Article pubs.acs.org/IECR
Fundamental Study of Indirect vs Direct Sulfation under Fluidized Bed Conditions Liyong Wang,† Shiyuan Li,‡ and Eric G. Eddings*,† †
Department of Chemical Engineering, University of Utah, Salt Lake City, Utah 84112, United States Institute of Engineering Thermophysics, Chinese Academy of Science, Beijing 100190, China
‡
ABSTRACT: Direct dry sorbent injection is a relatively simple and low-cost process for the capture of SO2 emissions, particularly for in-bed capture during fluidized bed combustion. SO2 capture by limestone may be accomplished by either direct or indirect sulfation, depending on gas-phase environment and operating conditions. In this paper, a combination of thermogravimetric analysis (TGA) and a bench-scale bubbling fluidized bed reactor were used to study sulfation, calcination and carbonation behavior of limestone at different CO2 concentrations and various temperatures to identify the relative contributions of indirect and direct sulfation reactions under air- or oxy-fired fluidized bed combustion conditions. Scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy analyses were also used to help identify product layer microstructure and composition. A significant temperature dependence on the sulfation of limestone was seen under oxy-fired conditions (when direct sulfation dominates); however, limited temperature dependence was seen for air-fired conditions (where indirect sulfation dominates). A distinct CaSO4 product layer is seen in SEM images for the air-firing condition that can significantly inhibit gas transport into internal pores. No such distinct product layer was seen for oxy-firing conditions. For the size range of limestone used in this study (600 to 1000 μm), indirect sulfation (dominant under air-fired conditions) appeared to be strongly limited by diffusion, whereas direct sulfation (dominant under oxy-fired conditions) was mainly under kinetic control or a mixture of kinetic and diffusion control.
1. INTRODUCTION Direct injection of limestone to the furnace under oxy-fuel combustion is a very attractive option for controlling SO2 emissions. Despite many earlier efforts (Borgwardt et al.,1,2 Dam-Johansen et al.,3,4 Silcox et al.,5−7 and Anthony et al.8,9), the mechanism for direct SO2 removal by limestone is still not fully understood. Generally, the sulfation mechanisms (direct or indirect) for limestone depend on whether the limestone passes through a calcination step. Usually, in fluidized bed combustion conditions, direct sulfation occurs under oxy-fuel combustion conditions due to CO2 inhibition of the calcination step, and indirect sulfation occurs under air-fired combustion conditions. During indirect sulfation, CaCO3 decomposes into CaO and CO2, followed by reaction of CaO and SO2 to form CaSO3, with subsequent oxidation to CaSO4, as shown in Mechanism I below. In direct sulfation, a high concentration of CO2 and/or a relatively low operating temperature (conditions that may be found under oxy-fired fluidized bed combustion) inhibits the decomposition of CaCO3. CaCO3 can thus directly react with SO2 to form CaSO3, with subsequent oxidation to CaSO4, as shown in Mechanism II below. I. Indirect sulfation (CaO−SO2) Step 1: calcination of CaCO3
CaSO3 +
1 O2 → CaSO4 2
II. Direct sulfation (CaCO3−SO2) Step 1: formation of sulfite CaCO3 + SO2 → CaSO3 + CO2 (g)↑
Step 2: oxidation of sulfite CaSO3 +
1 O2 → CaSO4 2
The operating temperature can have a significant effect on the sulfation mechanism. Direct sulfation can proceed at low temperatures (≤600 °C) under air-firing conditions; similarly, indirect sulfation can occur at high temperatures (≥920 °C) when oxy-firing. Numerous mechanisms have been proposed;5,10−17 however, there is not complete agreement on the details for the direct sulfation mechanism, particularly under oxy-firing conditions, as it can be difficult to confirm intermediate species along the way to forming CaSO4.18 The work of Liu et al.13,19,20 presented a drastic reduction of SO2 emissions during oxy-fuel firing. They proposed that limestone can maintain a high reactivity under a high CO2 partial pressure and suggested two conclusions: first, the high concentration of SO2 inhibits CaSO4 decomposition; second, the diffusion resistance through the CaSO4 layer is minimized
