Subscriber access provided by University of Rochester | River Campus & Miner Libraries
Process Engineering
Optimization of CO2 Capture from Simulated Flue Gas Using K2CO3/Al2O3 in a Micro Fluidized Bed Reactor Mohsen Amiri, and Shahrokh Shahhosseini Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.8b00789 • Publication Date (Web): 08 Jun 2018 Downloaded from http://pubs.acs.org on June 13, 2018
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
1
Optimization of CO2 Capture from Simulated Flue Gas
2
Using K2CO3/Al2O3 in a Micro Fluidized Bed Reactor
3
Mohsen Amiri, and Shahrokh Shahhosseini*
4
School of Chemical, Petroleum, and Gas Engineering, Iran University of Science and Technology,
5
P.O. Box 16765163, Tehran, 16846-13114, Iran.
6
ABSTRACT
7
The cost effective dry regenerable K2CO3/Al2O3 seems to be a promising sorbent for CO2 removal
8
from the flue gas of fossil fuel power plants. In this work, the characterization of carbonation
9
reaction and process optimization were performed in a so-called micro fluidized bed reactor
10
(MFBR), which has recently been applied to study gas–solid reactions. The sorbent was also
11
characterized by BET and SEM techniques. In addition, the most important gas-solid heterogeneous
12
models were evaluated and the kinetic parameters were determined by the model fitting approach.
13
Based on the kinetic study results, the homogeneous model (HM) and the shrinking core model
14
(SCM) were selected as the reaction models. Also, the effects of the independent variables including
15
temperature, gas flow rate, and vapor pretreatment amount on the responses (adsorption capacity and
1 ACS Paragon Plus Environment
Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
16
reaction rate constant) were investigated by the response surface methodology (RSM) coupled with
17
Box-Behnken-Design (BBD). Regarding to the analysis of variance (ANOVA) results, the
18
temperature and gas flow rate are the most important factors affecting the adsorption capacity and
19
the reaction rate constant, respectively. In addition, the semiempirical polynomials were developed
20
to find the optimum condition corresponding to the highest adsorption capacity and reaction rate.
21
Consequently, the optimum independent variables were 60 °C, 562 CCM and 22.2 mg H2O condition
22
for the temperature, gas flow rate, and vapor pretreatment amount. The best response values of 65.29
23
mg CO2/g sorbent and 0.3402 (min-1) were predicted for the adsorption capacity and reaction rate
24
constant at the optimum conditions which were verified experimentally. The presented results are
25
applicable and essential for future simulation and modeling CO2 capture in the fluidized bed reactor.
26
Keywords: CO2 capture; Micro fluidized bed reactor (MFBR); Optimization; Response surface
27
methodology (RSM); Adsorption; Kinetic.
28 29
1. INTRODUCTION
30
Global warming and climate change resulting from the greenhouse gas (GHG) emissions has
31
become a widespread concern in the last two decades.1 According to the prediction of
32
Intergovernmental Panel on Climate Change (IPCC) the temperature increases between 1.0 to 3.7 °C
2 ACS Paragon Plus Environment
Page 2 of 50
Page 3 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
33
through the 21 century.2, 3 Among the greenhouse gases, carbon dioxide (CO2) is the most significant
34
contributor to global warming.4, 5 The main source of anthropogenic CO2 emissions is the flue gas
35
from power plants burning fossil fuels.6, 7 CO2 capture and storage (CCS) is known as an imperative
36
option for decreasing CO2 emissions in recent years.8 Postcombustion is a key CCS technology
37
option for the conventional power plants.9
38
Postcombustion processes include mainly physical and chemical absorption, adsorption, cryogenic
39
and membrane processes.5 Among these technologies chemical absorption and adsorption are the
40
most appropriate ones for current power plants.1 Absorption using amine solutions is a matured
41
technology in petroleum, natural gas and power plants, although it has several issues including
42
solvent losses, corrosion, hazardous byproducts, and high-energy requirement for regeneration.5, 8, 10
43
Therefore, adsorption process using solid sorbents is an encouraging alternative in CO2 capture from
44
flue gas and also have a good potential in the future9.
45
Recently, multiple research groups have been involved in the development of solid physisorbent
46
for CO2 capture including activated carbonaceous materials, microporous/mesoporous silica or
47
zeolites and MOFs.
48
13% CO2, 9% H2O) due to the high attraction to the water vapor, low thermal and mechanical
49
stability, and low CO2/N2 selectivity or low CO2 adsorption capacities at moderately low CO2 partial
9, 11
However, these sorbents are not applicable under flue gas condition (78 N2,
3 ACS Paragon Plus Environment
Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
9, 11
Page 4 of 50
50
pressure.
