CO2 Sequestration by Direct GasâSolid Carbonation of Air Pollution

Energy & Fuels 2006, 20, 1933-1940

1933

CO2 Sequestration by Direct Gas-Solid Carbonation of Air Pollution Control (APC) Residues Renato Baciocchi,*,† Alessandra Polettini,‡ Raffaella Pomi,‡ Valentina Prigiobbe,§ Viktoria Nikulshina Von Zedwitz,§ and Aldo Steinfeld§,| Department of CiVil Engineering, UniVersity of Rome “Tor Vergata”, Via del Politecnico 1, 00133 Rome, Italy, Department of Hydraulics, Transportation and Roads, UniVersity of Rome “La Sapienza”, Via Eudossiana 18, 00184 Rome, Italy, Department of Mechanical and Process Engineering, ETH Zurich, 8092 Zurich, Switzerland, and Solar Technology Laboratory, Paul Scherrer Institute, 5232 Villigen, Switzerland ReceiVed March 30, 2006. ReVised Manuscript ReceiVed May 15, 2006

Direct gas-solid carbonation of alkaline residues from air pollution control (APC) systems was investigated with the aim of evaluating its contribution as a CO2 storage option as well as its effect on the properties of the residues, namely, the leaching behavior, which may affect the strategies for their disposal or reuse. APC residues from a medical solid-waste incinerator located in the vicinity of Rome, Italy, were selected for performing carbonation experiments with pure CO2 at reaction temperatures in the 200-500 °C range. The extent of calcium conversion to the carbonate form was found to be negligible below 300 °C, whereas the maximum conversion of 57% was already measured at 400 °C. This implies a storage capacity of 0.12 kg of CO2/kg of dry solid. Considering that the APC residues production in the EU25 equals 1260 kiloton/year, this translates in a CO2 storage potential for the European market of 0.15 megatons of CO2/year. On the basis of the ENV 12457-2 leaching test procedure performed after accelerated carbonation, the concentrations of Cd, Cu, and Cr in the leachate were below the limits imposed by the Italian regulation for disposal in nonhazardous waste landfills, whereas only the lead concentration still exceeded the corresponding limit value. These results indicate that accelerated carbonation of alkaline residues could be suitable to address the issue of CO2 storage, especially for niche applications such as at steel plants and solid-waste incinerators, where both the residues and the CO2 are present.

Introduction The rapid increase in the CO2 concentration in the atmosphere from the pre-industrial values of 280 ppm to the current values of 370 ppm has forced the international community toward adopting a series of actions, e.g., the Kyoto protocol, aimed at reducing anthropogenic emissions of greenhouse gases. Furthermore, because fossil fuels are projected to be a dominant energy resource in the 21st century,1 technologies for sequestering emissions from fossil fuel combustion in a safe and definitive manner are being developed and implemented. Among the different options, mineral carbonation has been proposed as a possible way for CO2 sequestration. In this process, CO2 reacts with alkaline elements, namely, Ca and Mg, forming the corresponding thermodynamically stable carbonates.2 The large worldwide availability of minerals rich in alkaline earth silicates provides a source of alkaline materials that exceeds that required * To whom correspondence should be addressed. Telephone: +39-0672597022. Fax: +39-06-72597021. E-mail: [email protected]. † University of Rome “Tor Vergata”. ‡ University of Rome “La Sapienza”. § ETH Zurich. | Paul Scherrer Institute. (1) Fauth D. J.; Baltrus J. P.; Soong Y.; Knoer J. P.; Howard B. H.; Graham W. J.; Maroto-Valer M. M.; Andresen J. M. Carbon Storage and Sequestration as Mineral Carbonates in EnVironmental Challenges and Greenhouse Gas Control for Fossil Fuel Utilization in the 21st Century; Springer, Berlin, Germany, 2002. (2) Huijgen, W. J. J.; Comans, R. N. J. Carbon Dioxide Sequestration by Mineral Carbonation: Literature ReView; Report ECN SF ECN-C--03016: Petten, The Netherlands, 2003.

