among various CO2-capture processes using aqueous ammonia. parameters are shown in Table 1. The internal diameters of
Darde et al.29 have presented thermodynamic analysis of the absorber 1 and absorber 2 are 0.15 and 0.16 m, respectively.
chilled-ammonia process. It calculates the activity coefficient for The packing height of the both absorbers is 1.47 m. The
the liquid phase using the extended UNIQUAC model and the packing height of absorber 2 can be varied within 1.47 m, which
gas phase fugacity using the Soave−Redlich−Kwong (SRK) allows investigation of the effect of packing height on CO2
equation for the volatile compounds. Mathias et al.30 have removal. There are two methods to generate flue gas: one is by
quantitatively evaluated the chilled-ammonia process for CO2 methane combustion in a furnace, and the other is by mixture
capture using thermodynamic analysis and process simulation. of air and CO2. The flow rates of methane, air, and CO2 are
The Electrolyte-NRTL model for the NH3−CO2−H2O system 0.8−1.0, 10.0−16.0, and 0−2.0 m3/h, respectively. The flue gas
has been used in the Aspen Plus simulator. It should be noted enters the lower part of absorber 1. The flow sheet incorporates
that the process simulation model used by Darde et al.29 and two absorbers, which can be run in series or in parallel. In the
Mathias et al.30 assumes physical and chemical equilibrium and case of series operation, the solvent enters absorber 2 first and
does not consider reaction kinetics and vapor-to-liquid mass then absorber 1. In the absorbers, the aqueous ammonia and
transfer rate in the absorber. It is known that the rate-based flue gas stream were in a counter flow pattern.
modeling approach is rigorous and offers higher model fidelity In the ammonia pilot plant CSIRO and Delta Electricity
over the traditional equilibrium-stage modeling approach.32 constructed,28 ammonia loss at the stripper gas outlet is
Therefore, developing the rate-based process modeling of CO2 reduced due to high pressure of the stripper. The ammonia
absorption with aqueous ammonia is necessary. concentration can be controlled at 200 ppm under a pressure of
Industrial or pilot plant data available for aqueous ammonia 850 kPa at temperatures of 20−25 °C. However, these pressure
separation processes are still limited in the public literature. and temperature conditions in the stripper condenser and
Besides, the research of the rate-based model for CO2 reflux lines lead to the precipitation of ammonium bicarbonate.
absorption using aqueous ammonia solution is still at the The buildup of solids can block the stripper condenser and
initial stage. The motivation of this paper is to investigate the reflux pipes, leading to a shutdown of the pilot plant. Blockage
feasibility of CO2 removal by an ammonia process in a pilot also occurs in instrument tubing, leading to false readings and
plant at laboratory scale. A rate-based absorption model, affecting the operation. In order to solve this problem, the
RateFrac, based on Aspen Plus process simulator, is used to
stripper and separation system of gas mixture, containing CO2
simulate the pilot plant, and the experimental results will
and ammonia, are adopted in this study. The pressure of
provide the validity of rate-based modeling of the CO2 capture
stripper and separation system is near the atmospheric pressure.
process by ammonia.
The lower pressure decreases the temperature of reboiler that is
about 85−95 °C. Low-grade thermal energy may be applied for
2. EXPERIMENTAL INSTALLATION the heat of reboiler, which will be a substitute for the steam
2.1. Description of the Laboratory Pilot Plant. A from steam turbine. The temperature of stripper condenser and
simplified flow sheet for the pilot rig is shown in Figure 1. The separation system is about 50−70 °C. The precipitation of
design of the pilot plant is based on a standard absorption and ammonium bicarbonate or ammonium carbonate will decom-
desorption flow sheet and partial or complete separation of gas pose in this temperature, so blockage of condenser and pipes
5310 dx.doi.org/10.1021/ie2030536 | Ind. Eng. Chem. Res. 2012, 51, 5309−5319
Industrial & Engineering Chemistry Research Article
will not be a problem. Ammonia loss at the stripper gas outlet is NH3 concentration in the liquid samples (apparent composi-
obvious at this pressure and temperature condition. The tion). Supposing components in the solution are H2O, NH3,
separation system is used to separate the CO2 and NH3 gas. CO2, H3O+, OH−, NH4+, NH2COO−, HCO3−, CO32−, N2,
The system is based on the principle that NH3 is absorbed by NH4HCO3, and CO2, loading of solutions in this paper refers
water easier than CO2. The ammonia gas is dissolved by CO2- to the molar ratio of CO2 to NH3 in the solvent as follows:
lean stream from reboiler at high gas velocity.33 After cooling
and ammonia makeup, the stream from separation system is CO2 loading = [CO2 ]tot /[NH3]tot
reused for absorption of CO2. {[CO2 ] + [NH2COO−] + [HCO3−] + [CO32 −]}
In order to simplify the operation, the absorption and =
stripping experiments were conducted separately, and only the {[NH3] + [NH 4+] + [NH2COO−]}
absorber data are reported here. Data for the pilot absorber unit (1)
and operating conditions are shown in Table 2. The pressure of
where [ ] is the concentration of components, mol/L. The total
CO2 and NH3 concentrations are attained by titration analysis.