An especially promising opportunity is the usage of dry alkali metal-based sorbents to
51
capture CO2 from flue gas stream. 8, 12, 13 Dry alkali metal carbonates like K2CO3 and Na2CO3 could
52
react with CO2 in the presence of water vapor and produce the alkali metal hydrogen carbonates salt
53
(KHCO3 or NaHCO3) at low temperatures through the reaction R1 14, 15:
54
M2CO3 + CO2 + H2O 2MHCO3
55
∆H = -141 kJmol-1 ⸫ M=K
56
In the carbonation step, CO2 could react with M2CO3 in the presence of water vapor at low
57
temperature range of (50-100 °C). MHCO3 decomposes at a moderate temperature change of 120-
58
200 °C and produce CO2/H2O mixture that could be converted into CO2 stream ready for
59
transportation and storage by condensing the water vapor.4, 12, 16 In addition, it is reported that K2CO3
60
shows more CO2 capture capacities than Na2CO3.16, 17 Researchers suggested several porous matrixes
61
as support including Al2O3, TiO2, Ac, SiO2, ZrO2, CaO and zeolites.18-23 Among them K2CO3/Al2O3,
62
appeared to possess a high porosity, mechanical strength and attrition resistance. It has a well-
63
developed microstructure and seems to be an almost perfect sorbent for fluidized bed CO2 capture
64
process.8, 17, 24
(R1) ∆H = -135 kJmol-1 ⸫ M=Na
65
According to the literature, most of the studies on potassium based sorbent have been conducted
66
using thermogravimetric analyzer (TGA)20, 25-30 or in a fixed bed reactor.15, 17-19, 23, 31-36 These studies
4 ACS Paragon Plus Environment
Page 5 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
67
provide a good comprehension of carbonation reaction behavior and the effect of operating factors.
68
The carbonation reaction of K2CO3/Al2O3 in a fixed bed reactor and the effect of operating condition
69
have been reported by Amiri et al.15 They found that the temperature and vapor pretreatment time are
70
the most significant carbonation process variables.
71
The so-called micro fluidized bed reactor (MFBR) has been successfully developed to study gas–
72
solid reactions characteristics and reaction kinetics.37-49 In comparison to TGA and fixed bed, MFBR
73
results in quicker heat and mass transfer between gas and particles, faster reaction heat removal and
74
less external interfacial diffusion limitations. MFBR also gives the capability of real-time
75
measurement of the gaseous product through an on-line gas analyzer.40, 49 Although few studies have
76
been performed on the CO2 capture using K2CO3/Al2O3 in fluidized bed reactors in bench scale,21, 50-
77
52
78
under different operating conditions in MFBR. The final goal for this technology is to develop a CO2
79
capture process with dual fluidized-bed reactors under optimum conditions to achieve maximum CO2
80
capture capacity and reaction rate. Performance studies describing the effect of different operating
81
variables on the adsorption capacity and reaction kinetics are essential to commercially apply CO2
82
capture with the fluidized bed technology.53
there is no systematic statistical study on carbonation reaction characteristics and reaction kinetics
5 ACS Paragon Plus Environment
Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
83
Page 6 of 50
Recently, response surface methodology (RSM) is successfully applied to estimate the individual
84
and interactive effects of multiple process variables and find the optimum conditions.15,
54-56
85
Therefore, in the present study the micro fluidized bed reactor (MFBR) was employed to evaluate the
86
carbonation characteristics of K2CO3/Al2O3 sorbent in the simulated flue gas. RSM with
87
Box−Behnken design (BBD) has been applied for the experimental design, proposing empirical
88
correlation and specification of the optimum values of the process variable (Temperature, gas flow
89
rate and vapor pretreatment time) for the desirable response variables (adsorption capacity and
90
reaction rate constant). Kinetic parameters were investigated based on the model-fitting approach.
91
The main objectives were to optimize the carbonation process in MFBR and evaluate the effect of
92
the process variables on adsorption capacity and reaction rate by proposing two semiempirical
93
correlations.
94 95 96
2. EXPERIMENTAL SECTION 2.1.
Solid Sorbent Preparation
97
The solid sorbent (K2CO3/ γ-Al2O3) was prepared by coating the conventional wet impregnation of
98
(K2CO3, Merck, 99% purity) on the porous particles (γ-Al2O3, Merck, 99% purity). The preparation
99
process was composed of four stages: (1) mixing and impregnation of in the aqueous mixture of
6 ACS Paragon Plus Environment
Page 7 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
100
deionized water and 16.2 g K2CO3 with a magnetic stirrer at 25°C for 14 h, (2) drying at 100°C in an
101
oven for 24 h, (3) calcination at 300°C for 4 h, and (4) graining and sieving of the particles for
102
fluidized bed tests. Additionally, the desired loading of K2CO3 on γ-Al2O3 particles was 35 wt. %.