for sequestering the total CO2 emitted by the combustion of the available fossil fuel reservoirs.2 However, capture of CO2 by mineral carbonation requires operation at high temperatures (180 °C) and pressures (15 MPa).3 Although several other options have been proposed,2 it is worth noting that the current stage of process development is still at the lab scale. An alternative source of materials is represented by alkaline wastes, which are available in relatively large amounts and are generally rich in Ca or Mg. Johnson4 proposed the use of several waste materials, including pulverized fuel ash (PFA) produced by coal-fired power stations, ground granulated blast furnace, and stainless steel slags (SSG) from the steel manufacturer industry, bottom and fly municipal solid-waste incineration ashes (MSWI-b and MSWI-f), as well as deinking ash resulting from the waste produced during the recycling of paper. Carbonation of these waste materials, previously mixed with water at different water/solid (w/s) ratios, was carried out in a 100% CO2 atmosphere at a pressure of 3 atm. The best performance was obtained with SSG, with a maximum 20% weight gain upon carbonation, achieved with a w/s ratio of 0.125. A lower but still notable weight gain of 13% was also observed for MSWIf. More recently, Fernańdez Bertos et al.5 investigated more specifically the carbonation behavior of bottom ash and air pollution control (APC) residues from different municipal solid(3) Wolf, G. H.; Chizmeshya, A. V. G.; Diefenbacher, J.; McKelvy, M. J. EnViron. Sci. Technol. 2004, 38, 932-936. (4) Johnson, D. C. SCI Lect. Pap. Ser. 2000, 108, 1-10. (5) Fernańdez Bertos, M.; Li, X.; Simons, S. J. R.; Hills, C. D.; Carey, P. J. Green Chem. 2004, 6, 428-436.

10.1021/ef060135b CCC: $33.50 © 2006 American Chemical Society Published on Web 07/07/2006

1934 Energy & Fuels, Vol. 20, No. 5, 2006

waste incinerators located in the U.K. The results from the experimental tests, performed at operating conditions very close to those reported by Johnson,4 resulted in a 7 and 3% weight gain for APC residues and bottom ash, respectively. A quite different approach was followed by Huijgen et al.,6,7 who performed the carbonation of steel slag slurries at a 10:1 w/s ratio. A 70% Ca conversion was obtained operating at 20 bar and 200 °C in a 100% CO2 atmosphere, on a solid fraction sieved below 106 µm. Stolaroff et al.8 assessed the feasibility of extracting free calcium from steel slag and concrete waste by dissolution experiments in aqueous solutions. According to these authors, the solution containing free calcium could then be used in a carbonation process for capturing CO2 directly from air. Well before its exploitation as a process for CO2 storage, carbonation was investigated for its effects on the mechanical properties of cementitious materials9-13 as well as on the leaching behavior of alkaline wastes14-20 and soils.17,21 Such studies assessed the effect of carbonation on the reduction of the mobility of inorganic trace contaminants from raw and solidified wastes as a result of pH changes and the formation of stable carbonate forms or chemical interactions (including sorption and coprecipitation) with the newly formed carbonate species. An alternative to the direct wet route is represented by direct gas-solid carbonation of alkaline residues. Such a carbonation route has been traditionally applied for the carbonation of raw materials, such as calcium and magnesium oxides.22,23 Namely, Bhatia and Perlmutter22 investigated the kinetics of CaO carbonation by modified TGA experiments performed in a CO2 stream at different operating temperatures, obtaining 70% calcium conversion at 500 °C but still a reasonably fast process rate and high carbonation efficiency when operating at 400 °C. More recently, Abanades et al.24 proposed the use of CaO in a fluidized bed to capture CO2 directly from combustion gases; (6) Huijgen W. J. J.; Witkamp, G.-J.; Comans, R. N. J. Mineral CO2 Sequestration in Alkaline Solid Residues; Report ECN SF ECN-RX--04079: Petten, The Netherlands, 2004. (7) Huijgen W. J. J.; Witkamp, G.-J.; Comans, R. N. J. EnViron. Sci. Technol. 2005, 39, 9676-9682. (8) Stolaroff, J. K.; Lowry, G. V.; Keith, D. W. Energy ConVers. Manage. 2005, 46, 687-699. (9) Johnstone, J. R.; Glasser, F. P. Proc. 9th Int. Congr. Chem. Cem. 1992, 5, 370-376. (10) Lange, L. C.; Hills, C. D.; Poole, A. B. EnViron. Sci. Technol. 1996, 30, 25-30. (11) Short, N. R.; Purnell, P.; Page, C. L. J. Mater. Sci. 2001, 36, 3541. (12) Shtepenko, O. L.; Hills, C. D.; Coleman, N. J.; Brough, A. EnViron. Sci. Technol. 2005, 39, 345-354. (13) van Gerven, T.; van Baelen, D.; Dutre´, V.; Vandecasteele, C. Cem. Concr. Res. 2004, 34, 149-156. (14) Reddy, K. J.; Gloss, S. P.; Wang, L. Water Res. 1994, 28, 13771382. (15) Meima, J. A.; van der Weijden, R. D.; Eighmy, T. T.; Comans, R. N. J. Appl. Geochem. 2002, 17, 1503-1513. (16) Ecke, H.; Menad, N.; Lagerkvist, A. J. EnViron. Eng. 2003, 129, 435-440. (17) Fernańdez Bertos, M.; Simons, S. J. R.; Hills, C. D.; Carey, P. J. J. Hazard. Mater. 2004, B112, 193-205. (18) Polettini, A.; Pomi, R. J. Hazard. Mater. 2004, B113, 209-215. (19) Rendek, E.; Ducom, G.; Germain, P. J. Hazard. Mater. 2006, B128, 73-79. (20) van Gerven, T.; van Keer, E.; Arickx, S.; Jaspers, M.; Wauters, G.; Vandecasteele, C. Waste Manage. 2005, 25, 291-300. (21) Perera, A. S. R.; Al-Tabbaa, A. Proceedings of the 4th BGA GeoenVironmental Engineering Conference: Integrated Management of Groundwater and Contaminated Land: Stratford-upon-Avon, U.K., June 28-30, 2004; pp 560-567. (22) Bhatia, S. K.; Perlmutter, D. D.; AIChE J. 1983, 29, 79-86. (23) Butt, D. P.; Lackner, K. S.; Wendt, C. H.; Conzone, S. D.; Kung, H.; Lu, Y.-C.; Bremser, J. J. Am. Ceram. Soc. 1996, 79, 1982-1898.

