Experimental Study on the Reuse of Spent Rapidly Hydrated Sorbent

Sep 19, 2011 - desulfurization technique using circulating fluidized bed system. The sorbent ... with other calcium-based sorbent modified by coal fly...
0 downloads 0 Views 4MB Size
ARTICLE pubs.acs.org/est

Experimental Study on the Reuse of Spent Rapidly Hydrated Sorbent for Circulating Fluidized Bed Flue Gas Desulfurization Yuan Li, Kai Zheng, and Changfu You* Key Laboratory for Thermal Science and Power Engineering of Ministry of Education, Department of Thermal Engineering, Tsinghua University, Beijing100084, China

bS Supporting Information ABSTRACT: Rapidly hydrated sorbent, prepared by rapidly hydrating adhesive carrier particles and lime, is a highly effective sorbent for moderate temperature circulating fluidized bed flue gas desulfurization (CFB-FGD) process. The residence time of fine calcium-containing particles in CFB reactors increases by adhering on the surface of larger adhesive carrier particles, which contributes to higher sorbent calcium conversion ratio. The circulation ash of CFB boilers (α-adhesive carrier particles) and the spent sorbent (β and γ-adhesive carrier particles) were used as adhesive carrier particles for producing the rapidly hydrated sorbent. Particle physical characteristic analysis, abrasion characteristics in fluidized bed and desulfurization characteristics in TGA and CFB-FGD systems were investigated for various types of rapidly hydrated sorbent (α, β, and γ-sorbent). The adhesion ability of γ-sorbent was 50.1% higher than that of α-sorbent. The abrasion ratio of β and γ-sorbent was 16.7% lower than that of α-sorbent. The desulfurization abilities of the three sorbent in TGA were almost same. The desulfurization efficiency in the CFB-FGD system was up to 95% at the bed temperature of 750 °C for the β-sorbent.

1. INTRODUCTION Dry flue gas desulfurization has the advantages of low investment, little water consumption, little equipment corrosion and small area requirements with dry powder byproduct. The main disadvantage is low gassolid sulfuration reaction activity, leading to a low calcium conversion ratio and low desulfurization efficiency.1,2 Thus, it is very important to develop an easily prepared and economical sorbent with high sulfur reaction activity for industrial application of dry flue gas desulfurization. Hou et al.3 developed a moderate temperature dry flue gas desulfurization technique using circulating fluidized bed system. The sorbent was prepared by rapidly hydrating the mixture of coal fly ash and lime by the mass ratio of 4:1 for 2 h at ambient temperature and then drying for 1 h at 150 °C.4,5 Larger ash particles were used as adhesive carrier particles, which carried the fine calcium-containing particles in the CFB reactor. Compared with other calcium-based sorbent modified by coal fly ash,610 the hydration time and temperature are significantly reduced for rapidly hydrated sorbent. The desulfurization efficiency for a pilot-scale CFB-FGD system was 6783% at the bed temperature of 600800 °C and the Ca/S ratio of 2.0.11 However, the particle abrasion of rapidly hydrated sorbent in CFB reactor reduced the residence time of the fine calcium-containing particles, which reduced the calcium conversion ratio of the sorbent. Lee12 used the circulation ash from CFB boilers as the adhesive carrier particles instead of coal fly ash and reduced the mass ratio of coal fly ash and lime to 2:1. Experimental results showed that the circulation ash enhanced the adhesion between the fine calcium-containing particles and the adhesive carrier r 2011 American Chemical Society

particles, reduced the sorbent abrasion ratio and increased the sorbent desulfurization ability in TGA. However, the source of the circulation ash is inconvenient, which would limit the application of this sorbent. One potential solution is to reuse the spent sorbent as adhesive carrier particles to prepare the rapidly hydrated sorbent. Particle physical characteristic analysis, abrasion experiment and desulfurization experiment in TGA were conducted for the rapidly hydrated sorbent prepared by the spent sorbent and lime; desulfurization experiments on a pilot-scale CFB-FGD system was conducted to investigate the desulfurization ability of the rapidly hydrated sorbent in the CFB-FGD system.