Table 2. Operation Conditions for the Absorber The ion and molecule concentrations are calculated by rate-
parameter based modeling.
liquid flow rate (L/h) 60−180 The CO2 removal efficiency in absorber refers to percentage
ammonia concentration (wt %) 7.6−8.5 of the amount of CO2 absorbed in the absorber per unit time
rich solution loading (mol/mol) 0.084−0.482 (kg/h) with respect to the CO2 mass flow rate at the inlet of
lean solution loading (mol/mol) 0−0.397 absorber. The calculation method is given as follows:
temperature lean stream to absorber 2 (°C) 18.0−29.0 y − y2
flue gas flow rate (m3/h) 9.0−16.0 ε= 1 × 100%
y1(1 − y2 ) (2)
CO2 concentration of flue gas (v/v, %) 6.9−15.7
temperature of flue gas to absorber 1 (°C) 17.7−138.0 where ε is the CO2 removal efficiency; y1 is the CO2 mole
fraction at the inlet of absorber 1; y2 is the CO2 mole fraction at
absorber is atmospheric pressure. The packing height of the outlet or some position of absorber.
absorber 2 can be varied which allows investigation of the Individual liquid and gas phase mass balances for CO2 could
effect of packing height on CO2 removal. The tests were be obtained on the basis of the content of CO2 in both phases.
performed with 7.6−8.5 wt % ammonia solutions. The accuracy of the balances is defined as percent deviation:
2.2. Measurements. The gas and liquid sampling points
were given in Figure 1. Five gas sampling points are arranged Liq CO − GasCO2
2
Devexp(moleflow) = × 100%
along absorber 1 height. The concentration of CO2 is measured max(Liq CO , GasCO2)
by an infrared gas analyzer (Beijing Huyuan Analytical 2
Instrument Institute Ltd.). At first, the sampling gas is induced (3)
into a small gas washing bottle loading sulfuric acid, where most where LiqCO2 is the increased amount of CO2 in absorber
of NH3 is absorbed. Then, the sampling gas enters into the gas liquid, mol/h; GasCO2 is the decreased amount of CO2 in
dryer filled with desiccant CaCl2, where water vapor and the left absorber gas, mol/h.
NH3 are absorbed. The concentration of NH3 is measured by a 2.3. Evaluation of Experimental Data with CO 2
laser analyzer (Beijing Define Technology Ltd.). Sampling pipe Absorption. The pilot plant experiments of absorption were
and measuring cell in the analyzer are heated to the conducted for all 19 runs. Their operational parameters are
temperature of 120−130 °C in order to minimize the reaction shown in Supporting Information, Table S1. The CO2 loading
between CO2 and NH3. The full scales of measurement ranges of CO2-rich solutions at absorber outlet was measured only for
of CO2 and NH3 analyzer are 20% (v/v) and 10% (v/v), cases 11−19 (as shown in Supporting Information, Table S2).
respectively. The CO2 and NH3 analyzers are calibrated to a Deviation can be calculated by eq 3. The highest absolute value
precision of about ±2% F.S. (full scale) after each run, with at of deviation is 9.35%. This systematic deviation is so small that
least 4−5 different concentrations covering the whole range of it is acceptable and actually can be taken as a support for the
interest. All temperature probes (PT-100 resistance temper- quality of the experimental data. This indicates that the liquid
ature sensors) were calibrated to a precision of about ±0.5 °C. and gas phase measurement methods are appropriate.