CaCO3 → CaO + CO2 (g)↑
Step 2: formation of sulfite Received: Revised: Accepted: Published:
CaO + SO2 → CaSO3
Step 3: oxidation of sulfite © 2015 American Chemical Society
3548
December March 17, March 26, March 26,
6, 2014 2015 2015 2015 DOI: 10.1021/ie504774r Ind. Eng. Chem. Res. 2015, 54, 3548−3555
Article
Industrial & Engineering Chemistry Research due to suppressed limestone calcination. The point was cited in multiple review papers.21,22 Snow et al.12 investigated the direct sulfation mechanism using thermogravimetric analysis (TGA) featuring a high CO2 concentration and Iceland spar limestone. They found that the diffusion resistance through the product layer at a high CO2 concentration (direct sulfation) was much lower than that with CaO under a N2 atmosphere (indirect sulfation). They postulated that the difference was due to CaSO4 product layer porosity variations in both scenarios, and suggested that CO2 generated during direct sulfation can keep the pore open or at least delay the pore closure or shrinking. However, Hu et al.18 doubted whether the CO2 generated is responsible for the porosity during direct sulfation. They argued that 1.5 mol of gaseous reactants were consumed to generate 1 mol of CO2. Thus, the direction of net flux was from surface to inner particle. The mechanism for sulfation of limestone under air-fired conditions is quite different from that under oxy-firing. The concept of a CaSO4 product layer in the context of a shrinking core model has been previously proposed to explain a direct sulfation mechanism.10,11 The purpose of this paper is to address whether the CaSO4 product layer is similar under conditions that lead to either direct or indirect sulfation.
Figure 1. Schematic of the bench-scale bubbling fluidized bed setup (BFB).
measured by a series of K-type thermocouples. The SO2 in the effluent stream was measured by a Magna-IR 550 Fourier transform infrared (FTIR) spectrometer. The preheated furnace was purged before experiments to ensure the desired air- (N2/O2) vs oxy- (CO2/O2) firing conditions. A constant inlet SO2 concentration was required in these experiments, and was achieved by use of a calibration gas that consisted of 1000 ppm of SO2 in either N2 or CO2. The experiments were run by loading 1.0 g of limestone into the reactor, which already contained approximately 300 g of preheated zirconium silicate bed material. The limestone was exposed to a N2 atmosphere for 15 min to allow for calcination. The CO2 concentration in the effluent gas was monitored by FTIR during this calcination period to ensure complete decomposition of the limestone. The CO2 signal would disappear after approximately 10 min, but the calcination period was run for the full 15 min to ensure complete decomposition. Upon completion of the calcination stage, the inlet gas flow was then switched to the desired premixed sulfurcontaining gas stream (either N2/O2/SO2 or CO2/O2/SO2) to commence the reaction period. The total reaction time for these semibatch experiments was approximately 60 min, and reactor temperatures of 765, 835, and 874 °C were used for both oxy- and air-firing environments. A summary of the experimental matrix is provided in Table 1. 2.3. Sulfated Limestone for Microstructural Examination. A FEI NovaNano FEG SEM 630 was employed to identify sulfated limestone microstructures. An energy dispersive X-ray spectrometer provided the sulfur distribution of the sulfated limestone. Sulfated limestone samples were mounted, cut in half, and polished gently to expose the inner microstructure for examination. The microstructures were characterized in terms of pore structure, size, formation, alignment, etc. The structural information was obtained to provide insight into the controlling mechanism of the indirect or direct sulfation reactions. Two SEM magnifications (20K×, 3.272K×) were utilized; the lower magnification provided an overview of the particles, whereas the higher magnification provided more details in specific locations as needed.