2. EXPERIMENTAL SECTIONS 2.1. Sorbent Preparation and Desulfurization Experiment on the CFB-FGD System. The lime used in the experiment was

from Laishui, Hebei in China. The circulation ash, which was circulated in the CFB boiler, was from the thermal power plant of Tsinghua University. Adhesive carrier particles were mixed with lime to prepare the rapidly hydrated sorbent. The sorbent preparation steps were as follows.4,12 The lime, adhesive carrier particles, and water were mixed in a hydration mixer with continuously stirring at ambient Received: May 23, 2011 Accepted: September 19, 2011 Revised: September 18, 2011 Published: September 19, 2011 9421

dx.doi.org/10.1021/es2017285 | Environ. Sci. Technol. 2011, 45, 9421–9426

Environmental Science & Technology

Figure 1. Particle size distributions from a laser diffraction instrument (Malvern Mastersizer 2000).

temperature to produce sorbent slurry. The water/solid mass ratio was 1:1 and the mass ratio of adhesive carrier particles and lime was 2:1.The hydration time was 2 h. Then, the sorbent slurry was dried in an infrared desiccator to a water content less than 10%. The drying time was 1.5 h at a drying temperature of 150 °C. Moderate temperature desulfurization experiment was conducted on the pilot-scale CFB-FGD system. The pilot-scale CFB-FGD system (Supporting Information (SI) Figure S1) mainly includes the sorbent preparation subsystem, the flue gas generator subsystem and the CFB reactor. Detailed descriptions of the system were given in refs 3,11. Flue gas generated by the oil burner was mixed with a small amount of air to produce 600800 °C simulated flue gas. SO2 was added to the flue gas before the CFB reactor. The CFB reactor riser was 6 m high with a diameter of 0.305 m and a flue gas flow rate of 300 N m3/h. The flue gas passed through the CFB reactor and reacted with the sorbent and then went through the cyclone separator and the bag filter before emitting from the stack. The sorbent particles collected in the cyclone separator were fed back into the reactor for further circulation or drained out of the system. The O2, CO2, and SO2 concentrations in the flue gas were measured online at the CFB reactor inlet and outlet by a PS3400 type gas analyzer. The desulfurization efficiency was directly calculated from the inlet and outlet SO2 concentrations. First, rapidly hydrated sorbent (α-sorbent) was prepared by the circulation ash (α-adhesive carrier particles) and lime. The α-sorbent was used in CFB-FGD desulfurization experiments and the first spent sorbent (β-adhesive carrier particles) was collected. Second, β-sorbent was prepared by β-adhesive carrier particles and lime. The β-sorbent was used in CFB-FGD desulfurization experiments and the second spent sorbent

ARTICLE

(γ-adhesive carrier particles) was collected. Third, γ-sorbent was prepared by γ-adhesive carrier particles and lime. The relationship among the adhesive carrier particles and sorbent were shown in SI Figure S2. The experimental conditions: bed temperature, T = 600800 °C; bed superficial velocity, U0 = 2.5 m/s; inlet SO2 concentration, CSO2‑in = 1500 ppm; inlet CO2 concentration, CCO2‑in = 10%; inlet O2 concentration, CO2‑in = 10%; bed pressure drop, ΔP = 1000 Pa; Ca/S ratio, Ca/S = 1.5 or 1.0. 2.2. Abrasion Experiment. The abrasion of rapidly hydrated sorbent had significant influence on the sorbent desulfurization performance.12,13 A fluidized particle abrasion test bed (SI, Figure S3) was developed to simulate the sorbent abrasion characteristics in a CFB.14,15 The sorbent used for the abrasion experiment was first calcined for 30 min at 750 °C to simulate moderate temperature conditions. The fan (250W) pushed the air at ambient conditions into the bed (inner diameter 90 mm and height 1100 mm) to fluidize the precalcined rapidly hydrated sorbent. The large particles remained in the bed while the fine particles were entrained in the cyclone separator (d99 = 40 μm). Some fine particles from the fluidized bed were collected by the cyclone separator while other fine particles were collected by the bag filter (efficiency 99%). The test began with 200 g of sorbent being put into the bubbling fluidized bed with an air flow rate of 5 N m3/h. Some minutes later, the fan was stopped and the particles in the cyclone separator and the bag filter were measured. The total abrasion time was set to be 60 min.12 An abrasion ratio, Ra, was defined to reflect the sorbent mass change in the fluidized abrasion test bed. Ra ¼