After a change in operational conditions, the plant is run for
at least 30 min to ensure steady state operation. Liquid samples 3. MODEL DEVELOPMENT
are withdrawn from the system for analysis, and operational The absorber was modeled with RateFrac, the rate-based
data are recorded at the same time. The liquid sampling points distillation module of Aspen Plus. The details of RateFrac
are arranged at the absorber inlet and outlet. Each time, the model can be found in Aspen Plus Documentation.34 This
liquid sample titration analysis is conducted within 10 min. approach required that a rigorous thermodynamic model, a
Standard wet chemistry titration is performed with an rigorous rate model, and transport property models be
automatic titrator (Metrohm 877 Titrino plus) in the analytic implemented in Aspen Plus.
lab for the determination of CO2 and NH3 content in the liquid 3.1. Thermodynamic Model. The RateFrac model used
samples. The titration method is summarized as follows: the rigorous Electrolyte-NRTL thermodynamics to predict
Hydrochloric acid titrating solution (0.1 N HCl) is added to an enthalpies, entropy, equilibrium vapor pressures of NH3, CO2,
aliquot of a diluted sample with a dropwise add velocity. When and H2O, and solution speciation. Thermodynamic models for
the first and second inflection points of titration curve are electrolyte systems require a chemistry model. The chemistry
attained, the usages of HCl are EP1 mL and EP2 mL, model of NH3−CO2−H2O system is presented in Table 3.
respectively. EP2-EP1 and EP2 are used to calculate CO2 and Chemical equilibrium is assumed in the chemistry model. CO2
5311 dx.doi.org/10.1021/ie2030536 | Ind. Eng. Chem. Res. 2012, 51, 5309−5319
Industrial & Engineering Chemistry Research Article
where r is the rate of reaction; k is the pre-exponential factor; n 4. COMPARISON BETWEEN EXPERIMENTS AND
is the temperature exponent; E is the activation energy. The RATE-BASED MODEL PREDICTIONS
factors k, n, and E are given in Table 6. The kinetic parameters As mentioned above, pilot plant at laboratory scale had been
for reactions 4−7 in Table 5 are derived from the work of constructed and operated to test the ammonia process. Rate-
Pinsent et al.41 based rigorous simulation of the process is beneficial to
comprehend and estimate the CO2 absorption processes with
Table 6. Parameters k, n, E in Eq 5 NH3. The experimental data provides the verification of rate-
reaction ID k n E, cal/mol
based NH3 model. The model is meant to be used as a guide
for modeling more sophisticated CO2 capture processes with
4 4.32 × 1013 0 13249
NH3 for process development, debottlenecking, and plant and
5 2.38 × 1017 0 29451
equipment design.
6 1.35 × 1011 0 11585
4.1. CO2 Removal Efficiency. The absorption experiments,
7 4.75 × 1020 0 16529
operational parameters, and the CO2 removal efficiency ε
simulation results for all 19 cases are shown in Supporting
RateFrac model in Aspen Plus provides four different flow Information, Table S1. The CO2 removal efficiencies of
models, which are Mixed, CouterCurrent, Vplug, and VPlug- experimental and simulated data of cases 1−16 are compared
Pavg. CounterCurrent flow model is chosen to determine the in Figure 4. It can be seen that the correspondence between
bulk properties required to evaluate the mass and energy fluxes
and reaction rates. The bulk properties for each phase are an
average of the inlet and outlet properties.
The RateFrac module of Aspen Plus is used to integrate the
heat transfer and multicomponent mass-transfer relationships
for a packed absorber. Mass transfer rate depends on the binary
mass transfer coefficients, the averaged mole fraction driving
force, and interfacial area. The flow model determines the
averaged mole fraction driving force. The mass transfer
coefficients and interfacial area are calculated by empirical
correlation formula.42 The packing is set as CY structured
packing. Its correlation parameters are shown in Table 1.
Interfacial area factor is used to match the experimental value.
Figure 4. Comparison of the experimental and the simulated CO2
Besides, liquid holdup is an important parameter. In the removal efficiency.
simulation model, gas phase reactions are ignored. Liquid
holdup is used to calculate the reactions of bulk liquid. Liquid experimental and simulated values is good at the different
holdup is suitable to be set as 3−5% of free volume,43 and 4% operational conditions of cases 1−16.
of free volume is chosen in this paper. The heat transfer The profile of the experimental and the simulated ε along
coefficient may be estimated by Chilton-Colburn method.44 packing height is shown in Figure 5. The packing height is
Heat transfer area is assumed as calculated interfacial area. Gas-
film and liquid-film mass- and heat-transfer coefficients were
obtained as a function of liquid and gas flow rates and
properties with the default correlations of RateFrac.