Baciocchi et al.

up to 70% carbonation conversion was obtained for the fresh material, but a reduction in the conversion yield was observed after several carbonation-calcination cycles, required to recover the raw material from the process. The results obtained from direct gas-solid carbonation of natural alkaline oxides provide the basis for applying this route to the carbonation of alkaline residues as well. Such materials are mainly composed of calcium and magnesium oxides or hydroxides. The advantage of using alkaline residues rather than natural alkaline oxides is that, unlike the latter, they are not a raw material but a waste stream. Therefore, the process could be envisioned as a sequestration rather than a capture process. Besides, the carbonation process could at the same time improve the leaching properties of the waste material, potentially allowing for either utilization of the carbonated material for civilengineering application or mitigation of impacts at the final disposal site. Relevant work dealing with direct gas/solid carbonation of alkaline residues was performed by Jia et al.,25 who performed pressurized thermogravimetric analysis (TGA) experiments of hydrated and nonhydrated fluidized bed combustion (FBC) ash, achieving CaO conversion efficiencies up to 60% when operating at temperatures above 400 °C and in a 100% CO2 atmosphere. Nevertheless, in this work, no further independent evidence of the degree of carbonation was provided nor was any information given on the effect of the carbonation reaction on metal-leaching behavior. The objective of this work is to investigate the efficacy and efficiency of direct gas-solid carbonation of APC residues collected from a medical waste incinerator located near Rome, Italy. Carbonation experiments performed under different operating conditions in a modified muffle furnace are presented and discussed. The carbonation degree is then evaluated on the basis of different independent measurements (weight gain, calcination tests, and TGA tests), made on the material before and after carbonation. The leaching behavior of the tested material prior and after the carbonation process is also investigated to elucidate its influence on the mobility of trace metals. Experimental Section APC ash was sampled at a medical solid-waste incinerator located in the vicinity of Rome from the baghouse section, following a contact reactor for acid gases and organic micropollutants abatement using Ca(OH)2 and activated carbon. When the fresh APC ash was received at the laboratory, it was homogenized using a quartering procedure, oven-dried at 60 °C to constant weight, and finally transferred into sealed containers, where it was kept until the time of testing to prevent contact with atmospheric CO2. Characterization of APC ash involved the determination of element and major anion content, mineralogical composition, acid neutralization capacity (ANC), as well as leaching behavior. The elemental composition of APC ash was determined using an alkaline digestion procedure with Li2B4O7 in platinum melting pots at 1050 °C, followed by dissolution of the molten material in a solution of 10% (w/w) HNO3/nanopure water and measurement of element concentrations by means of an atomic absorption spectrometer equipped with air-acetylene flame. The concentration of chloride and acid-soluble sulfate was determined using the standard Italian UNI 8520 test procedures, which involve dissolution with hot deionized water and 3% HCl, respectively. The mineralogical composition of the material was evaluated by X-ray diffraction (XRD) analysis using Cu KR radiation, as well as by TGA and differential thermogravimetry (DTG). The acid neutralization behavior of the material was evaluated through the ANC test as (24) Abanades, J. C.; Anthony, E. J.; Lu, D. Y.; Salvador, C.; Alvarez, D. AIChE J. 2004, 50, 1614-1622. (25) Jia L.; Anthony E. J. Fuel 2000, 79, 1109-1114.

