Systematic Approach to Determination of Maximum Achievable

Article pubs.acs.org/est

Systematic Approach to Determination of Maximum Achievable Capture Capacity via Leaching and Carbonation Processes for Alkaline Steelmaking Wastes in a Rotating Packed Bed Shu-Yuan Pan,† Pen-Chi Chiang,† Yi-Hung Chen,‡ Chun-Da Chen,§ Hsun-Yu Lin,§ and E.-E. Chang*,∥ †

Graduate Institute of Environmental Engineering, National Taiwan University, 71 Chou-Shan Rd., Taipei City, Taiwan 10673, Taiwan (R.O.C.) ‡ Department of Chemical Engineering and Biotechnology, National Taipei University of Technology, 1, Sec. 3, Zhongxiao E. Road, Taipei City, Taiwan 10608, Taiwan (R.O.C.) § Chins Steel Corporation, 1 Chung Kang Road, Hsiao Kang, Kaohsiung 81233, Taiwan (R.O.C.) ∥ Department of Biochemistry, Taipei Medical University, 250 Wu-Hsing Street, Taipei City, Taiwan 110, Taiwan (R.O.C.) S Supporting Information *

ABSTRACT: Accelerated carbonation of basic oxygen furnace slag (BOFS) coupled with cold-rolling wastewater (CRW) was performed in a rotating packed bed (RPB) as a promising process for both CO2 fixation and wastewater treatment. The maximum achievable capture capacity (MACC) via leaching and carbonation processes for BOFS in an RPB was systematically determined throughout this study. The leaching behavior of various metal ions from the BOFS into the CRW was investigated by a kinetic model. In addition, quantitative X-ray diffraction (QXRD) using the Rietveld method was carried out to determine the process chemistry of carbonation of BOFS with CRW in an RPB. According to the QXRD results, the major mineral phases reacting with CO2 in BOFS were Ca(OH)2, Ca2(HSiO4)(OH), CaSiO3, and Ca2Fe1.04Al0.986O5. Meanwhile, the carbonation product was identified as calcite according to the observations of SEM, XEDS, and mappings. Furthermore, the MACC of the lab-scale RPB process was determined by balancing the carbonation conversion and energy consumption. In that case, the overall energy consumption, including grinding, pumping, stirring, and rotating processes, was estimated to be 707 kWh/t-CO2. It was thus concluded that CO2 capture by accelerated carbonation of BOFS could be effectively and efficiently performed by coutilizing with CRW in an RPB.

1. INTRODUCTION The increased global average CO2 concentration in the atmosphere is likely to cause further warming and induce many changes in the global climate system.1 As a result, carbon capture, utilization, and storage (CCUS) technologies are key strategies to attenuate the impacts of global warming during the transition period for developing sustainable energy technologies.2−5 Among the CCUS technologies, accelerated carbonation (also referred to as mineral sequestration) of natural minerals and/or industrial alkaline wastes is attractive because gaseous CO2 is fixed as a solid precipitate and rarely released after mineralization due to the thermodynamic stability of solid carbonates.6−8 Because the carbonation reaction is regarded as © 2013 American Chemical Society

diffusion controlled (i.e., mass-transfer limited), a rotating packed bed (RPB) reactor was introduced to improve the mass transfer rate among phases due to its high centrifugal forces and excellent micromixing ability.9,10 As a result, the carbonation conversion of basic oxygen furnace slag (BOFS) in an RPB was found to be greater than that in an autoclave or slurry reactor.4,11−14 In addition, the carbonation reaction could be further enhanced by coupling with the cold-rolling wastewater Received: Revised: Accepted: Published: 13677

July 26, 2013 October 27, 2013 October 28, 2013 October 28, 2013 dx.doi.org/10.1021/es403323x | Environ. Sci. Technol. 2013, 47, 13677−13685