3.3. Transport Property Models. Rate-based process
models require transport property models applied for heat
transfer, mass transfer, interfacial area, liquid holdup, etc.
Transport property models include viscosity, thermal con-
ductivity, diffusion coefficient, and surface tension.
For the liquid viscosity, the Jones-Dole electrolyte correction
model is used to calculate the corrected value of viscosity of
liquid mixture. The interaction parameters between NH3 and
H2O are regressed against experimental NH3−H2O viscosity
data from Pleskov and Igamberdyev45 and Frank et al.46 The Figure 5. Comparison between the experimental and the simulated ε
Jones-Dole model parameters of CO32− and HCO3− are profile for case 11.
regressed against K2CO3−H2O viscosity data from Pac et al.47
counted from the bottom of packing in the absorber. In cases
and KHCO3−H2O viscosity data from Palaty,48 respectively.
For NH4+, Aspen Plus databank values are used. 1−11, only absorber 1 is used, the packing height is 1.47 m, and
For the liquid thermal conductivity, the Riedel electrolyte the ammonia concentration of inlet liquid is 4.677 mol/L. Five
correction model is used.49 Thermal conductivity of compo- gas sampling points were arranged along absorber 1 height. The
nent NH2COO− does not have the experimental data, and sampling gas may contain droplets, which would cause reaction
value of zero is used. Electrolyte solutions diffusion coefficient of CO2 and ammonia in sampling pipes. Therefore, in the
is calculated using the Nernst−Hartley model. Diffusion middle section of absorber 1, the measured ε is higher than the
coefficient of component NH2COO− does not have the simulated ε. At the outlet of absorber, droplets and reaction in
experimental data, and nominal value is used. Liquid surface sampling pipes may be ignored. In general, the tendency of the
tension is calculated by the Onsager−Samaras model. experimental and the simulated ε is in good agreement.
5313 dx.doi.org/10.1021/ie2030536 | Ind. Eng. Chem. Res. 2012, 51, 5309−5319
Industrial & Engineering Chemistry Research Article
Figure 10. Effects of the packing height on ε, CO2 loading, NH3 concentration of outlet gas, and temperature.
Figure 11. Effects of the NH3 concentration of CO2-lean solutions on ε, CO2 loading, NH3 concentration of outlet gas, and temperature.
aqueous ammonia solution can be applied for process of CO2 reacted with ammonia is increasing, and reaction heat
simulation and optimization analysis. release is increasing. The release heat makes the temperature of
CO2-rich solutions increase. It is concluded that there is a best
5. PROCESS OPTIMIZATION WITH RATE-BASED packing height, which is 3−4 m in this operation condition.
MODEL SIMULATIONS 5.2. Optimization of Concentration. Effects of the NH3
Process optimization can be divided into two main procedures: concentration of CO2-lean solution on ε, CO2 loading, NH3
absorption and desorption. In this study, only the absorption concentration of outlet gas, and temperature of rich solution are
process optimization is investigated in order to attain the shown in Figure 11. It can be seen that the concentration of
control mechanism and optimization approach. Moreover, aqueous ammonia solution plays an important role on the CO2
process optimization with rate-based model simulations will removal efficiency. With the NH3 concentration of CO2-lean
provide theoretical guidance for energy-saving technology. For solution increasing, the ε increases to a high level and keeps a
the CO2 absorption, main optimization target parameters are ε, stable value, the CO2 loading of rich solution decreases, the
CO2 loading of CO2-rich solution, and NH3 concentration of NH3 concentration of outlet gas increases quickly, and the
outlet gas. The optimization parameters and their setting temperature of rich solution reaches to a peak value. When the
operational parameters for process optimization with the rate- NH3 concentration of CO2-lean solution is higher than 4 mol/
based model are shown in Supporting Information, Table S3. L, improving the concentration of aqueous ammonia has a little
5.1. Optimization of Packing Height. Packing height has effect on the CO2 removal efficiency under the condition of this
effects on construction cost, operation, and process conditions. study, but the NH3 concentration of outlet gas increases
Figure 10 gives the effects of the packing height on ε, CO2 sharply. As mentioned above, on the basis of the premise that
loading, NH3 concentration of outlet gas, and temperature of the required ε could be realized, it is necessary to reduce the
rich solution. Figure 10a shows that the packing height has an NH3 concentration of outlet gas and improve the CO2 loading
obvious effect on the ε. It can be seen that the ε and the CO2 of CO2-rich solution as much as possible. NH3 leakage will give
loading of CO2-rich solutions increase to stable values with the rise to NH3 loss and environmental pollution. When the NH3
packing height increasing. Mass transfer rate is related to concentration of CO2-lean solution is 3.5−4 mol/L, the
interfacial area. Improving packing height leads to the temperature of rich solution reaches to its highest value,
increment of interfacial area obviously, which enhances the which is beneficial for the regeneration of CO2-rich solution.