2. EXPERIMENTAL APPARATUS AND METHOD The investigation of limestone calcination is a prerequisite step for a mechanistic study of indirect and direct sulfation reactions. A TGA apparatus was used in these experiments to identify the relationship among calcination, temperature, and CO2 partial pressure. The sulfation behavior of limestone was studied using a bench-scale bubbling fluidized bed (BFB) reactor. The microstructure and sulfur distribution of the sulfated limestone was explored by scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDS). A more detailed description of these experimental setups is provided below. 2.1. Experimental Setup for Calcination vs Carbonation Studies. A TA Instruments SDT Q-600 TGA apparatus was used to study the calcination and carbonation behavior of limestone. We used reagent grade calcium carbonate (CaCO3) chips to provide a uniform representation of limestone, and the particle sizes were between 600 and 1000 μm. The limestone sample (70 mg) was loaded into a ceramic cup in the TGA furnace. The environmental gases were mixtures of N2 and CO2, and the mixture was controlled by calibrated gas rotameters. The CO2 concentration was adjusted to either 0, 50, 82, or 100%. The temperature profile for all of the TGA experiments was as follows: (1) The reactor temperature was increased from room temperature to 910 °C using a heating rate of 5 °C/min. (2) Then, the reactor temperature was maintained at 910 °C for 60 min. (3) After that, the reactor temperature was decreased from 910 to 700 °C using a heating rate of 5 °C/min. 2.2. Experimental Setup for Direct and Indirect Sulfation Studies. The experiments for limestone sulfation under air- vs oxy-firing conditions were performed in a benchscale bubbling fluidized bed reactor that has been described previously.23,24 The reactor, shown in Figure 1, consists of a vertical, cylindrical, stainless-steel chamber with a combustion chamber of inner diameter 44 mm, which is contained within an electrically heated furnace. There is a perforated plate at the base of the reactor that serves as a distributor for the fluidizing gas. The temperatures of the bed material and the gas were
3. RESULTS AND DISCUSSION 3.1. Calcination vs Carbonation Studies. The process of sulfation includes the important but competing reactions of calcination and carbonation. The TGA experiments performed very clearly illustrate how the rates for these two reactions vary with CO2 concentration and particle temperature, as shown in 3549
DOI: 10.1021/ie504774r Ind. Eng. Chem. Res. 2015, 54, 3548−3555
Article
Industrial & Engineering Chemistry Research Table 1. Experimental Matrix for Known SO2 Concentrations in N2/CO2 Environments Reacting with Limestone experimental parameters temperature in fluidized bed reactor (°C) O2 concentration (vol %) calibration gas (vol %) SO2 concentration in calibration gas (ppm) SO2 concentration in final mixture (ppm) bed material limestone (Alfa Aesar, 471-34-1) limestone weight (g) limestone size (mm) total gas flow rate (L/min)
765, 835, 874 20 80 1000 800 300 g of zirconium silicate (BSLZ-3) lab grade chips, calcium carbonate >99% 1.0 0.6−0.99 mm 3.0
Figure 2. Calcination and carbonation of limestone at varying concentrations of CO2 in N2.
Figure 2. In the figure, it is quite clear that limestone calcination is occurring in the 100% N2 environment, as illustrated by the decrease in sample weight. The limestone weight loss begins at a TGA reactor temperature of approximately 620 °C. The limestone sample loses 44% of its initial weight, and the calcination is essentially complete, because CO2 accounts for 44% of the total weight of the initial CaCO3. No change is observed after 44% of the weight is lost. Because reactiongenerated CO2 is purged in the TGA, no sign of carbonation is observed, even during the cooling-down period. If CO2 increases from 0 to 50%, as shown in Figure 2, the temperature at which calcination initiates is higher due to the increased CO2 partial pressure. Because of the presence of significant quantities of CO2, the carbonation reaction is in competition with the calcination reaction. As the temperature increases, the rate of calcination becomes greater than that of carbonation, and a weight loss is observed. When the final temperature of 910 °C is reached, about 44% of the total weight is lost, indicating a completion of calcination. The temperature is held for 1 h and no further weight loss is observed. During the temperature ramp-down, carbonation eventually becomes possible and the sample weight starts to increase again. The