mc þ mb  100% m0

ð1Þ

where mc represents the mass of particles collected by the cyclone separator, mb represents the mass of particles collected by the bag filter, and m0 is the initial mass of calcined sorbent. 2.3. Desulfurization Experiment in TGA. First, about 10 mg of sorbent was preheated to 750 °C in TGA under N2 atmosphere. After the sorbent weight became stable, the reaction gas was introduced into the TGA at a gas flow rate of 100 mL/min for a reaction time of 80 min. The reaction gas consisted of SO2 (1500 ppm), O2 (5%), and N2 (balance gas). CO2 was not added to the reaction gas because CO2 does not obviously affect the sorbent desulfurization ability at 750 °C since almost all desulfurization reaction products were CaSO4 at this condition.16 The sorbent desulfurization ability in TGA, Rd (mg/g), was represented by the mass of SO2 absorbed by per gram of sorbent. Since the desulfurization reaction product was CaSO4 for this condition, CaO(solid) + SO2(gas) + 1/2O2(gas) f CaSO4(solid), Rd can be calculated by eq 2. Rd ¼

ðmi  m0 Þ  64  1000 80  m0

ð2Þ

where mi represents the sorbent mass at a given time (mg) and m0 represents the initial mass of precalcined sorbent (mg). 2.4. Particle Characterization. A laser diffraction instrument (Malvern Mastersizer 2000) was used to evaluate the particle size distribution of the samples. A scanning electron microscope (SEM: KYKY-2800) was used to observe the particle surface micrographs. The specific surface areas and pore volume distribution were measured by the nitrogen adsorption method 9422

dx.doi.org/10.1021/es2017285 |Environ. Sci. Technol. 2011, 45, 9421–9426

Environmental Science & Technology

ARTICLE

Figure 2. Typical SEM (KYKY-2800) images of various particles.

using an ASAP2010 type BET analyzer and AUTOSCAN-33 type mercury porosimetry. A Bruker D8 Advance X-ray diffractometer (XRD) was used to characterize the sample components.

3. RESULTS AND DISCUSSION 3.1. Particle Physical Characteristics. The particle size distributions of various particles are shown in Figure 1. The particle size distributions of the three adhesive carrier particles were similar with an average diameter of about 250 μm and the Ca(OH)2 was smaller than 10 μm. Compared with the corresponding adhesive carrier particles, the volume ratio of the particles smaller than 10 μm increased significantly for the sorbent, indicating that part of the fine calcium-containing particles did not adhere to the adhesive carrier particles. Typical SEM images of various particles are shown in Figure 2. The three adhesive carrier particles had irregular, rough and porous surfaces. Many fine calcium-containing particles were adhered to the surfaces of the three adhesive carrier particles. The XRD results are shown in Figure 3. α-adhesive carrier particles mainly contain SiO2(q, Quartz), Al6Si2O13 (Mullite) and little CaAl2Si2O8 (a, Anorthite). β-adhesive and γ-adhesive carrier particles have similar components, including CaSO4 (s, Anhydrite), little CaCO3 (c, Calcite), and CaO (e, Lime). α-sorbent mainly contains Ca(OH)2 (p, Portlandite), SiO2+Al6Si2O13 (q), CaCO3 (c) and little cementitious components. Cementitious components, such as CaAl2Si2O8 3 4H2O (g, hydrated anorthite), were the products of pozzolanic reaction.1719 β-sorbent and γ-sorbent also have similar components, including CaSO4 (s) and more cementitious components (CaAl2Si2O8 3 4H2O, et al.). The main reason was that CaSO4 can promote the pozzolanic reaction between lime and adhesive carrier particles.19 The specific surface areas and pore volumes of various particles are shown in Table 1. Figure 4 shows the sorbent pore volume distributions. The measurement uncertainties of these results were 5%. The specific surface area of α-adhesive carrier particles was larger than that of β and γ-adhesive carrier particles and coal