mass transfer. When the packing height is higher than 4.0 m, Thus, the best NH3 concentration of CO2-lean solution is 3.5−
the increased mass transfer area has a little effect on the CO2 4 mol/L, which is the obvious difference comparing aqua
absorption. Figure 10b shows that the NH3 concentration of ammonia in the chilled ammonia process; its NH3 concen-
outlet gas has a little augment with the packing height tration may reach up to 28%(w/w). When the ε increases to a
increasing, and the temperature of rich solution increases from high level and keeps a stable value, reaction heat release also
25 to 32.5 °C. With the packing height increasing, the amount keeps a constant value, but the heat consumption due to the
5315 dx.doi.org/10.1021/ie2030536 | Ind. Eng. Chem. Res. 2012, 51, 5309−5319
Industrial & Engineering Chemistry Research Article
Figure 12. Effects of the CO2 concentration of flue gas on ε, CO2 loading, NH3 concentration of outlet gas, and temperature.
Figure 13. Effects of the temperature of liquid on ε, CO2 loading, NH3 concentration of outlet gas, and temperature.
Figure 14. Effects of the temperature of flue gas on ε, CO2 loading, NH3 concentration of outlet gas, and temperature.
evaporation of NH3 increases when the NH3 concentration of concentration of flue gas increasing, the amount of CO2
CO 2 -lean solution increases. It is found that the NH3 reacted with ammonia is increasing, but the flow rate and the
concentration of flue gas is high at the high NH3 concentration concentration of aqua ammonia is not changed. Therefore, the
of CO2-lean solution. Besides, the specific heat of ammonia ε is declined, and reaction heat release is increasing. The release
solution also increases a little with the NH3 concentration of heat makes the temperature of CO2-rich solutions increase. In
CO2-lean solution increase. Therefore, the temperature of CO2- order to attain large CO2 loading of rich solution, on the basis
rich solution decreases when the NH3 concentration of CO2- of the premise that the required ε is higher than 90%, the best
lean solution is greater than 4 mol/L. CO2 concentration of flue gas is chosen as 12−13%.
Figure 12 shows effects of the CO2 concentration of flue gas 5.3. Optimization of Temperature. In the process of
on ε, CO2 loading, temperature of rich solution, and NH3 CO2 absorption, temperature is an important parameter
concentration of outlet gas. With the inlet concentration of influencing reaction kinetics, heat and mass transfer rate, and
CO2 increasing, the CO2 removal efficiency decreased a little, diffusion rate. Effects of the temperature of liquid on ε, CO2
the CO2 loading of CO2-rich solutions increases, the temper- loading, NH3 concentration of outlet gas, and temperature of
ature of rich solution increases, and the NH3 concentration of rich solution are shown in Figure 13. It can be seen that the ε
outlet gas keeps a nearly constant value. With the CO2 reaches to the highest value at the temperature of 33 °C. In the
5316 dx.doi.org/10.1021/ie2030536 | Ind. Eng. Chem. Res. 2012, 51, 5309−5319
Industrial & Engineering Chemistry Research Article
Figure 15. Effects of the flow rate of flue gas on ε, CO2 loading, NH3 concentration of outlet gas, and temperature.
literature of Diao et al.,17 their experimental results also show beneficial for the ammonia evaporation. In this operation
that the CO2 removal efficiency reaches its highest value at 33 condition, the best flow rate of flue gas is about 15−16 m3/h.