CO2 Sequestration by Direct Gas-Solid Carbonation

Energy & Fuels, Vol. 20, No. 5, 2006 1935

Table 1. Chemical Composition of APC Ash (Dry Weight Basis) element Al Ca Cd Cr Cu Fe Mg Mn Ni Pb Zn

concentration (mg/kg) 3195.0 276 294 116.1 55.1 825.7 937.2 6187.0 29.0 42.0 707.9 5436.3

anion Cl-

SO4-2

concentration (g/kg) 307.5 20.9

proposed by the Canadian Wastewater Technology Centre,26 which involves contacting 11 subsamples of the material with HNO3 solutions with increasing acid concentration. The leaching behavior was evaluated from the results of the ANC test reported as heavymetal-concentration-versus-pH curves, as well as by the ENV 12457-2 leaching test. The accelerated carbonation treatment was applied as direct gas/ solid carbonation at different temperatures and contact times to evaluate the ash susceptibility to the process and the optimal operating conditions. The carbonation experiments were conducted in a muffle furnace that was modified to allow for feeding a constant 100% CO2 flow and for distributing uniformly the gas flow within the muffle chamber. Treatment temperature values of 200, 300, 400, and 500 °C were selected for the carbonation experiments, and the treated samples were referred to as S2, S3, S4 and S5, respectively. Residence time equal to 6 hours. Because of the semibatch configuration of the carbonation reactor and the elevated treatment temperature, no adjustment could be made to the moisture content of the material, and this, although recognized as an important parameter for carbonation,5,16,27 was not included among the process variables. Before the onset of the experiments, the carbonation chamber was flushed with 100% CO2 for 10 min to purge air out from inside. Control experiments were also performed at the same operating conditions but using an inert gas (N2) flux. To evaluate the extent of carbonation, the treated material was characterized for weight gain upon carbonation and for final carbonate content, which was measured by means of calcimetry. This involved contacting the material with a standard volume of a HCl solution in a closed chamber and subsequently measuring the volume of CO2 evolved from the reaction between carbonate phases and HCl. Additional information on the production of major mineral phases as a result of carbonation was derived from TGA/DTG analyses as well as from a TGA technique coupled with an on-line gas chromatograph for the determination of CO2 in the gases evolved from the sample. The influence of accelerated carbonation on the leaching behavior of the material was evaluated through the previously described ANC and ENV 12457-2 leaching tests.