Environmental Science & Technology

Article

and 840−1190 μm, and then dried at 105 °C for 3 h in an oven. Physicochemical properties of BOFS with different particle sizes used in this study were presented in Table S1 (see Supporting Information). The detailed methods of particle size distribution, density, and BET surface analyses for fresh BOFS were found in the literature.11,12 The BOFS is rich in CaO based on the XRF analysis (i.e., 48.16% for less than 125 μm, 47.24% for 125−350 μm, 46.49% for 350−500 μm, 45.30% for 500−840 μm, and 43.59% for 840−1190 μm). On the other hand, both DIW and alkaline CRW were used as liquid agents to evaluate the respective leaching behavior of various metal ions from BOFS matrix into solution. Physicochemical properties of CRW were presented in Table S2 (see Supporting Information), which indicated that the measured pH values of CRW ranged from 11.20 to 11.87. It suggests that the alkaline properties of both CRW and BOFS should be beneficial to carbonation. 2.2. Leaching and Carbonation Experiments. The leaching experiments of BOFS were carried out using different liquid agents, i.e., DIW and CRW, to evaluate the leaching behavior of various ions into different solutions. In all leaching experiments, 100 g of BOFS was mixed in 2 L of CRW by mechanical stirring at 500 rpm for 90 min. After leaching experiments, the leachate was filtered through PTFE membrane filters (Millipore, 45-μm pore size and 47-mm diameter) to separate BOFS from solution. The leachate was then acidified with HNO3 (15.2 N) to a pH less than 2 to avoid further chemical precipitation. The concentration of various metal ions, e.g., Ca, Na, K, Zn, Pb, Al, Fe, Ni, Mg, and Cr, in the solution after acidification was analyzed with inductively coupled plasma atomic emission spectroscopy (ICP-AES, JOBIN YVON, JY24). The carbonation experiments of BOFS with different particle sizes were conducted in an RPB, as shown in Figure S1 (see Supporting Information), with a batch of slurry containing CRW and a fixed amount of BOFS at a given liquid-to-solid (L/ S) ratio (mL g−1). In this study, high-purity CO2 and N2 gases were mixed together with mass flow controllers (MFC) to simulate the CO2 concentration in hot-stove flue gas (i.e., 30 vol%), according to our previous study at China Steel Corp.12 CO2 was injected from the outside of the RPB reactor continuously at 101.3 kPa and a specific constant flow rate with a reaction time of 20 min. The detailed experimental procedures regarding leaching, carbonation, and analytical methods can be found in our previous research work.36 2.3. QXRD Using Rietveld Refinement. QXRD by the Rietveld method was carried out to determine the phase fraction of mineral crystal in BOFS. The XRD data of all BOFS samples were obtained by a Bruker D8 Advance iffractometer under the conditions described in Table S3 (see Supporting Information). The X-ray source was a Cu anode operated at 40 kV and 40 mA using Cu Kα radiation. Data collections were carried out between 20° and 80° in 2θ with a step of 0.01° and count time of 7.8 s per step. The Rietveld method was performed by the General Crystal Structure Analysis System (GSAS) software with the EXPGUI program, as shown in Figure S2 (see Supporting Information). GSAS was developed by Larson and Von Dreele37 for fitting atomic structural models to single crystal and powder diffraction data. The crystal structure parameters used to interpret the XRD patterns in GSAS were taken from the ICSD (Inorganic Crystal Structure Database). The collection codes for each structure: brownmillerite (Ca2Fe1.014Al0.986O5, code

(CRW) because of its alkaline property.12 Consequently, the CO2 removal efficiency (i.e., the percentage of CO2 removal from the emission source) in the flue gas by BOFS/CRW in the RPB process was 96−99% with a retention time of less than 1 min under ambient temperature and pressure conditions.12 However, the proposed RPB process would consume electricity for equipment such as the rotating bed and pumps. Therefore, the energy consumption for each unit in the RPB process should be critically evaluated to maximize the overall CO2 capture capacity. On the other hand, considerable research has been carried out on solid wastes and/or industrial byproducts such as BOFS from iron and steel manufacturing industries in various domains: accelerated carbonation,15−19 utilization assessment,20−22 landfilling,23 and environmental impact.24,25 Although these research studies were performed for a variety of purposes, a common point is that the material characterization, including physical, chemical, and mineralogical properties, should be determined in advance. However, quantitative analysis of mineral crystals in solid waste is very difficult to conduct with accuracy and precision due to their complex composition. The Rietveld method has been shown to be a powerful tool for quantitative analysis and crystal structure refinement based on X-ray diffraction (XRD) patterns,26 which has been widely used on various well-known materials27 and solid wastes.28−32 The quantitative analysis of mineral crystals using the Rietveld method can be executed by a number of efficient programs such as GSAS,29,30 SIROQUANT,29,33,34 and Maud.31 Because the Rietveld method is a full-pattern analysis of an XRD diffractogram, the relative weight fractions of crystalline phases in a multiphase sample can be calculated directly using scale factors for the respective calculated intensities. For instance, the study reported by Kuusik et al.35 indicated that the chemical and quantitative XRD (QXRD) analyses were in relatively good agreement on the composition of oil-shale ash, and the latter can be used for preliminary and rapid analyses. Similar observations were reported by Mahieux et al.,31 which indicates that consistent results were obtained by both physicochemical analysis and the Rietveld method for the mineral composition of sewage sludge ash (SSA) and municipal solid waste incineration fly ash (MSWI-FA). This suggests that the mineral composition of complex mineral waste should be accurately quantified by the Rietveld method. In this study, the leaching behavior of various metal ions from BOFS matrix into different types of liquid agents such as deionized water (DIW) and CRW was evaluated using a firstorder kinetic model. Meanwhile, the process chemistry of accelerated carbonation for BOFS with CRW in an RPB was investigated by evaluating the weight fraction of various mineral phases before and after carbonation using the Rietveld method. In addition, qualitative characterization of BOFS before and after carbonation was carried out using SEM and XEDS mappings. Furthermore, the maximum achievable capture capacity (MACC) for CO2 capture by the lab-scale RPB and the energy consumption was estimated to evaluate its feasibility in planning a pilot-scale plant.