°C, which further validate the simulation results of rate-based
NH3 model. As shown in Figure 13a,b, when the temperature 6. CONCLUSIONS
of liquid increases from 13 to 56 °C, the CO2 loading of rich
solution and the NH3 concentration of outlet gas grow rapidly Experimental studies and rate-based process simulations of CO2
in the same way, and the temperature of rich solution increases absorption with aqueous ammonia solutions have been carried
to a high level and keeps a constant value. The growth of the out. A pilot plant at laboratory scale for absorption of CO2 has
CO2 loading of rich solution arises from increment of the NH3 been constructed and operated to test the feasibility of CO2
concentration of outlet gas. Considering the ε and the NH3 removal by aqueous ammonia solutions. CO2 removal
concentration at the outlet, the best temperature of liquid is efficiencies and temperature profiles along the packing height
lower than 30 °C. When the temperature of CO2-lean solution are obtained. The effects of the CO2 loading of CO2-lean
is 33 °C, the highest temperature in the packed tower is 47.1 solutions on CO2 removal efficiency, NH3 concentration of
°C. outlet gas, and CO2 loading of CO2-rich solutions are
In the packed absorber, the temperatures of gas and liquid investigated. Experimental results show that the best CO2
increase with the temperature of CO2-lean solution increasing. loading of CO2-lean solutions is 0.124−0.150 in order to
The high temperature accelerates mass transfer rate, heat achieve 90% of CO2 removal efficiency. In order to minimize
transfer rate, diffusion rate, and reaction rate, which is beneficial ammonia loss and reduce the heat duty of regenerator, it is
to increase the CO2 removal efficiency, while the backward necessary to reduce the NH3 concentration of outlet gas and
reaction rates speed up at the same time, which will lead to the improve the CO2 loading of CO2-rich solutions as much as
ε decreasing. These inverse mechanisms result in the peak value possible. Besides, a rate-based absorption model, RateFrac,
of the CO2 removal efficiency. based on the Aspen Plus process simulator, is developed to
Effects of the temperature of flue gas on ε, CO2 loading, and simulate the pilot plant. Simulation results of the rate-based
NH3 concentration of outlet gas are shown in Figure 14. When model are in good agreement with the experimental data.
the temperature of flue gas increases from 25.5 to 150 °C, the ε, Further, optimization analysis for operating parameters of the
the CO2 loading, and the NH3 concentration of outlet gas pilot plant is carried out. The operation parameters include the
remain almost unchanged. Whereas the temperature of rich
packing height of the absorber, the ammonia concentration of
solution increases from about 31 to 37 °C, the increment of
lean solution, the CO2 loading of lean solution, the temperature
liquid temperature is not obvious due to the small specific heat
ratio between flue gas and solution. Therefore, before flue gas of liquid, and the flue gas flow rate. It is shown that process
entering into absorber, it does not need a cooling operation, optimization with the rate-based model can provide theoretical
and increasing the temperature of flue gas appropriately will be guidance for energy-saving technology.
beneficial to the removal of CO2.
5.4. Optimization of Gas Flow Rate. Figure 15 shows
effects of the flow rate of flue gas on ε, CO2 loading, and NH3
■
*
ASSOCIATED CONTENT
S Supporting Information
concentration of outlet gas. It is found that the CO2 removal
Details of operational parameters and results for the experi-
efficiency decreases from 100% to about 40% when the flow
rate of flue gas increases from 4 to 48 m3/h. The CO2 loading ments and simulations. This information is available free of
of rich solution grows quickly at first and later increases to a charge via the Internet at http://pubs.acs.org.
nearly stable value. The NH3 concentration of outlet gas
reaches to a peak value when the gas flow rate is about 30 m3/h
due to the high value of liquid temperature at the top of
■ AUTHOR INFORMATION
Corresponding Author
absorber as shown in Figure 15b. At the top of absorber (the
*Telephone: (8610) 6277 2112. Fax: (8610) 6278 1824. E-
first stage in rate-based model), absorbent contacts with flue gas
mail: [email protected].
that will leave the absorber immediately. Liquid and vapor tend
to thermodynamic equilibrium. Temperature has a remarkable Notes
effect on the vapor liquid equilibrium. A temperature increase is The authors declare no competing financial interest.
5317 dx.doi.org/10.1021/ie2030536 | Ind. Eng. Chem. Res. 2012, 51, 5309−5319
Industrial & Engineering Chemistry Research
■
Article
ACKNOWLEDGMENTS (12) Adams, D.; Davison, J. Capturing CO2. IEA Greenhouse Gas
R&D Programme Report; IEA: Cheltenham, U.K., 2007.