fly ash. The main reason was that β and γ-adhesive carrier particles had been abraded in the CFB reactor which decreased the specific surface area of the particles. The pore volumes and pore volume distributions of α, β, and γ-adhesive carrier particles were similar and larger than those of coal fly ash. α, β, and γ-adhesive carrier particles had similar porous structures which were not significantly influenced by particle abrasion while the coal fly ash particles were more compact. The specific surface areas and pore volumes of α, β, and γ-sorbent were also similar and larger than those of coal fly ash sorbent, which could improve the desulfurization ability. 3.2. Adhesion Characteristic of Adhesive Carrier Particles. VPM10 represents the volume ratio of particles smaller than 10 μm to the total volume of particles, which can be obtained from the particle size distribution. ΔVPM10, the increase of VPM10 compared with the sorbent and the adhesive carrier particles was calculated, which is listed in Table 2. Since the lime particles were smaller than 10 μm, the lime particles that did not adhere to the adhesive carrier particles led to ΔVPM10. The adhesion characteristic of the adhesive particles was represented by ΔVPM10. The ΔVPM10 of γ-sorbent compared with γ-adhesive carrier particles was smallest, indicating that γ-adhesive carrier particles had the best adhesion characteristics. Compared with α-adhesive carrier particles, the adhesion characteristic of γ-adhesive carrier particles increased 50.1%, which indicated that the adhesion characteristic of adhesive carrier particles would increase when the spent sorbent was used as adhesive carrier particles. As shown in Table 1 and Figure 4, the specific surface of α-adhesive carrier particles was larger than those of the other three adhesive carrier particles. The pore volume and pore volume distribution of coal fly ash was smaller than those of α, β, and γ-adhesive carrier particles. The adhesion characteristic of coal fly ash (ΔVPM10 = 36.57%)13 was much worse than that of the other three adhesive carrier particles, indicating that the pore structure of the adhesive carrier particles had significant influence on the adhesion characteristic of the adhesive carrier particles while the influence of the specific surface area was negligible. 9423

dx.doi.org/10.1021/es2017285 |Environ. Sci. Technol. 2011, 45, 9421–9426

Environmental Science & Technology

ARTICLE

Figure 3. X-ray diffraction (Bruker D8 Advance) results of various particles.

Table 1. Particles Specific Surface Area and Pore Volume by the Nitrogen Adsorption Method at 77 K α-adhesive carrier particles

β-adhesive carrier particles

γ-adhesive carrier particles

Coal fly ash

specific surface area, m2/g

8.10

2.96

3.05

2.19

pore volume, cm3/g

0.0136

0.0121

0.0128

0.00491 coal fly ash sorbent

α-sorbent

β-sorbent

γ-sorbent

specific surface area, m2/g

17.5

15.0

16.7

5.60

pore volume, cm3/g

0.0540

0.0644

0.0495

0.0391

XRD results in Figure 3 show that β and γ-adhesive carrier particles had large amount of CaSO4 (Figure 5-b,c), which promoted the pozzolanic reaction and increased the production of cementitious components (Figure 5-e, f, CaAl2Si2O8 3 4H2O et al.).20 These cementitious components enhanced the adhesion force between the fine calcium-containing particles and the

adhesive carrier particles. Thus, the adhesion characteristic of β and γ-adhesive carrier particles increased. 3.3. Abrasion Resistant Characteristic of Sorbent. Figure 5 shows the abrasion test results of the three sorbent. The abrasion ratios of β and γ-sorbent were similar in 60 min (∼8.5%), which were lower than that of α-sorbent (10.2%). The sorbent abrasion 9424

dx.doi.org/10.1021/es2017285 |Environ. Sci. Technol. 2011, 45, 9421–9426

Environmental Science & Technology

ARTICLE

Figure 5. Abrasion ratios of various sorbent.

Figure 4. Pore volume distributions of various particles by the nitrogen adsorption method at 77 K.

Table 2. VPM10 Content of Various Particles adhesive carrier α-adhesive carrier β-adhesive carrier γ-adhesive carrier particles particles particles particles VPM10/%

1.33

2.27

sorbent

α-sorbent

β-sorbent

γ-sorbent

VPM10/%

13.2

9.76

5.92

ΔVPM10/%

11.9

7.49

5.92

0.00

resistant characteristic increased 16.7% due to the spent sorbent reuse as adhesive carrier particles. The reason was that the improved adhesion characteristic of β and γ-adhesive carrier particles enhanced the adhesion force between the fine calciumcontaining particles and the adhesive carrier particles, which was good for improving the abrasion resistant characteristic of the sorbent, increasing the residence time of fine calcium-containing particles in CFB and increasing the calcium conversion ratio of the sorbent. 3.4. Desulfurization Ability in TGA. The desulfurization abilities of α, β, and γ-adhesive carrier particles and sorbent in TGA are shown in Figure 6-a. The desulfurization abilities of the lime in various sorbent were calculated by deducting the desulfurization abilities of adhesive carrier particles from that of the sorbent, as shown in Figure 6-b. The initial reaction rate (010 min) of α-sorbent, which mainly was influenced by the specific surface area of the sorbent, was larger than those of β and γ-sorbent. The desulfurization abilities of the three sorbent were all about 313 mg/g in 80 min. The desulfurization ability of α-adhesive carrier particles was almost zero, while that of the β and γ-adhesive carrier particles was about 30 mg/g due to the existence of CaCO3 (c) and CaO (e) in the components of β and γ-adhesive carrier particles (Figure 3-b, c). The desulfurization abilities of the lime in α, β, and γ-sorbent were about 900 mg (SO2)/g (lime), while that of coal fly ash