Results and Discussion Characterization of APC Residues. Table 1 reports the chemical composition of APC ash, while Figure 1 depicts the results from thermal analysis in terms of TGA and DTG curves as well as CO2 evolution as a function of the temperature. Chemical composition data reveal that the major constituents of APC ash are by far calcium and chlorine, accounting for more than 58% of the total mass of the residue. Relatively high concentrations were also measured for volatile heavy metals, such as Zn, Pb, Cu, and Cd, which are present in APC residues (26) Stegemann, J. A.; Coˆte´, P. L. InVestigation of Test Methods for Solidified Waste CharacterizationsA CooperatiVe Program, Appendix B: Test Methods for Solidified Waste EValuation; Environment Canada, unpublished Manuscript Series Document TS-15, 1991. (27) Fernańdez Bertos, M.; Scuzzarella, A.; Simons, S. J. R.; Hills, C. D.; Carey, P. J. Proceedings of AIChE 2004 Annual Meeting; November 7-12, Austin, TX, 2004.

Figure 1. Results of TGA/DTG analysis for sample S0. (a) Sample weight (s), derivative weight (gray line) (W0 ) 5.33 mg; heating rate ) 10 °C/min), and (b) CO2 concentration (- - -) of gases leaving the TGA furnace (W0 ) 38.83 mg; heating rate ) 20 °C/min; gas flow rate ) 238 N mL min-1).

as a result of volatile metal evaporation under the elevated temperature conditions of the incinerator. Additional information on the mineralogical characteristics of APC ash was derived from TGA/DTG analyses, which indicated the presence of two peaks on the DTG curve at temperatures of about 500 and 650 °C, respectively, with relatively large related weight losses. The first DTG peak was associated to Ca(OH)2, with a weight loss of 5.0%, corresponding to a Ca(OH)2 content of 20.5%. The second peak, on the basis of the evidence from CO2 evolution measurements, was associated to CaCO3, the presence of which in APC ash is presumably a result of a certain degree of carbonation occurring before the time of testing, possibly in the APC system. The total volume of CO2 measured for the second peak was 0.94 mL, which corresponds to a CaCO3 content of 10.8%, in very close agreement with the amount calculated from the weight loss associated to the second peak, which was equal to 10.6%. Tentative partitioning of Ca among the different chemical species was calculated on the basis of the data discussed above, XRD measurements, as well as ENV 12457 leachate composition. The calculations were done to determine the amount of Ca available for the carbonation process [i.e, in the form of Ca(OH)2/CaO] and the consequent Ca conversion yield. XRD analysis of untreated APC ash (not reported here) revealed the presence of crystalline phases including chloride (halite, NaCl, and sylvite, KCl) and sulfate minerals (anhydrite, CaSO4) in addition to Ca(OH)2. Such species are formed during flue gas treatment as a consequence of acid gas neutralization reactions and are concentrated in the solid residue from the APC section. It was assumed that chloride species were completely solubilized during 24 h distilled water leaching. Under the hypothesis that Na and K in the eluate from the ENV 12457 leaching test were


Baciocchi et al.

Table 2. Calculated Ca Speciation in APC Ash species

amount (% dry weight)

Ca(OH)2 CaCl2

20.5 32.0

CaCO3 CaSO4

10.8 3.0

CaO

10.9

measurement/assumption TGA/DTG analysis indirect calculation based on Clcontent and leaching results TGA/DTG analysis direct calculation based on SO4-2 content indirect calculation based on residual Ca