2. MATERIALS AND METHODS 2.1. Materials. Both the alkaline BOFS and CRW were provided by China Steel Corp. (Kaohsiung, Taiwan). The BOFS was ground and sieved into different particle sizes, i.e., less than 125 μm, 125−350 μm, 350−500 μm, 500−840 μm, 13678

dx.doi.org/10.1021/es403323x | Environ. Sci. Technol. 2013, 47, 13677−13685


Article

98836); portlandite (Ca(OH)2, code 73468); calcite (CaCO3, code 169933); wollastonite (CaSiO3, abbreviated as C1S, code 240469); β-larnite (Ca2SiO4, abbreviated as C2S, code 245080); α-dicalcium silicate hydrate (Ca2(HSiO4)(OH), abbreviated as C2-S-H, code 75277); wustite (FeO, code 633038). The atom parameters including scale factors, background coefficients, zero-shifting error, lattice parameters, profile shape parameters, atomic site occupancies, and phase fractions in the unit cell were refined simultaneously in GSAS. The Rietveld algorithm utilizes the least-squares method to minimize the weighted sum of squared differences (M) between the observed (YO,i) and calculated (YC,i) intensity values in XRD diffractograms, as shown in eq 1:38,39 i=1

M=

∑ wi[YO,i − YC,i]2

(1)

n

where wi is the weight of each observation point and n is the number of observation points. Accordingly, the standard uncertainty for YO,i, i.e., σ[YO,i], can be obtained by measuring the YO,i intensity for an infinite number of times. To evaluate the goodness of model fit, a statistical method, i.e., “Chi squared” or χ2 test, was introduced as follows:39 χ2 =

1 n

i=1

∑ n

Figure 1. Leaching concentrations of calcium ion in DI and CRW under different particle sizes of BOFS for 90 min. Percentages shown on each bar represent the fraction of leaching concentration in totally leaching concentration.

thereby resulting in a greater carbonation reaction rate and higher CO2 capture capacity. Accordingly, the leaching concentrations of the major elements such as Ca, Na, K, Zn, Pb, Al, Fe, Ni, Mg, and Cr in CRW were measured with different particle sizes of BOFS at different leaching times, as shown in Figure 2. The leaching concentrations of various metal ions were observed to increase rapidly in the first 10 min and then gradually approach a maximum concentration. Therefore, the leaching kinetics of various metal ions (i) was evaluated by the mass loss-based method, as shown in eq 3:

2

[YO, i − YC, i] σ 2[YO, i]

(2)

The Rietveld refinement would gradually make the χ2 convergent to 1 during the refinement process. It was noted that χ2 would never drop below or equivalent to 1, if the crystallographic model is correct and chemically reasonable.39 2.4. SEM, XEDS, and Mapping. The fresh and carbonated BOFS samples were examined qualitatively by a scanning electron microscope (SEM)-equipped X-ray energy dispersive spectrometer (XEDS). The morphology and elemental distribution analyses of BOFS were carried out with a JEOL JSM-6340F field-emission SEM equipped with XEDS. The samples before and after carbonation were mounted with double-sided carbon tape on an aluminum stub and coated with a thin layer of platinum. The operating conditions were kept at a constant voltage of 15 kV and current density of 45 μA/cm2, with the electron beam directed at 90° to the specimen. In addition, the elemental composition of BOFS was detected by the XEDS. Mapping of Ca, Mg, Fe, Si, C, and O was carried out to investigate the distribution of these elements on the sample particles.

ri =

dC i = k i[Cmax ,i − C i]ni dt

(3)

where the ri is the leaching rate of various metal ions, ki is the rate constant of leaching, Cmax,i (mg/L) is the maximum leaching concentration for various metal ions in solution, Ci (mg/L) is the leaching concentration of various metal ions where the background concentration of metal ions originally in the solution was subtracted, and ni is the order of leaching reaction of various metal ions. Finally, eq 3 can be integrated as follows: C i = Cmax ,i[1 − e−k it ],

for

n=1

C i = Cmax ,i − [Cmax ,i1 − n − (1 − n)k it ]1/(1 − n) ,

3. RESULTS AND DISCUSSION 3.1. Leaching Behavior of BOFS in CRW. In general, the process chemistry of accelerated carbonation for BOFS can be divided into three categories: (a) gaseous CO2 dissolution into aqueous solution, (b) leaching of metal ions from BOFS into aqueous solution, and (c) carbonate precipitation. The leaching concentration of calcium ions in DIW and CRW was compared using different particle sizes of BOFS, as shown in Figure 1. The results indicate that a maximum Ca concentration of 2600 ppm was measured in the alkaline CRW with a particle size of BOFS less than 125 μm. The leaching of metal ions from the solid waste into solution would be higher with smaller particle size. Meanwhile, the high concentration of Na+ and Cl− in CRW might accelerate the leaching behavior of Ca-bearing phases in BOFS. This suggests that the leaching concentration of calcium ions in CRW should be higher than that in DIW,

n≠1

(4a)

for (4b)