This research was supported by Beijing Municipal Commission (13) Gabrielsen, J.; Svendsen, H. F.; Michelsen, M. L.; Stenby, E. H.;
for Science & Technology under Grant No. Z08040902950803.
■
Kontogeorgis, G. M. Experimental validation of a rate-based model for
CO2 capture using an AMP solution. Chem. Eng. Sci. 2007, 62, 2397−
NOMENCLATURE 2413.
A, B, C, D = adjustable parameters (14) Bai, H.; Yeh, A. C. Removal of CO2 greenhouse gas by ammonia
DevExp = deviation of individual liquid and gas phase mass scrubbing. Ind. Eng. Chem. Res. 1997, 36, 2490−2493.
balances for CO2, % (15) Yeh, A. C.; Bai, H. Comparison of ammonia and monoethanol-
E = activation energy amine solvents to reduce CO2 greenhouse gas emissions. Sci. Total
GasCO2 = decreased amount of CO2 in absorber gas, mol/h Environ. 1999, 228, 121−133.
k = pre-exponential factor (16) Yeh, J. T.; Resnik, K. P.; Rygle, K.; Pennline, H. W. Semi-batch
absorption and regeneration studies for CO2 capture by aqueous
K = equilibrium constant
ammonia. Fuel Process. Technol. 2005, 86, 1533−1546.
LiqCO2 = increased amount of CO2 in absorber liquid, mol/h (17) Diao, Y. F.; Zheng, X. Y.; He, B. S.; Chen, C. H.; Xu, X. C.
n = temperature exponent Experimental study on capturing CO2 greenhouse gas by ammonia
r = rate of reaction scrubbing. Energy Convers. Manage. 2004, 45, 2283−2296.
R = universal gas constant (18) Mani, F.; Peruzzini, M.; Barzagli, F. The role of zinc(II) in the
T = absolute temperature absorption-desorption of CO2 by aqueous NH3, a potentially cost-
T0 = reference temperature effective method for CO2 capture and recycling. ChemSusChem 2008,
y1 = CO2 mole fraction at the inlet of absorber 1, 228−235.
y2 = CO2 mole fraction at the outlet or some position of (19) Niu, Z. Q.; Guo, Y. C.; Lin, W. Y. Experimental studies on
absorber removal of carbon dioxide by aqueous ammonia fine spray. Sci. Chin.
ε = CO2 removal efficiency, % Tech. Sci. 2010, 53, 117−122.
(20) Mani, F.; Peruzzini, M.; Stoppioni, P. CO2 absorption by
Subscripts aqueous NH3 solutions: speciation of ammonium carbamate,
eq = at equilibrium
■
bicarbonate and carbonate by 13C NMR study. Green Chem. 2006,
8, 995−1000.
REFERENCES (21) Kim, Y. J.; You, J. K.; Hong, W. H.; Yi, K. B.; Ko, C. H.; Kim, J.
(1) Houghton, J. T.; Ding, Y.; Griggs, D. J.; Noguer, M.; van der N. Characteristics of CO2 absorption into aqueous ammonia. Sep. Sci.
Linden, P. J.; Dai, X.; Maskell, K.; Johnson, C. A., Eds.; Climate Technol. 2008, 43, 766−777.
Change 2001: The Scientific Basis. Third Assessment Report of the (22) Park, H.; Jung, Y. M.; You, J. K.; Hong, W. H.; Kim, J. N.
IPCC; Cambridge University Press: Cambridge and New York, 2001. Analysis of the CO2 and NH3 reaction in an aqueous solution by 2D
(2) Kanniche, M.; Gros-Bonnivard, R.; Jaud, P.; Valle-Marcos, J.; IR COS:formation of bicarbonate and carbamate. J. Phys. Chem. A
Amann, J. M.; Bouallou, C. Pre-combustion, post-combustion and oxy- 2008, 112, 6558−6562.
combustion in thermal power plant for CO2 capture. Appl. Therm. Eng. (23) Qin, F.; Wang, S. J.; Hartono, A.; Svendsen, H. F.; Chen, C. H.
2010, 30, 53−62. Kinetics of CO2 absorption in aqueous ammonia solution. Int. J.
(3) Noorman, S.; Gallucci, F.; Annaland, M.; van, S.; Kuipers, H. J. A. Greenhouse Gas Control 2010, 4, 729−738.