in the form of the corresponding chloride salts, the amount of NaCl and KCl in APC ash was calculated to be 16.8 and 0.2%, respectively. The assumption was then made that the residual chloride content (accounting for 20.5% of APC ash) was in the form of CaCl2, while the entire sulfate content was present as CaSO4. The amount of Ca remaining according to the above calculations (which was equal to 7.8%) was finally assumed to be in the form of CaO. Table 2 summarizes the speciation of Ca determined, along with the hypotheses made for the calculations. The amount of chemical forms available for carbonation determined from Table 2 is thus 31.4% as Ca(OH)2 plus CaO or 18.9% as Ca. The presence of any Mg mineral potentially liable to carbonation was not considered in the present study because of the much lower Mg content (0.6%) in APC ash as compared to Ca content (27.6%). The ANC behavior of APC ash is shown in Figure 4. For the untreated material, the ANC curve indicates the presence of a very large plateau at a pH value of about 12.5 units, typical of solution pH control by Ca(OH)2, which was responsible for an acid neutralization capacity of about 8.8 milliequiv/g of dry material. This corresponds to a Ca(OH)2 content of 17.6% (expressed as Ca), which is reasonably close to the value of 18.9% determined using the approximate method described previously. From the ANC results, it is also evident that the whole acid neutralization capacity of APC ash was dictated by Ca(OH)2; after solubilization of this phase, no additional plateau was visible on the ANC curve reported in Figure 4, so that pH dropped steeply from 12.5 down to less than 1 unit upon the addition of about 3 milliequiv of H+/g. Carbonation of APC Residues. The results of the carbonation experiments are reported in Figure 5a in terms of weight gain observed at the different operating conditions tested. The results reported refer to samples S3, S4, and S5, whereas the results for sample S2 are not shown because the weight gain was not detectable. The weight gain was found to be dependent upon the operating temperature, increasing from 3.1% for sample S3 to slightly less than 10% for samples S4 and S5. Conversely, exposing the APC residues to a nitrogen flow resulted in a reduction in the sample weight that was slightly influenced by the operating temperature, increasing from 1.2% at 300 °C to about 2.5% at 400 and 500 °C. Such a weight reduction was probably due to the dehydration process, involving the loss of both sample humidity and bound water (e.g., in calcium hydroxide). Because it is reasonable to assume that the dehydration process still takes place during accelerated carbonation and that the extent of water loss is the same as under N2 and CO2 atmosphere, the actual weight gain upon carbonation was calculated using the principle of superimposition of the effects. As shown in Figure 5a, the resulting actual weight change ranged from 4.4% at 300 °C to about 12% at 400 and 500 °C. The CO2 content of the APC residues before and after treatment in CO2 atmosphere, shown in Figure 5b, clearly confirmed the extent of the carbonation reactions. The CO2 content did not notably increase after carbonation at 200 °C, in


agreement with the undetectable weight gain observed at this temperature. In contrast, the CO2 content increased up to 18% at higher temperatures. Figure 6 compares the weight gain calculated from the difference in CO2 content after and before the carbonation treatment (derived from Figure 5b) with the actual weight gain (reported in Figure 5a). Although the quantitative agreement is not perfect, it is worth pointing out that the two independent measurements suggest that the weight gain of the APC residue may be entirely attributed to the increase in CO2 content. Besides, the trend line is respected, with an increase in weight gain for samples S3 and S4 and almost no effect or a slightly negative one for sample S5. The TGA/DTG curves relevant to sample S3, reported in Figure 2, show that the first DTG peak at 500 °C, associated to Ca(OH)2, was smaller than that observed for untreated APC ash (sample S0, see Figure 1). Furthermore, a larger second peak at 650 °C, associated to CaCO3, was observed, with an estimated calcium carbonate content of 25.7 wt % (11.3% CO2), which is in agreement with the data obtained from the calcimetry test (Figure 5b) and discussed above. The data obtained from the GC analysis, also shown in Figure 2, confirmed that the CO2 amount in the S3 sample was higher than for the untreated fly ash (sample S0), although the quantitative agreement between TGA/DTG and GC data was in this case rather poor, with the latter measurements providing a CO2 amount of 18.1 wt %. The TGA curve for sample S4, depicted in Figure 3, showed a different behavior than those curves for samples S0 and S3, displaying a sudden 20% weight loss at temperatures between 20 and 120 °C, ascribed to the loss of moisture that is probably more strongly retained in sample S4 because of a modification of the material porosity. In addition, the DTG curve of Figure 3 shows no peak at 500 °C, thus suggesting complete conversion



Figure 4. ANC curves for untreated and carbonated APC residues.