Table S4 (see Supporting Information) presents the values of Cmax, k, n, and determination coefficient (r2) for various metal ions leaching from different particle sizes of BOFS into CRW, which indicates that the leaching concentration of various metal ions can be well expressed by eqs 4a or 4b, with r2 values ranging from 0.95 to 0.99. In general, the leaching behaviors of Zn, Pb, Al, Ni, Cr, and Mg fit eq 4a quite well regardless of particle size. On the other hand, the leaching behaviors of Ca, Na, K, and Fe were found to be more sensitive to the concentration driving force than that of others because the obtained n values were greater than 1. In general, both values of Cmax and k were observed to increase as the particle size of BOFS decreased. Ca (876−2690 mg/L), Na (827−1177 mg/ 13679



Article

Figure 2. Effect of leaching time and particle size of BOFS on leaching concentration of different metal ions (e.g., Ca, Na, K, Zn, Pb, Al, Fe, Ni, Mg, and Cr) in CRW for 90 min.

L), and K (201−232 mg/L) ions were found to be the major ions leaching from BOFS. Because BOFS contained a great amount of calcium hydroxide (i.e., 19.82 ± 0.85%) based on TG analysis,11 it might account for most of the Ca leaching concentration. It also was noted that the reactive Ca-bearing phases were lime, portlandite, and larnite, whereas brownmillerite was generally the most insoluble phase in BOFS.40 On the other hand, it seems that the leaching concentrations of Zn (63.2−79.0 mg/L) and Pb (19.0−43.0 mg/L) did not vary with time. Compared to Ca, Na, and K ions, Zn and Pb are considered minor elements that might not affect the carbonation reaction. Al, Fe, and Mg are considered slightly released cations, because the measured concentrations (i.e., 14.1−22.6 mg/L for Al, 9.3−9.9 mg/L for Fe, and 1.7−3.0 mg/ L for Mg) in the solution remain low during the entire leaching time. It could be explained by the low mobility of Mg−Fe−Al− Si-O, in which Mg and Al are commonly associated with Febearing phases in BOFS. In addition, Mg release is quite low (i.e., 3.0 ppm for 10.3) Ca2+(aq) + HCO3−(aq) → CaCO3(s) + H+(aq) (major reaction at 6.3 < pH < 10.3) Ca2(HSiO4)(OH)(s) + 4 Cl−(aq) + 3 H2O(l) → 2 CaCl2 (aq) + H4SiO4 (s) + 4 OH−(aq) CaSiO3(s) + 2 Cl−(aq) + 3 H2O(aq) → CaCl2 (aq) + H4SiO4 (s) + 2OH−(aq) Ca2(Fe, Al)2O5(s) + 4 Cl−(aq) + 4 H2O(l) → 2 CaCl2 (aq) + H4(Fe, Al)2O5(s) + 4 OH−(aq) CaCl2 (aq) → Ca2+(aq) + 2 Cl−(aq) CaO(s) + CO2(g) + H2O(l) → CaCO3(s) + H2O(l)

−81.89 −109.48

−26.90 −56.10

6 7

−68.49 13.07 27.92

−35.25 −47.41 11.63

8 9 10

−17.53b

−

11

8.34

60.06

12

−

−

13

0.15 −178.30

−0.04 −130.41

14 15

Ca(OH)2(s) + CO2(g) + H2O(l) → CaCO3(s) + 2 H2O(l) Ca2(HSiO4)(OH)(s) + 2 CO2(g) + H2O(l) → 2 CaCO3(s) + H4SiO4(aq) CaSiO3(s) + CO2(g) + 2 H2O(l) → CaCO3(s) + H4SiO4(aq) Ca2(Fe, Al)2O5(s) + 2 CO2(g) + 2 H2O(l) → 2 CaCO3(s) + H4(Fe, Al)2O5(s)

−113.13 −203.45 −87.92 −

−73.08 − −43.49 −

16 17 18 19

a

Thermodynamic properties of minerals and related substances were at 298.15 K and 1 atm. bHeat of formation for C2-S-H is adapted from Newman.42

lead to the formation of CaCl2 (eqs 11 to 13), which would further dissociate back to Cl− and release Ca2+, as shown in eq 14. Therefore, the carbonation of Ca2Fe1.014Al0.986O 5, C2-S-H, and C1S phases in BOFS could be enhanced by CRW, which might be attributed to the presence of Cl− ions in CRW and then the increase in the Ca2+ leaching capacity. A similar observation reported in the literature44−46 suggests that the presence of inorganic ionic species in solution such as Na+ and Cl− can promote the dissolution of silicate-bearing minerals due to the formation of surface complexes, leading to the reductive (and oxidative) dissolution of minerals. It was thus concluded that the overall process chemistry of aqueous carbonation for BOFS in CRW could be expressed as eqs 15 to 19. In addition, Figure 3a and 3b show the cross-section observations of BOFS before and after carbonation, respectively, by SEM/XEDS. It was clearly observed that, before carbonation, the entire BOFS is rich in calcium-ferrous-silicate and/or calcium-magnesium-silicate but without carbon element. After carbonation, the BOFS exhibits rhombohedral crystals, with a size of 1 to 3 μm, formed uniformly on the surface of the BOFS, exhibiting a CaCO3 layer (reacted outside) and a metal-rich core (unreacted inside). Furthermore, Figure S6a and S6b (see Supporting Information) present the elemental mappings of the fresh and carbonated BOFS, respectively, where Ca, Mg, Fe, Si, C, and O were recorded during the scanning of samples. The distribution of the above chemical elements was observed to be inhomogeneous from particle to particle. Meanwhile, the distribution of the calcium is quite concentrated in the cases of both fresh and carbonated BOFS. Generally, the distribution percentage of the carbon on the surface of the carbonated BOFS is found to be higher than that on the fresh BOFS, which indicates that the CO2 can be captured successfully by the carbonation reaction. The cubicshaped crystals coating the surface of carbonated BOFS were