weight increases from 56% to 76%, before a third weight plateau forms. It is not exactly clear what causes the first step in the weight loss profiles observed for runs involving CO2 in Figure 2. A possible explanation is the competition between carbonation and calcination reactions. The calcination process begins at a lower temperature, similar to what is observed with 0% CO2, but soon the competition with the carbonation reaction effectively balances the calcination rate. The temperature continues to rise and once a high enough temperature is reached, the calcination rate begins to exceed that of the carbonation reaction, and we again see continued weight loss. When we use a higher CO 2 concentration (82%), carbonation becomes more competitive. An initial phase of calcination is seen in Figure 2 before it levels off at about the same temperature as in the 50% CO2 case. A further weight loss follows, but starts at a higher temperature (898 °C) than in the 50% CO2 case (862 °C). The higher starting temperature is the result of a higher CO2 concentration. Carbonation and calcination compete with each other, and the relative contribution of each rate depends upon the temperature. At a high enough temperature, the rate of calcination exceeds that of carbonation, and the limestone decomposes until reaching 3550
DOI: 10.1021/ie504774r Ind. Eng. Chem. Res. 2015, 54, 3548−3555
Article
Industrial & Engineering Chemistry Research
Table 2. Primary Solid Products from Equilibrium Calculations for CaCO3 in Mixtures of CO2 in N2 at Temperatures Used in TGA Experiments CO2 concentration (%) temperature (°C)
0 750 800 850 910
CaO CaO CaO CaO
(100%) (100%) (100%) (100%)
50
81.9
100
CaCO3 (100%) CaCO3 (100%) CaO (100%) CaO (100%)
CaCO3 (100%) CaCO3 (100%) CaCO3 (100%) CaO (100%)
CaCO3 (100%) CaCO3 (100%) CaCO3 (100%) CaO (100%)
another weight plateau, whereas calcination is complete for these conditions. A considerably longer time is required for calcination to complete, and the steep slope seen with 0 and 50% CO2 is no longer seen due to increased competition with the reverse reaction. As the temperature decreases, the weight percent of the limestone increases from 59 to 82% as carbonation becomes more competitive again. When we use 100% CO2, carbonation becomes even more favorable. At T = 910 °C, the calcination rate still exceeds that of carbonation; however, the advantage is so small that after 60 min, only a 15% weight loss is recorded. Once the cooling period begins, the weight of limestone increases from 79 to 88%, due to the increased contribution of carbonation at the slightly lower temperatures. In summary, carbonation competes with calcination under various CO2 concentrations. Higher CO2 concentrations delay calcination and promote carbonation even at high temperatures. These experimental results are consistent with results of thermochemical equilibrium calculations, as shown in Table 2. The table indicates the primary solid-phase equilibrium product for the various temperatures and gas phase concentrations used in the TGA experiments. At the highest temperature (910 °C), CaCO3 is completely decomposed into CaO over the full range of CO2 concentrations (0, 50, 82, 100%). At the lowest temperature (750 °C), the CaCO3 is fully calcined to CaO in the absence of CO2 (0% case, but 100% CaCO3 is predicted in the presence of the high CO2 concentrations used in the experiments (50, 82, 100%). At the intermediate temperature of 850 °C, the CaCO3 is fully calcined to CaO at 50% CO2, but remains uncalcined at the higher CO2 levels (82, 100%). 3.2. Indirect Sulfation and Direct Sulfation Studies. The bubbling fluidized bed (BFB) reactor was used to study the sulfation behavior of limestone in either air- or oxy-fired conditions at different temperatures, as described above. The competition between calcination and carbonation reactions during the reaction of SO2 with limestone gave rise to different sulfation behavior, as will be discussed below. The experimental results are presented in terms of the degree of sulfation at time t, f t, which can be calculated using eq 1: ft =
PMCaCO3 0 RTmCaCO 3
∫0
t
Figure 3. Effect of temperature on the degree of limestone sulfation for air-fired conditions (SO2/O2/N2).
Figure 4. Effect of temperature on the degree of limestone sulfation for oxy-fired conditions (SO2/O2/CO2).