103
2.2.
Sorbent Characterization
104
A PHS-1020 (PHSCHINA) system with N2 adsorption-desorption was applied to analyse surface
105
area and pore size distributions. The microscopic shape of the sorbent was examined by a Philips
106
XL30 scanning electron microscope (SEM). The impregnated amount of the K2CO3 on Al2O3 was
107
determined by a PHILIPS PW1480 X-ray fluorescence (XRF) system.
108
2.3.
Apparatus and Procedure
109
The cyclic carbonation reactions under isothermal condition were performed in a micro fluidized
110
bed reactor (MFBR) unit, as illustrated in Figure 1. The experimental apparatus primarily composed
111
of five segments: (1) gas supply section, (2) steam generator system, (3) temperature-controlled bath,
112
(4) micro fluidized bed reactor (MFBR), and (5) online CO2 analysis in the outlet stream. Feed gases
113
(N2 and CO2) were provided from high-purity (vol. 99.999%) cylinders and transmitted to the MFBR
114
with separate mass flow controllers (MFC). In addition, water feed was prepared by a high-precision
115
liquid pump and then preheated by heat tracing of the tube line to ensure full water vaporization
116
before mixing with N2 and CO2. Isothermal condition of reactor was established with a circulating 7 ACS Paragon Plus Environment
Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
117
fluid through the MFBR jacket. The reactor temperature was measured by two thermocouples at the
118
entrance and output of the MFBR. Dimensionally, the internal height and diameter of the bed zone
119
were initially 20 mm and 12 mm, respectively. Moreover, CO2 composition in the treated outlet
120
stream was continuously recorded by an online infrared (IR) analyzer (Vaisala, Finland,
121
measurement limit vol. 0-20%). Furthermore, the carbonation and regeneration tests were performed
122
at 60-80 and 300 °C, respectively. Based on the results of the previous study, H2O/CO2 mole ratio
123
was kept as one with a balanced amount of N2 and the cyclic tests were performed by the same
124
regenerated sorbent due to complete regeneration of the sorbent at 300 °C.15 The particle size of bed
125
material was in the range 90−106 µm. According to the Ergun equation, the minimum fluidization
126
velocity (Umf) and corresponding minimum fluidization flow rate of the sorbent particles were
127
calculated as 1.91 cm/s and 129.8 mL/min, respectively. Therefore, the total gas flow rate of the inlet
128
stream through MFBR was designed in the range of 450-650 mL/min to ensure fluidization of
129
particles.
8 ACS Paragon Plus Environment
Page 8 of 50
Page 9 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
130 131
Figure 1. Schematic diagram of the micro fluidized bed reactor (MFBR) apparatus
132
At the beginning of each carbonation experiment about 2 g of solid adsorbent was loaded into the
133
MFBR. Then, N2 gas was passed through the MFBR in order to reach the isothermal condition of the
134
sorbet particles and to prevent water vapor condensation at low temperature. Thereafter, N2 stream
135
with water vapor (mol. 10%) was passed through the bed till the amount of H2O during the vapor
136
pretreatment period reached to 12-24 mg. Subsequently, CO2 was added to the gas mixture of N2 and
137
H2O vapor and the carbonation reaction began through the sorbent fluidized bed and CO2 was
138
removed. Finally, the sorbent regenerated at 300 °C and prepared for the next experimental run. The
139
reaction characteristics including kinetic parameters were determined by analyzing the data collected
140
using the online gas analyzer.
9 ACS Paragon Plus Environment
Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 10 of 50
141
3. THEORETICAL BASIS
142
3.1.
143
CO2 capture capacity during carbonation reaction, Ac (mg CO2/g sorbent), was calculated through
144 145
Adsorption Capacity and Conversion
eq 1:
Ac =
1000C i w
t
∫ Q (1 −ψ (t) ) ρ.dt 0
mg CO2 g sorbent
(1)
146
Where w, Ci, Q, t, , and are the sorbent mass (g), inlet CO2 concentration (vol %), total gas
147
flow rate (cm3/min), time (min), dimensionless outlet CO2 concentration (C/Ci) and CO2 density
148
(1.55 g/cm3), respectively. Furthermore, CO2 removal conversion degree, X, was determined by eq
149
2: t
X =
150
∫0 (C i −C )Q dt tf
∫0 (C i −C )Q dt
t
=
∫ (1 −ψ (t ) ) dt ∫ (1 −ψ (t ) ) dt 0 tf
(2)
0
Where C, and tf are the outlet CO2 concentration (vol %) and final time (min), respectively.