and Ca2Fe1.04Al0.986O5 in BOFS can be regarded as the major species reacting with CO2 to form CaCO3 precipitation. According to the above observation, the process chemistry of accelerated carbonation for BOFS coutilizing with CRW was summarized and proposed as shown in Table 1. At first, the leaching of Ca(OH)2 and CaO in BOFS would directly generate the Ca2+ and OH− in the solution, as shown in eqs 5 and 6, respectively. Second, gaseous CO2 can rapidly dissolve into the alkaline solution (i.e., CRW), where the predominant carbonate ions (CO32−) could reduce the pH of the solution, as indicated in eq 7. Because the CO2 continuously dissolved into the solution during the carbonation, the pH value would decrease gradually to 6.3,8 where the bicarbonate ions (HCO3−) were found to be dominated, as shown in eq 8. Finally, CaCO3 would be formed by reacting the calcium ions with the carbonate ions (CO32−) and bicarbonate ions (HCO3−) in the solution under a high pH and circum-neutral condition, respectively, as expressed by eqs 9 and 10. On the other hand, Ca2Fe1.014Al0.986O5, C2-S-H, and C1S phases do not react with water at normal temperature and pressure; therefore, the leaching of Ca2+ and O2− (or OH−) from the BOFS solid matrix into solution is very low. However, after carbonation, the contents of the above mineral phases (i.e., Ca2Fe1.014Al0.986O5, C2-S-H, and C1S) were reduced significantly according to the results of the Rietveld refinement. Therefore, it suggests that CaCO3 should be directly generated in the course of carbonation reaction without the formation of intermediate products such as Ca(OH)2, in the cases of Ca2Fe 1.014Al0.986O5, C2-S-H , and C1S phases, which was in good agreement with the findings reported in the literature.43 Additionally, the leaching concentration of calcium species in CRW was found to be higher than that in DIW due to the high concentration of chloride ions (Cl−) in CRW. As a result, the above mineral phases might react with the Cl− in the CRW and 13681



Article

Figure 3. Cross-section observations of BOFS (a) before and (b) after carbonation by SEM/XEDS.

RPB could be obtained by considering both the carbonation conversion and energy consumption. It was noted that, because the small CaCO3 particles that formed on the surface of BOFS accounted for the formation of a reacted layer around the reacting particles according to the SEM observations, further leaching of reactive oxide species from the inner unreacted core of BOFS was hindered. Therefore, the carbonation kinetics can be expressed by an “exponential growth to maximum” model due to the formation of a product layer, during the carbonation reaction, as reported in the literature.8,9,13 On the other hand, the overall energy consumption of the RPB process was found

composed of calcium, carbon, and oxygen elements, indicating the formation of calcium carbonate (calcite). As a result, the observations of SEM/XEDS and mappings are in good agreement with the results of QXRD using the Rietveld method. It suggests that the BOFS should be carbonated successfully with CO2 in an RPB, and the carbonated products are mainly calcite (CaCO3) according to the observation of XRD, SEM-XEDS, and mapping. 3.3. Determination of Maximum Achievable Capture Capacity (MACC). In this study, the maximum achievable capture capacity (MACC) of BOFS coupled with CRW in an 13682



Article

Figure 4. (a) Maximum achievable capture capacity (MACC) and (b) estimation of energy consumption for accelerated carbonation of BOFS coupled with CRW in an RPB (as indicated by red line).

Information) by balancing “exponential growth of capture capacity (positive capture)” and “linear increase of energy consumption (negative capture)”.

to increase linearly with the increase of reaction time. Therefore, the “MACC” could be systematically determined and graphically presented in Figure S7 (See Supporting 13683