dependence on temperature also suggests a diffusion-controlled regime. The degree of sulfation for oxy-firing conditions is shown in Figure 4, which reveals a more clear dependence on temperature. Similar to the air case, the level of sulfation is quite low. A more detailed comparison between air and oxy cases is provided in Figures 5 and 6. At 765 °C, the rate of sulfation in the presence of CO2 is much less than in the presence of N2. After 1 h, the degree of sulfation when oxyfiring is about 12%, as compared to air-firing at about 15%. By contrast, a higher degree of sulfation is observed for oxy combustion at 874 °C. The conversion of limestone reaches more than 17%, higher than the 14% obtained in the air case. Different sulfation mechanisms dominate for air- and oxy-fired conditions. The indirect sulfation mechanism likely dominates when air-firing, whereas the direct mechanism likely dominates in oxy combustion. Also, it appears that the mechanism during air combustion is diffusion controlled, as discussed earlier. We suspect that the direct sulfation mechanism is mainly kinetically controlled, because the temperature dependence is much more significant. The reaction of SO2 and CaCO3 is slow under oxy-
Q ·(co − ct ) × 10−6dt (1)
The degree of sulfation obtained in the BFB as a function of time for both air- and oxy-fired combustion conditions is illustrated in Figures 3 and 4. When air firing, the reactor temperature shows no significant effect on the degree of sulfation between 765 and 874 °C. The degree of sulfation is quite low at 15% after 1 h. It is well established that significant pore plugging will limit the degree of sulfation because of the larger molar volume of CaSO4. Therefore, the reaction will take place at the particle surface, and will be inhibited inside of a thin outer layer. The low degree of sulfation shown in Figure 3 is consistent with this hypothesis. In addition, the very limited 3551
DOI: 10.1021/ie504774r Ind. Eng. Chem. Res. 2015, 54, 3548−3555
Article
Industrial & Engineering Chemistry Research
Figure 5. Comparison of the degree of limestone sulfation between N2/O2/SO2 and CO2/O2/SO2 environments at T = 765 °C.
Figure 6. Comparison of the degree of limestone sulfation between N2/O2/SO2 and CO2/O2/SO2 environments at T = 874 °C.
firing conditions. This lower reaction rate will result in a longer reaction window prior to pore plugging. Therefore, SO2 can be transported more deeply inside the particle due to a reduced diffusion limitation, and allow for the formation of CaSO4 over a larger internal surface area. 3.3. Microstructure of Sulfated Limestone by SEM and EDS. Differences in the fundamental sulfation mechanisms for oxy vs air combustion can also be evaluated by analysis of the microstructures (pore size, percent of void, alignment, sulfur distribution) using SEM and EDS. As mentioned previously, when operating under air-fired conditions (indirect sulfation), a rapid formation of CaSO4 near the surface will plug pores and inhibit sulfation of the inner portion of the particle. Thus, an evaluation of the microstructure should indicate a more dense product layer near the surface of the particle, and also a less uniform distribution of sulfur as a function of radius. For oxy-fired conditions, we have hypothesized that the overall rate is mainly under kinetic control, or a combination of kinetic and diffusion control, but not limited by diffusion alone. In this case, SO2 will penetrate deeper into the particle and will not plug pores due to CaSO4 formation as quickly. If this is true, we would not expect to see a dense outer product layer for these conditions, but instead, a more uniform distribution of CaSO4 (as evidenced by the presence of sulfur) as a function of radius. SEM imaging was performed on a large sampling of particles for each of the various conditions studied. Each mounted sample had on the order of 50 limestone particles, and 10−20 particles were closely examined for each condition. Figure 7 presents SEM photomicrographs of a representative sulfated
Figure 7. SEM photomicrograph of sulfated limestone with a magnification rate of 20000× with 800 ppm of SO2, 20% O2, and 80% N2, T = 874 °C (air-fired conditions). (a) At the edge of a particle; (b) in the center of a particle.