Figure 6. Sorbent desulfurization abilities in TGA.

sorbent was 725 mg (SO2)/g (lime). This was because the pore volumes and pore volume distributions of the α, β, and γsorbent, especially the medium pore (250 nm) distribution that had significant influence on the sulfur reaction,20 were similar and larger than those of the coal fly ash sorbent, as shown in Figure 4-b. The sorbent desulfurization ability in TGA, where the abrasion phenomena would be negligible, provides ideal capacity for the sorbent desulfurization ability. Desulfurization experiments in the pilot-scale CFB-FGD system could provide more actual desulfurization ability of the sorbent. 3.5. Desulfurization Efficiency of the CFB-FGD System. The desulfurization efficiency at various bed temperatures for the CFB-FGD system is shown in Figure 7. The desulfurization efficiency in the CFB-FGD system increased from 68% to 83% when the α-sorbent was used instead of the coal fly ash sorbent at the bed temperature of 750 °C and the Ca/S ratio of 1.5. The corresponding sorbent calcium conversion ratio increased from 45% to 55%, confirming that the circulation ash, used as adhesive carrier particles instead of coal fly ash, can 9425

dx.doi.org/10.1021/es2017285 |Environ. Sci. Technol. 2011, 45, 9421–9426

Environmental Science & Technology

ARTICLE

’ ACKNOWLEDGMENT This research was supported by the Special Funds for Major State Basic Research Projects (No. 2006CB200305). ’ REFERENCES

Figure 7. Desulfurization efficiency of CFB-FGD system with various sorbent (U = 2.5 m/s, Fbed = 16.7 kg/m3, CSO2 = 1500 ppm, CCO2 = 10%, CO2 = 10%).

increase the sorbent calcium conversion ratio. The desulfurization efficiency of the CFB-FGD system with β-sorbent was about 10% higher than that with the α-sorbent. The calcium conversion ratio of β-sorbent achieved 80% at the bed temperature of 750 °C and the Ca/S ratio of 1.0. The reason was that the adhesion characteristic of β-adhesive carrier particles and the abrasion resistant characteristic of β-sorbent were improved, which significantly increased the residence time of the fine calciumcontaining particles in CFB. By comparing the particle components collected by the bagfilter in Figures 5-g and 5-h, the bag-filter particles from αsorbent contained large amount of Ca(OH)2 (p), while there was nondetectable diffraction of Ca(OH)2 in the bag-filter particles for β-sorbent. This phenomenon proved that the residence time of the fine calcium-containing particles of α-sorbent in CFB was relatively shorter than that of β-sorbent; the fine calcium-containing particles were quickly collected by the bag filter after the sorbent was fed in and could not react with SO2 effectively.

’ ASSOCIATED CONTENT

bS

Supporting Information. Schematic of the pilot-scale CFB-FGD system (Figure S1); Relationship between α,β,γadhesive carrier particles and α,β,γ-sorbent (Figure S2); Schematic of the fluidized abrasion test bed (Figure S3). This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Phone: +86-10-62785669; fax:+86-10-62770209; e-mail: youcf@ tsinghua.edu.cn.