M. Experimental investigation of a CuO/Al2O3 oxygen carrier for (24) Zeng, Q.; Guo, Y. C.; Niu, Z. Q.; Lin, W. Y. Mass transfer
chemical-looping combustion. Ind. Eng. Chem. Res. 2010, 49, 9720− coefficients for CO2 absorption into aqueous ammonia solution using
9728. a packed column. Ind. Eng. Chem. Res. 2011, 50, 10168−10175.
(4) deMontigny, D.; Tontiwachwuthikul, P.; Chakma, A. Comparing (25) Gal, E. Ultra cleaning combustion gas including the removal of
the absorption performance of packed columns and membrane CO2. World Intellectual Property, Patent WO 2006022885, 2006.
contactors. Ind. Eng. Chem. Res. 2005, 44, 5726−5732. (26) McLarnon, C. R.; Duncan, J. L. Testing of ammonia based CO2
(5) Tomizaki, K.; Shimizu, S.; Onoda, M.; Fujioka, Y. Heats of capture with multi-pollutant control technology. Energy Proc. 2009, 1,
reaction and vapor-liquid equilibria of novel chemical absorbents for
1027−1034.
absorption/recovery of pressurized carbon dioxide in integrated coal
(27) Park, S. Y.; Yi, K. B.; Ko, C. H.; Park, J. H.; Kim, J. N.; Hong, W.
gasification combined cycle-carbon capture and storage process. Ind.
H. Selection of optimal operating conditions for a continuous CO2-
Eng. Chem. Res. 2010, 49, 1214−1221.
capture process using an aqueous ammonia solution. Energy Fuels
(6) Hoff, K. A.; Juliussen, O.; Falk-Pedersen, O.; Svendsen, H. F.
Modeling and experimental study of carbon dioxide absorption in 2010, 24, 3704−3709.
aqueous alkanolamine solutions using a membrane contactor. Ind. Eng. (28) Yu, H.; Morgan, S.; Allport, A.; Cottrell, A.; Do, T.; McGregor,
Chem. Res. 2004, 43, 4908−4921. J.; Wardhaugh, L.; Feron, P. Results from trialling aqueous NH3 based
(7) Ho, M. T.; Allinson, G. W.; Wiley, D. E. Reducing the cost of post-combustion capture in a pilot plant at Munmorah power station:
CO2 capture from flue gases using pressure swing adsorption. Ind. Eng. Absorption. Chem. Eng. Res. Des. 2011, 89, 1204−1215.
Chem. Res. 2008, 47, 4883−4890. (29) Darde, V.; Thomsen, K.; Well, W. J. M.; Stenby, E. H. Chilled
(8) Kumar, K.; Dasgupta, C. N.; Nayak, B.; Lindblad, P.; Das, D. ammonia process for CO2 capture. Int. J. Greenhouse Gas Control 2010,
Development of suitable photobioreactors for CO2 sequestration 4, 131−136.
addressing global warming using green algae and cyanobacteria. (30) Mathias, P. M.; Reddy, S.; O’Connell, J. P. Quantitative
Bioresour. Technol. 2011, 102, 4945−4953. evaluation of the chilled-ammonia process for CO2 capture using
(9) Zhang, J.; Lee, J. W. Enhanced kinetics of CO2 hydrate formation thermodynamic analysis and process simulation. Int. J. Greenhouse Gas
under static conditions. Ind. Eng. Chem. Res. 2009, 48, 5934−5942. Control 2010, 4, 174−179.
(10) Tobiesen, F. A.; Juliussen, O.; Svendsen, H. F. Experimental (31) Puxty, G.; Rowland, R.; Attalla, M. Comparison of the rate of
validation of a rigorous desorber model for CO2 post-combustion CO2 absorption into aqueous ammonia and monoethanolamine.
capture. Chem. Eng. Sci. 2008, 63, 2641−2656. Chem. Eng. Sci. 2010, 65, 915−922.
(11) Ma’mun, S.; Jakobsen, J. P.; Svendsen, H. F.; Juliussen, O. (32) Zhang, Y.; Chen, H.; Chen, C. C.; Plaza, J. M.; Dugas, R.;
Experimental and modeling study of the solubility of carbon dioxide in Rochelle, G. T. Rate-based process modeling study of CO2 capture
aqueous 30 mass % 2-((2-aminoethyl)amino)ethanol solution. Ind. with aqueous monoethanolamine solution. Ind. Eng. Chem. Res. 2009,
Eng. Chem. Res. 2006, 45, 2505−2512. 48, 9233−9246.