Article

Table 2. Summary of Power Consumption and Cost Estimation for 1 Ton of CO2 Captured by the Proposed RPB Process process

equipment

grinding stirring

crusher (ball mill)a laboratory stirrer (Corning, Model PC-410)b motor (Masterflex, digital economy driver with model 77800-50)b RPB (TECO, 3-phase induction motor, AEAJQU)b

pumping rotation

product specification/ equipment scale

power consumption per unit process (kWh) (1)

treated capacity (t/h) (2)

capacity per unit process (t/h) (3)

scale factor (−) (4)

power consumption (kWh) (1) × (2) × (4)/(3)

cost estimation ($)c

≤125 μm lab-scale

43.28 0.00169

5.13 107.69

1 0.0021

1 0.7

222.03 60.82

17.74 4.86

lab-scale

0.00334

107.69

0.0021

0.7

120.02

9.59

lab-scale

0.00848

107.69

0.0021

0.7

304.37

24.32

707.24

56.51

total a

Estimated from Bond’s equation and work index of BOFS is assumed to be 60.8 kWh/ton according to the available data provided by China Steel Corp. bMeasured from existing equipment. cTaiwan average electricity price for industry in 2011 was US $0.0799 per kWh.

the pumps and the grinding process to the total energy consumption is 17% and 31%, respectively. Although the RPB process for carbonation of BOFS/CRW would require additional electricity, it could effectively neutralize the alkaline CRW (down to a pH value of 6.3) and improve the properties of BOFS for further utilization, because the free-CaO and Ca(OH)2 in fresh BOFS could be totally eliminated after carbonation.8,36 The treatment cost for waste stabilization is expected to decrease because the RPB process does not need to introduce additional chemicals or steam; it only needs to utilize waste-CO2 as a reaction agent. It was thus concluded that the developed RPB process for carbonation of BOFS and CRW should be a feasible technology for CO2 capture and waste treatment for further discharge and/or utilization.

In this case, energy consumption of the proposed lab-scale RPB process was evaluated based on the assumption of the treatment of 1 ton CO2, where the grinding process, stirring process, pumps, and rotation of the packed bed were taken into account. The power consumption (W) for the BOFS grinding (crushing) process can be calculated by Bond’s equation as shown in eq 20,47 which has been widely used in the literature:20,48,49 ⎛ 10 W = Wi ⎜⎜ − ⎝ DP80

10 DF80

⎞ ⎟⎟ ⎠

(20)

where the DF80 (μm) and DP80 (μm) are the 80% passing size of feed and product BOFS, respectively, and Wi (kWh/ton) is the work index of ground material. In this study, the work index of BOFS was estimated to be 60.8 kWh/ton based on the available data provided by China Steel Corp. On the other hand, the power consumption for the stirring process, pumps, and rotation of the packed bed was estimated by multiplying the operating voltage to the operating amplitude of the existing equipment. Because the energy consumption of processes would increase with the operating time increases, the overall MACC of BOFS should be achieved at the “maximum point”, as shown in Figure 4a, by considering both the carbonation rate and energy consumption. The results indicate that the operating time for reaching the MACC was 8.5 min using CRW in an RPB (1 → 2 and then we can obtain the point A). In that case, the required amount of BOFS for capturing 1 ton of CO2 by the developed RPB process was estimated to be 5.13 ton (3 → 5 and then we can get the point B), under which the MACC was approximate 0.195 ton CO2 per ton BOFS. It suggests that the carbonation of BOFS coupled with CRW with a particle size less than 125 μm exhibits a relatively higher performance to achieve the lower energy consumption with higher CO2 capture capacity. Accordingly, the energy consumption of the proposed RPB process using direct carbonation of BOFS with various particle sizes under different reaction times can be determined from Figure 4b. As mentioned before, the capture capacity of BOFS at its maximum reaction rate (operated for 8.5 min using CRW) was estimated to be 0.195 ton CO2 per ton BOFS (from point 6 to point 8). In this case, the total energy consumption of the lab-scale RPB process is estimated to be 707 kWh/t-CO2 (from point 9 to point 10), which is counted for US $57/tCO2, as listed in Table 2. In addition, the rotation of the packed bed is found to be the most energy-intensive process (i.e., 43% of total). Meanwhile, the fraction of energy consumption for

■

ASSOCIATED CONTENT

S Supporting Information *

Additional information as noted in the text. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Phone: +886-2-23769236; fax: +886-2-27361661; e-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Great appreciation goes to China Steel Corp. and National Science Council of Taiwan (R.O.C.) for financial support under grant numbers RE100641 and NSC 102-3113-P-007-007, respectively, and to Dr. Ming-Wen Chu in the Center for Condensed Matter Sciences, National Taiwan University, for technical support on XRD analysis.

■

REFERENCES

(1) IPCC. IPCC Fourth Assessment Report (AR4); Intergovernmental Panel on Climate Change: Cambridge, 2007. (2) CSLF. In Focus: What is Carbon Utilization?; Carbon Sequestration Leadership Forum (CSLF): Washington, DC,2011. (3) IPCC. IPCC Special Report on Carbon dioxide Capture and Storage; Intergovernmental Panel on Climate Change: Cambridge, 2005. (4) Chang, E. E.; Chen, C. H.; Chen, Y. H.; Pan, S. Y.; Chiang, P. C. Performance evaluation for carbonation of steel-making slags in a slurry reactor. J. Hazard. Mater. 2011, 186 (1), 558−64. 13684