(1) Matsushima, N.; Nishioka, Y. L.; Nishioka, M.; Sadakata, M.; Qi, H. Y.; Xu, X. C. Novel dry-desulfurization process using Ca(OH)2/fly ash sorbent in a circulating fluidized bed. Environ. Sci. Technol. 2004, 38 (24), 6867–6874. (2) Pandey, R. A.; Biswas, R.; Chakrabrti, T.; Devotta, S. Flue gas desulfurization: Physicochemical and biotechnological approaches. Crit. Rev. Environ. Sci. Technol. 2005, 35 (6), 571–622. (3) Hou, B.; Qi, H. Y.; You, C. F. Dry desulfurization in a circulating fluidized bed (CFB) with chain reactions at moderate temperatures. Energy Fuels 2005, 19 (1), 73–78. (4) Zhang, J.; Zhao, S. W.; You, C. F.; Qi, H. Y.; Chen, C. H. Rapid hydration preparation of calcium-based sorbent made from lime and fly ash. Ind. Eng. Chem. Res. 2007, 46 (16), 5340–5345. (5) Zhang, J.; You, C. F.; Zhao, S. W.; Chen, C. H.; Qi, H. Y. Characteristics and reactivity of rapidly hydrated sorbent for semidry flue gas desulfurization. Environ. Sci. Technol. 2008, 42 (5), 1705–1710. (6) Ali, Al-S.; Hitoki, M. Comparative reactivity of treated FBC-and PCC-fly ash for SO2 removal. Can. J. Chem. Eng. 1995, 73 (11), 678–685. (7) Paolo, D. Investigation of the SO2 adsorption properties of Ca(OH)2-fly ash systems. Fuel 1996, 75 (6), 713–716. (8) Ishizuka, T.; Tsuchiai, H.; Murayama, T.; Tanaka, T.; Hattori, H. Preparation of active absorbent for dry-type flue gas desulfurization from calcium oxide, coal fly ash, and gypsum. Ind. Eng. Chem. Res. 2000, 39 (5), 1390–1396. (9) Fernandez, J.; Renedo, M. J. Kinetic modelling of the hydrothermal reaction of fly ash, Ca(OH)2 and CaSO4 in the preparation of desulfurant sorbents. Fuel 2004, 83 (45), 525–532. (10) Lee, K. T.; Mohamed, A. R.; Bhatia, S.; Chu, K H. Removal of Sulfur Dioxide by fly ash/CaO/CaSO4 sorbents. J. Chem. Eng. 2005, 114 (13), 171–177. (11) Zhang, J.; You, C. F.; Qi, H. Y.; Hou, B.; Chen, C. H.; Xu, X. C. Effect of operating parameters and reactor structure on moderate temperature dry desulfurization. Environ. Sci. Technol. 2006, 40 (13), 4300–4305. (12) Li, Y.; You, C. F.; Song, C. X. Adhesive carrier particles for rapidly hydrated sorbent for moderate-temperature dry flue gas desulfurization. Environ. Sci. Technol. 2010, 44, 4692–4696. (13) Li, Y.; Song, C. X.; You, C. F. Experimental study on abrasion characteristics of rapidly hydrated sorbent for moderate temperature dry flue gas desulfurization. Energy Fuels 2010, 24, 1682–1686. (14) Chu, C. Y.; Hsueh, K. W.; Hwang, S. J. Sulfation and abrasion of calcium sorbent in a bubbling fluidized bed. J. Hazard. Mater. 2000, B80, 119–133. (15) Chu, C. Y.; Hwang, S. J. Abrasion and sulfation of calcium sorbent and solids circulation rate in an internally circulating fluidized bed. Powder Technol. 2002, 127, 185–195. (16) Hou, B.; Qi, H. Y.; You, C. F.; Fan, B. G.; Xu, X. C. The efficiency of CO2 and NO to dry desulfurization process at medium temperature. J. Eng. Thermo. 2004, 25 (6), 1061–1064. (17) Matinez, J. C.; Izquierdo, J. F.; Cunill, F.; Tejero, J.; Querol, J. Reactivation of fly ash and Ca(OH)2 mixtures for SO2 removal of flue gas. Ind. Eng. Chem. Res. 1991, 30 (9), 2143–2147. (18) Tsuchiai, H.; Ishizuka, T.; Ueno, T.; Hattori, H.; Kita, H. Highly active absorbent for SO2 removal prepared from coal fly ash. Ind. Eng. Chem. Res. 1995, 34 (4), 1404–1411. (19) Tsuchiai, H.; Ishizuka, T.; Nakamura, H.; Ueno, T.; Hattori, H. Study of flue desulfurization absorbent prepared from coal fly ash: Effects of the composition of the absorbent on the activity. Ind. Eng. Chem. Res. 1996, 35 (7), 2322–2326. (20) Renedo, M. J.; Fernandez, J.; Garea, A.; Ayerbe, A.; Irabien, J. A. Microstructural changes in the desulfurization reaction at low temperature. Ind. Eng. Chem. Res. 1999, 38 (4), 1384–1390. 9426

dx.doi.org/10.1021/es2017285 |Environ. Sci. Technol. 2011, 45, 9421–9426