151
10 ACS Paragon Plus Environment
Page 11 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
152
153 154
Energy & Fuels
3.2.
Kinetic Method
Generally, the overall kinetic rate of heterogeneous gas solid reactions under isothermal condition as a function of reaction temperature T (K) and conversion X, can be defined by eq 3:
dX = K(T) F(X) dt 155 156
(3)
Where F(X) is the differential reaction rate function depending on the mechanism and K(T) is the reaction rate constant determined by the Arrhenius equation in eq 4: K(T ) = K o e - E a
RT
(4)
157
Where Ko, R, and Ea are pre-exponential factor (s-1), gas constant (8.314 Jmol-1K-1), and apparent
158
activation energy (Jmol-1), respectively. In addition, the integral reaction model G(X) can be
159
determined by integrating eq 3 in the form of eq 5. G(X ) =
∫
X
o
dX = K(T) t F(X )
(5)
160
Accordingly, based on the conventional model-fitting method57, the experimental conversion and
161
time data can be used in eq 4 and K(T) is determined as the slop of linear curve G(X) ~ t. There are
162
twenty mechanism functions that have been commonly used for the kinetic parameters specification
163
of heterogeneous solid-state reactions38, 49, 53, 57-59, as illustrated in Table 1. In current study, the form
164
of G(x) with the highest average correlation coefficient (R2) in all tests is selected to be the
165
mechanism function that characterizes the carbonation reaction kinetics.
11 ACS Paragon Plus Environment
Energy & Fuels 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
166
Page 12 of 50
Table 1. Typical Mechanism Function Using in Gas-Solid Reactions No. Model Power law models 1 2 3 4 Phase interfacial reaction 5 Zero order (plate) 6 2D (cylindrical) 7 3D (spherical) Diffusion models 8 1D 9 2D Va lensi equation 10 3D(Jander) 11 3D(A-J) 12 3D(Ginstling–Broushtein) Reaction-order models 13 First order 14 3/2 order 15 Second order 16 Third order Nucleation models 17 n=1.5 18 n=2 19 n=3 20 n=4
F(X)
G(X)
Ref
4X3/4 3X2/3 2X1/2 2/3X-1/2
X1/4 X1/3 X1/2 X3/2
57
1 2(1-X)1/2 3(1-X)2/3
X 1-(1-X)1/2 1-(1-X)1/3
57
1/(2X) [-ln(1-X)]-1 3/2(1-X)2/3[1-(1-X)1/3]-1 3/2(1-X)(2/3)[(1+X)1/3-1]-1 3/2[(1-X)-1/3-1]-1
X2 X+(1-X)ln(1-X) [1-(1-X)1/3]2 [(1+X)1/3-1]2 (1-2/3X)-(1-X)2/3
1-X (1-X)3/2 (1-X)2 (1-X)3
-ln(1-X) 2[(1-X)-1/2-1] (1-X)-1-1 0.5[(1-X)-2-1]
1.5(1-X)[-ln(1-X)]1/3 2(1-X)[-ln(1-X)]1/2 3(1-X)[-ln(1-X)]2/3 4(1-X)[-ln(1-X)]3/4
[-ln(1-X)]2/3 [-ln(1-X)]1/2 [-ln(1-X)]1/3 [-ln(1-X)]1/4
57 57 57
57 57
57 38, 57 38, 57 38, 49 38, 57
57 57 57 57
38, 57 38, 57, 58 38, 57, 58 38, 57, 58
167 168
4. EXPERIMENTAL DESIGN: METHODOLOGY
169
Response surface method (RSM) was used for experimental design and optimization of the
170
carbonation process. RSM is a collection of statistic and arithmetical techniques beneficial for
171
optimizing the processes, which is used widely in recent years
172
variables, alone or with their interactions, on the measured responses are explained by RSM.
173
Furthermore, a multivariate empirical model based on the independent variables and their
174
interactions is generated as a quadratic polynomial equation shown in eq 6. 12 ACS Paragon Plus Environment
60
. The effects of independent
Page 13 of 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Energy & Fuels
3
3
3
y = α 0 + ∑ α i x i2 + ∑∑ α ij x i x j + e i =1
(6)
i =1 j