of Ca(OH)2, while the CaCO3 peak corresponded to a higher associated weight loss when compared to the other samples. The weight decrease at 650 °C was equal to 14.7%, assuming total water content of 20% on the basis of TGA observations, implying that the CO2 content of dry sample S4 was 18.4%, corresponding to 41.8% as CaCO3. Such figures are in close agreement with those obtained from the integration of the CO2 evolution diagram, also reported in Figure 3, indicating a 20.1% CO2 content, i.e., a 45.6% CaCO3 content. Finally, it is worth pointing out that even in this case the results obtained from TGA/DTG measurements provided similar values for the CO2 content if compared to those determined by the calcimetry experiments, reported in Figure 6. On the basis of these results and relying on Ca speciation reported in Table 1, a 100 g APC residue sample contains roughly 0.47 mol of Ca. Carbonation of APC ash at 400 °C results in a 12% weight increase, which, as confirmed by different independent measurements, may be entirely attributed to CO2 uptake as CaCO3. This corresponds to 12 g, i.e., 0.27 mol, of CO2. Therefore, the Ca conversion yield is equal to 57%. A similar yield was obtained at 500 °C, and a smaller yield was obtained at 300 °C. As Bhatia and Perlmutter suggested,22 the occurrence of incomplete conversion in the gas-solid reaction may be attributed to the closure of small pores, which are characterized by a higher specific surface area, so that only the larger slowly reacting pores are left. The maximum conversion yield attainable depends upon the cumulative small pores volume, i.e., on the solid properties. The limit conversion yield estimated by Bhatia and Perlmutter for CaO was found to be 71%. In a recent work, Abanades and Alvarez28 found an experimental maximum conversion yield for CaO

powder of roughly 60-65%, depending upon the carbonation conditions. Both figures are not far from that found for the APC residues investigated in this work; this suggests that the operating conditions adopted in the present work probably allowed for the maximum conversion yield to be attained for the investigated materials when applying carbonation as a gassolid reaction process. The results commented above and derived from weight change and calcimetric measurements as well as thermal analysis were confirmed by the acid neutralization behavior of the accelerated carbonated material, reported in Figure 4. It was found that the ANC of sample S2 was very similar to sample S0, as indicated by the presence of a plateau of comparable width at a pH value of 12.5; considering that, as observed earlier, such a pH plateau is the result of pH control by Ca(OH)2, this is an additional indication of the very poor carbonation degree of sample S2. In the case of treatment at 300 °C, a large initial pH plateau was still present, although with a lower width (6.4 milliequiv of H+/g, as opposed to 8.8 milliequiv of H+/g for sample S0). This confirms that accelerated carbonation at 200 and 300 °C was not capable of producing any macroscopic change in the mineralogical composition of APC ash. Conversely, the ANC behavior was dramatically modified by the treatment at 400 °C, while for sample S4, the plateau at pH 12.5 completely disappeared, indicating consumption of Ca(OH)2 as a consequence of the carbonation treatment; another large plateau at pH values in the range of 8.0-4.5 units (typical of the carbonate/bicarbonate buffer system) was visible on the ANC curve. The amount of acid required to overcome such a plateau was calculated to be 9.6 milliequiv of H+/g, corresponding to a carbonate content of 28.8% as CO3-2 or 48.0% as CaCO3. Considering the poor resolution of the titration curve as derived from the ANC test, this value is in reasonable agreement with that (43.5%) determined from TGA analysis. Carbonation is widely reported in the literature to be capable of affecting the release of trace metals from different types of combustion ashes, likely because of the combination of different mechanisms, such as pH changes, formation of metal carbonate salts, reduction in permeability as a consequence of calcite precipitation in the pore spaces, as well as interaction (including sorption, surface precipitation, and complexation) with newly formed solid phases. In particular, an immobilization effect of carbonation was observed by several investigators14-16,19,20,29 for such metals as Cd, Cu, Pb, Mn, and Zn, while Cr, Mo, and Sb were variously reported to be either mobilized or immobilized as a result of carbonation. However, the influence of (28) Abanades, J. C.; Alvarez D. Energy Fuels 2003, 17, 308-315.


Baciocchi et al.

Figure 7. Results from the ENV 12457 leaching test for untreated and carbonated samples.

Figure 5. Effect of carbonation. (a) Weight change under CO2 flux (light gray bars), weight change under N2 flux (gray bars), and net weight change (black bars). (b) CO2 content calculated from calcimetry tests.