limestone particle under air-fired conditions (N2/O2/SO2) using a magnification of 20K×. Images were recorded at the edge of the particle (a) and near the center of the particle (b). Comparison of these two figures indicates a notable difference in pore structure between the outer edge and center of the particle. Figure 7a provides evidence of a relatively compact pore structure, which is consistent with the formation of a CaSO4 product layer that has a greater volume than the unreacted CaO, thus filling in pore space. Thus, it is apparent that in the presence of N2/O2/SO2, the indirect sulfation process is providing a diffusional resistance to additional sulfation in the center of the particle. 90% of the particles imaged for the air-fired conditions illustrated this type of product layer. The depth of the sulfation layer was measured using metric tools within the SEM imaging software, and it was found to be somewhat variable, even in adjacent regions on the same particle. We attribute this variability to the irregular surface of the limestone particles, coupled with the thin CaSO4 layer, which could be worn off through attrition by extended interaction (up to 1 h) with the much harder zirconium silicate bed material in the fluidized bed reactor. However, the measurements indicated the dense product layer to be on the order of 10 to 20 μm for the conditions studied. It should be 3552
DOI: 10.1021/ie504774r Ind. Eng. Chem. Res. 2015, 54, 3548−3555
Article
Industrial & Engineering Chemistry Research
O2/SO2). Comparison of these two images indicates distinct differences from the air-fired sulfated limestone particle. The pore structure between the edge and center of the particle is similar, and provides evidence of reaction throughout the particle (compare Figures 7a and 8a). CaSO 4 product formation is seen throughout, and has not formed the very compact pore structure at the surface seen in the air-fired conditions that can provide a diffusional barrier. Thus, it is apparent that in the presence of CO2/O2/SO2, the direct sulfation process does not appear to form a diffusional barrier and that the reaction is not dominated by diffusion control, and is under more kinetic control or a combination of kinetic and diffusion control, consistent with the temperature sensitivity seen in the BFB experiments. The differences in diffusional resistance observed in these experiments is also due to the larger particle sizes typical of fluidized bed operation, and would be less pronounced for limestone injection using much finer particles; e.g., with dry sorbent injection in downstream sulfur control operations. The sulfur distribution in the limestone particles was scanned by EDS, to provide further insight into the extent of sulfation within the particles. The sulfur distributions are shown in Figure 9 for air-fired (a) and oxy-fired (b) conditions. Elemental sulfur is represented by the green color, and the EDS photomicrographs were prepared with a lower magnification rate of 3272× to display a larger cross section of the particle. The differences in sulfur distribution are subtle in these images, as there is an overall background level (as noted in the region outside of each particle) that complicates the analysis. However, it is clear that there is a highly concentrated region of sulfur at the surface of the limestone particle from the air-fired experiments (Figure 9a). The presence of cracks or fissures in the particle also allow for some sulfation along those discrete regions as well. For the oxy-fired particle (Figure 9b), there is also some accumulation of sulfur near the surface, although not as intensely as in the airfired case. Although subtle, there is an increased presence of sulfur within the particle relative to the air-fired case, and a very significant presence of sulfur in some internal cracks or voids. The overall conversion for sulfation of both particles is quite low (less than 20%), and thus it is more difficult to identify the distinctions between the indirect (diffusion-controlled) and direct (less diffusion-controlled) sulfur distributions. Lower reaction rates for the direct sulfation reaction led to less
noted that these results were obtained in the absence of steam, which could have an impact on the observations reported here. Figure 8 provides images from the edge (a) and center (b) of a particle that was sulfated under oxy-fired conditions (CO2/
Figure 8. SEM photomicrograph of sulfated limestone with a magnification rate of 20000× with 800 ppm of SO2, 20% O2, and 80% CO2, T = 874 °C (oxy-fired conditions). (a) At the edge of a particle; (b) in the center of a particle.
Figure 9. EDS photomicrographs of sulfur distribution at a magnification rate of 3272 × , T = 874 °C. (a) Air-fired conditions with 800 ppm of SO2, 20% O2, and 80% N2; (b) oxy-fired conditions with 800 ppm of SO2, 20% O2, and 80% CO2. 3553
DOI: 10.1021/ie504774r Ind. Eng. Chem. Res. 2015, 54, 3548−3555
Article
Industrial & Engineering Chemistry Research
and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof. The authors thank Dana Overacker and Dr. Hongzhi Zhang for all their advice, guidance, and help.
product layer formation at the particle surface and thus less of a diffusion barrier, such that SO2 could penetrate more effectively into the particle and provide a somewhat more uniform CaSO4 conversion throughout the particle, instead of high conversion only near the surface resulting in the dense surface layer. The direct sulfation produced a higher “activity” than the indirect sulfation after a long reaction time under the conditions we tested due to the ability of SO2 to continue to diffuse into the particle at longer times in the absence of the dense surface product layer.