Article

(5) Tan, C.; Chen, J. Absorption of carbon dioxide with piperazine and its mixtures in a rotating packed bed. Sep. Purif. Technol. 2006, 49 (2), 174−180. (6) Huijgen, W. J. J.; Comans, R. N. J. Carbonation of steel slag for CO2 sequestration: Leaching of products and reaction mechanisms. Environ. Sci. Technol. 2006, 40 (8), 2790−2796. (7) Lackner, K. S. A guide to CO2 sequestration. Science 2003, 300 (5626), 1677−1678. (8) Pan, S. Y.; Chang, E. E.; Chiang, P. C. CO2 capture by accelerated carbonation of alkaline wastes: A review on its principles and applications. Aerosol Air Qual. Res. 2012, 12, 770−91. (9) Chen, Y. H.; Huang, Y. H.; Lin, R. H.; Shang, N. C. A continuous-flow biodiesel production process using a rotating packed bed. Bioresour. Technol. 2010, 101 (2), 668−73. (10) Yu, C. H.; Cheng, H. H.; Tan, C. S. CO2 capture by alkanolamine solutions containing diethylenetriamine and piperazine in a rotating packed bed. Int. J. Greenhouse Gas Control 2012, 9, 136− 147. (11) Chang, E. E.; Pan, S. Y.; Chen, Y. H.; Tan, C. S.; Chiang, P. C. Accelerated carbonation of steelmaking slags in a high-gravity rotating packed bed. J. Hazard. Mater. 2012, 227−228, 97−106. (12) Pan, S. Y.; Chiang, P. C.; Chen, Y. H.; Tan, C. S.; Chang, E. E. Ex Situ CO2 capture by carbonation of steelmaking slag coupled with metalworking wastewater in a rotating packed bed. Environ. Sci. Technol. 2013, 47 (7), 3308−15. (13) Chang, E. E.; Pan, S. Y.; Chen, Y. H.; Chu, H. W.; Wang, C. F.; Chiang, P. C. CO2 sequestration by carbonation of steelmaking slags in an autoclave reactor. J. Hazard. Mater. 2011, 195, 107−114. (14) Chang, E. E.; Wang, Y. C.; Pan, S. Y.; Chen, Y. H.; Chiang, P. C. CO2 Capture by using blended hydraulic slag cement via a slurry reactor. Aerosol Air Qual. Res. 2012, 12, 1433−1443. (15) Chang, E. E.; Chiu, A. C.; Pan, S. Y.; Chen, Y. H.; Tan, C. S.; Chiang, P. C. Carbonation of basic oxygen furnace slag with metalworking wastewater in a slurry reactor. Int. J. Greenhouse Gas Control 2013, 12, 382−389. (16) Gunning, P. J.; Hills, C. D.; Carey, P. J. Accelerated carbonation treatment of industrial wastes. Waste Manag. 2010, 30 (6), 1081−90. (17) Park, A.; Fan, L. mineral sequestration: physically activated dissolution of serpentine and pH swing process. Chem. Eng. Sci. 2004, 59 (22−23), 5241−5247. (18) Santos, R. M.; François, D.; Mertens, G.; Elsen, J.; Van Gerven, T. Ultrasound-intensified mineral carbonation. Appl. Therm. Eng. 2013, 57 (1−2), 154−163. (19) Teir, S.; Eloneva, S.; Fogelholm, C.; Zevenhoven, R. Fixation of carbon dioxide by producing hydromagnesite from serpentinite. Appl. Energy 2009, 86 (2), 214−218. (20) Eloneva, S.; Said, A.; Fogelholm, C.-J.; Zevenhoven, R. Preliminary assessment of a method utilizing carbon dioxide and steelmaking slags to produce precipitated calcium carbonate. Appl. Energy 2012, 90 (1), 329−334. (21) Santos, R. M.; Ling, D.; Sarvaramini, A.; Guo, M.; Elsen, J.; Larachi, F.; Beaudoin, G.; Blanpain, B.; Van Gerven, T. Stabilization of basic oxygen furnace slag by hot-stage carbonation treatment. Chem. Eng. J. 2012, 203, 239−250. (22) Sorlini, S.; Sanzeni, A.; Rondi, L. Reuse of steel slag in bituminous paving mixtures. J. Hazard. Mater. 2012, 209−210, 84−91. (23) Diener, S.; Andreas, L.; Herrmann, I.; Ecke, H.; Lagerkvist, A. Accelerated carbonation of steel slags in a landfill cover construction. Waste Manag. 2010, 30 (1), 132−9. (24) Huntzinger, D. N.; Eatmon, T. D. A life-cycle assessment of Portland cement manufacturing: comparing the traditional process with alternative technologies. J. Cleaner Prod. 2009, 17 (7), 668−675. (25) Xiao, L. S.; Wang, R.; Chiang, P. C.; Pan, S. Y.; Guo, Q. H.; Chang, E. E., Comparative life cycle assessment (LCA) of accelerated carbonation processes using steelmaking slag for CO2 fixation.. Aerosol Air Qual. Res. (2013). (26) Liu, H.; Kuo, C. Quantitative multiphase determination using the Rietveld method with high accuracy. Mater. Lett. 1996, 26, 171− 175.