■
Variable Definition
co ct
4. CONCLUSIONS Experimental results were presented on the differences between air-fired and oxy-fired conditions for limestone calcination and carbonations reactions, as well as limestone sulfation reactions. The relevance of indirect vs direct sulfation mechanisms for the reaction of SO2 with limestone was also discussed in the context of the two firing conditions. It was observed that the temperature for onset of calcination increases with increasing CO2 concentration, and that the degree of calcination decreases with increasing CO2 concentration. It was also observed that the temperature for the onset of recarbonation of calcined limestone increases with increasing CO2 concentration. The effect of temperature on the degree of sulfation of limestone was much greater under direct sulfation conditions (oxy-firing) than under indirect sulfation conditions (air-firing). Due to the larger particle sizes used for fluidized bed conditions (600−1000 μm), it was possible to form an external product layer for these particles relative to fine limestone particles. For air-fired conditions, the product layer at the edge of the particle showed limited porosity, and thus, the sulfation process was strongly limited by diffusion, which has lower temperature sensitivity than a kinetically controlled process. In oxy-fired conditions, there was greater porosity at the surface and a more uniform distribution of sulfation within the particle. SEM images confirmed the presence of a compact product layer at the surface of limestone particles for air-firing conditions that blocks gas transport into pores. The images showed that the surface layer was much less compact when oxy firing. It is anticipated that the competition between calcination and carbonation reactions in the presence of CO2 inhibits calcination and promotes more emphasis toward the direct sulfation reaction when operating under oxy-fired conditions. Lower reaction rates for the direct sulfation reaction led to less product layer formation at the surface and thus less of a diffusion barrier, such that SO2 could penetrate more effectively into the particle and provide a more uniform CaSO4 conversion throughout the particle, instead of high conversion only near the surface, resulting in the dense surface layer. It should be noted that these results were obtained in the absence of steam, which could have an impact on the observations reported.
■
TABLE OF NOMENCLATURE
ft m0CaCO3 MCaCO3 P Q R T
■
Initial SO2 concentration (ppm) before reacting with limestone SO2 concentration (ppm) after reacting with limestone at t time Degree of sulfation at time t Initial mass of CaCO3 (kg) Molecular weight of CaCO3 (kg/kg mol) Pressure in the reactor (Pa) Total flow rate at reactor T and P (m3/s) Universal gas constant ((Pa·m3)/(Kelvin·kg mol)) Temperature in the reactor (Kelvin)
REFERENCES
(1) Borgwardt, R. H.; Rochelle, G. T. Sintering and sulfation of calcium silicate-calcium aluminate. Ind. Eng. Chem. Res. 1990, 29 (10), 2118−2123. (2) Dong, Y.; Borgwardt, R. H. Biomass reactivity in gasification by the hynol process. Energy Fuels 1998, 12 (3), 479−484. (3) Hu, G.; Dam-Johansen, K.; Wedel, S. Kinetics of the direct sulfation of limestone at the initial stage of crystal growth of the solid product. AIChE J. 2011, 57, 1607−1616. (4) Hu, G.; Shang, L.; Dam-Johansen, K.; Wedel, S.; Hansen, J. P. Initial kinetics of the direct sulfation of limestone. AIChE J. 2008, 54, 2663−2673. (5) Silcox, G. D.; Kramlich, J. C.; Pershing, D. W. A mathematical model for the flash calcination of dispersed calcium carbonate and calcium hydroxide particles. Ind. Eng. Chem. Res. 1989, 28 (2), 155− 160. (6) Milne, C. R.; Silcox, G. D.; Pershing, D. W.; Kirchgessner, D. A. Calcination and sintering models for application to high-temperature, short-time sulfation of calcium-based sorbents. Ind. Eng. Chem. Res. 1990, 29 (2), 139−149. (7) Milne, C. R.; Silcox, G. D.; Pershing, D. W.; Kirchgessner, D. A. High-temperature, short-time sulfation of calcium-based sorbents. 