(27) Zhou, Y.; Song, W.; Zeng, X.; Xie, C. Quantitative X-ray Rietveld analysis of metallic aluminum content in nano-aluminum powders. Mater. Lett. 2012, 67 (1), 177−179. (28) Bodénan, F.; Deniard, P. Characterization of flue gas cleaning residues from European solid waste incinerators: Assessment of various Ca-based sorbent processes. Chemosphere 2003, 51 (5), 335− 347. (29) Dermatas, D.; Dadachov, M. S. Rietveld quantification of montmorillonites in lead-contaminated soils. Appl. Clay Sci. 2003, 23 (1−4), 245−255. (30) Fernández-Jimenez, A.; Delatorre, A.; Palomo, A.; Lopezolmo, G.; Alonso, M.; Aranda, M. Quantitative determination of phases in the alkali activation of fly ash. Part I. Potential ash reactivity. Fuel 2006, 85 (5−6), 625−634. (31) Mahieux, P. Y.; Aubert, J. E.; Cyr, M.; Coutand, M.; Husson, B. Quantitative mineralogical composition of complex mineral wastescontribution of the Rietveld method. Waste Manag. 2010, 30 (3), 378−88. (32) Santos, R. M.; Van Bouwel, J.; Vandevelde, E.; Mertens, G.; Elsen, J.; Van Gerven, T. Accelerated mineral carbonation of stainless steel slags for CO2 storage and waste valorization: Effect of process parameters on geochemical properties. Int. J. Greenhouse Gas Control 2013, 17, 32−45. (33) Uibu, M.; Kuusik, R.; Andreas, L.; Kirsimäe, K. The CO2binding by Ca-Mg-silicates in direct aqueous carbonation of oil shale ash and steel slag. Energy Procedia 2011, 4, 925−932. (34) Liu, Q.; Maroto-Valer, M. M. Studies of pH buffer systems to promote carbonate formation for CO2 sequestration in brines. Fuel Process. Technol. 2012, 98, 6−13. (35) Kuusik, R.; Uibu, M.; Kirsimäe, K. Characterization of oil shale ashes formed at industrial-scale CFBC boilers. Oil Shale 2005, 22, 407−420. (36) Chang, E. E.; Chen, T. L.; Pan, S. Y.; Chen, Y. H.; Chiang, P. C. Kinetic modeling on CO2 capture using basic oxygen furnace slag coupled with cold-rolling wastewater in a rotating packed bed. J. Hazard. Mater. 2013, 260, 937−946. (37) Larson, A. C.; Von Dreele, R. B. Generalized crystal structure analysis system. LAUR 1986, 86−748. (38) Rietveld, H. M. A profile refinement method for nuclear and magnetic structures. J. Appl. Crystallogr. 1969, 2, 65−71. (39) Toby, B. H. R factors in Rietveld analysis: How good is good enough? Powder Diffr. 2012, 21 (01), 67−70. (40) Doucet, F. J. Effective CO2-specific sequestration capacity of steel slags and variability in their leaching behaviour in view of industrial mineral carbonation. Miner. Eng. 2010, 23 (3), 262−269. (41) De Windt, L.; Chaurand, P.; Rose, J. Kinetics of steel slag leaching: Batch tests and modeling. Waste Manag. 2011, 31 (2), 225− 35. (42) Newman, E. S. Heats of formation of xonotlite, hillebrandite, and foshagite. J. Resea. Natio. Burea. Stand. 1956, 57, 27−30. (43) Saito, T.; Sakai, E.; Morioka, M.; Otsuki, N. Carbonation of rCa2SiO4 and the mechanism of vaterite formation. J. Adv. Concrete Technol. 2010, 8 (3), 273−280. (44) Brady, P. V. Physics and chemistry of mineral surfces; CRC Press LLC: Boca Raton, FL, 1996; p 368. (45) Hangx, S. J. T. Behaviour of the CO2-H2O system and preliminary mineralisation model and experiments; Shell International Exploration and Production (leader CATO WP 4.1), December, 2005; p 43. (46) Krevor, S. C. M.; Lackner, K. S. Enhancing serpentine dissolution kinetics for mineral carbon dioxide sequestration. Int. J. Greenhouse Gas Control 2011, 5 (4), 1073−1080. (47) Bond, F. C. Crushing and grinding calculations. Part 1. Br. Chem. Eng. 1961, 6, 378−385. (48) Huijgen, W. J. J.; Ruijg, G. J.; Comans, R. N. J.; Witkamp, G. J. Energy consumption and net CO2 sequestration of aqueous mineral carbonation. Ind. Eng. Chem. Res. 2006, 45 (26), 9184−9194. (49) Kodama, S.; Nishimoto, T.; Yamamoto, N.; Yogo, K.; Yamada, K. Development of a new pH-swing CO2 mineralization process with a recyclable reaction solution. Energy 2008, 33 (5), 776−784. 13685


Systematic Approach to Determination of Maximum Achievable

Recommend Documents