2. Experimental data and theoretical model predictions. Ind. Eng. Chem. Res. 1990, 29 (11), 2201−2214. (8) Wang, C.; Jia, L.; Tan, Y.; Anthony, E. J. Influence of water vapor on the direct sulfation of limestone under simulated oxy-fuel fluidizedbed combustion (FBC) conditions. Energy Fuels 2011, 25 (2), 617− 623. (9) Jia, L.; Tan, Y.; Anthony, E. J. Emissions of SO2 and NOx during oxy-fuel CFB combustion tests in a mini-circulating fluidized bed combustion reactor. Energy Fuels 2010, 24, 910−915. (10) Zhong, Q. Direct sulfation reaction of SO2 with calcium carbonate. Thermochim. Acta 1995, 260, 125−136. (11) Hajaligol, M. R.; Longwell, J. P.; Sarofim, A. F. Analysis and modeling of the direct sulfation of calcium carbonate. Ind. Eng. Chem. Res. 1988, 27 (12), 2203−2210. (12) Snow, M.; Longwell, J. P.; Sarofim, A. F. Direct sulfation of calcium carbonate. Ind. Eng. Chem. Res. 1988, 27, 268−213. (13) Liu, H.; Katagiri, S.; Kaneko, U.; Okazaki, K. Sulfation behavior of limestone under high CO2 concentration in O2/CO2 coal combustion. Fuel 2000, 79 (8), 945−953. (14) Illerup, J. B.; Dam-Johansen, K.; Lundén, K. High-temperature reaction between sulfur dioxide and limestoneVI. The influence of high pressure. Chem. Eng. Sci. 1993, 48 (11), 2151−2157. (15) de Diego, L. F.; Rufas, A.; García-Labiano, F.; de las ObrasLoscertales, M.; Abad, A.; Gayán, P.; Adánez, J. Optimum temperature for sulphur retention in fluidised beds working under oxy-fuel combustion conditions. Fuel 2013, 114, 106−113.
AUTHOR INFORMATION
Corresponding Author
*E. G. Eddings. E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This material is based upon work supported by the Department of Energy under Award Number DE-NT0005015. The views 3554
DOI: 10.1021/ie504774r Ind. Eng. Chem. Res. 2015, 54, 3548−3555
Article
Industrial & Engineering Chemistry Research (16) Lupiáñez, C.; Guedea, I.; Bolea, I.; Díez, L. I.; Romeo, L. M. Experimental study of SO2 and NOx emissions in fluidized bed oxyfuel combustion. Fuel Process. Technol. 2013, 106, 587−594. (17) Duan, L.; Jiang, Z.; Chen, X.; Zhao, C. Investigation on water vapor effect on direct sulfation during wet-recycle oxy-coal combustion. Applied Energy 2013, 108, 121−127. (18) Hu, G.; Dam-Johansen, K.; Wedel, S.; Peter Hansen, J. Review of the direct sulfation reaction of limestone. Prog. Energy Combust. Sci. 2006, 32 (4), 386−407. (19) Liu, H.; Shiro, K.; Ken, O. Drastic SOx removal and influences of various factors in O2/CO2 pulverized coal combustion system. Energy Fuels 2001, 15, 403−412. (20) Liu, H.; Okazaki, K. Simultaneous easy CO2 recovery and drastic reduction of SOx and NOx in O2/CO2 coal combustion with heat recirculation. Fuel 2003, 82 (11), 1427−1436. (21) Stanger, R.; Wall, T. Sulphur impacts during pulverised coal combustion in oxy-fuel technology for carbon capture and storage. Prog. Energy Combust. Sci. 2011, 37 (1), 69−81. (22) Toftegaard, M. B.; Brix, J.; Jensen, P. A.; Glarborg, P.; Jensen, A. D. Oxy-fuel combustion of solid fuels. Prog. Energy Combust. Sci. 2010, 36 (5), 581−625. (23) Can, G.; Salatino, P.; Scala, F. A single particle model of the fluidized bed combustion of a char particle with a coherent ash skeleton: Application to granulated sewage sludge. Fuel Process. Technol. 2007, 88 (6), 577−584. (24) Sánchez, A.; Eddings, E.; Mondragón, F. Fourier transform infrared (FTIR) online monitoring of NO, N2O, and CO2 during oxygen-enriched combustion of carbonaceous materials. Energy Fuels 2010, 24 (9), 4849−4853.
3555
DOI: 10.1021/ie504774r Ind. Eng. Chem. Res. 2015, 54, 3548−3555