Figure 6. Net weight gain for samples S3, S4, and S5 as calculated from calcimetry tests (gray bars) and batch carbonation experiments (black bars).

carbonation on the leaching characteristics of MSWI residues has not been explained thus far. In the present study, the effect of carbonation on metal leaching was investigated both at the natural pH of the material and as a function of solution pH. Heavy-metal concentrations in the eluate from the ENV 12457 leaching test are depicted in Figure 7, which shows that in the case of Cu, Pb, and Zn accelerated carbonation at 400 °C (29) Bone, B. D.; Knox, K.; Picken, A.; Robinson, H. D. Proceedings of Sardinia 2003, Ninth International Waste Management and Landfill Symposium; October 6-10, S. Margherita di Pula, Cagliari, Italy, 2003; session A10 (on CD ROM).

resulted in strongly reduced leachate concentration values. In particular, for sample S0, Zn exceeded the limit concentration of 5 mg/L prescribed for disposal in nonhazardous waste landfills. After carbonation at 400 °C, the Zn concentration was reduced to 1.1 mg/L, thus meeting the regulatory requirements. In the case of Pb, the leachate concentration for sample S0 was 60.2 mg/L, thus even exceeding the limit value of 5 mg/L prescribed for disposal in hazardous waste landfills. The strong reduction in Pb release observed for sample S4 was such that the eluate concentration (5.4 mg/L) complied with regulatory requirements for hazardous waste landfills but still exceeded the threshold value of 1 mg/L imposed for nonhazardous waste landfills. A different behavior was observed for Cr, with an increased leachate concentration of 0.41 mg/L for sample S4 as opposed to 0.12 mg/L for sample S0. However, for both untreated and carbonated APC ash, Cr release during the ENV 12457 test was below the corresponding limit value of 1 mg/L for disposal in nonhazardous waste landfills. A better insight into the leaching behavior of untreated and carbonated APC ash can be provided when complementing the ENV 12457 leaching data with the overall leaching curve as a function of pH as obtained from the ANC test (see Figure 8). A first large difference in the solubility curve was observed for Ca in the case of fully carbonated APC ash (sample S4): in this case, the strongly reduced Ca leachate concentrations in the whole pH range investigated suggest solubility control by calcite, which is much less soluble than portlandite. Among the trace metals, in the pH range of 10-12.5, the concentration of Cd was found to be lower than the analytical detection limit of 20 µg/L (which was used to plot the leaching curves) for all of the investigated samples, and for this reason, it was not possible to infer about the extent to which Cd leaching at the natural pH of the material is affected by carbonation. Nevertheless, for pH values below 8 units, all of the leaching curves for Cd were very similar, indicating no changes in solubility-controlling minerals as a result of accelerated carbonation. Similarly, in the case of Cu, the shape of the leaching curves was not modified by carbonation. Although some investigators hypothesized the possibility of precipitation of malachite in leachates from carbonated MSWI APC ash,16 other authors, when characterizing the precipitate from MSWI bottom ash leachates, stated that no evidence of Cu carbonate mineral formation could be gained.15 As a consequence, the reduced Cu release from sample S4 reported in Figure 7 was most likely a result of the



Figure 8. Metal leaching as a function of pH (ANC test) for untreated and carbonated samples.

pH decrease toward the range of minimum solubility. In the case of Pb, the leaching curve for sample S4 displayed lower leachate concentrations in the pH range of 6-12.5 if compared to the other materials. The results from geochemical modeling using Visual MINTEQ, a Windows version of the MINTEQA2 geochemical speciation code,30 indicated possible solubility control by PbCO3 in the pH range of 6-12.5; nevertheless, at lower pH values, Pb(OH)2 was still found to control the total Pb concentration in solution. The effect of accelerated carbonation on Cr leaching could not be entirely resolved from the leaching test results, because of some dispersion of the measured concentration values, particularly for sample S4. However, in agreement with the findings from other authors,31 it appears that carbonation lead to increased Cr leaching for different pH domains. The formation of any Cr carbonate species is excluded at both the trivalent and hexavalent oxidation states of this metal; it is tempting to hypothesize that the carbonation process may have affected the availability of Al(0) in the material. Metallic Al, which is commonly recognized in MSWI ash32,33 as small droplets condensed from evaporated metal, which can potentially originate from aluminum foils used in food packaging, is (30) Allison, J. D.; Brown, D. S.; Novo-Gradac, K. J. MINTEQA2/ PRODEFA2. A Geochemical Assessment Model for EnVironmental Systems. Version 3.0 User’s Manual, U.S. Environmental Protection Agency: Athens, GA, 1991. (31) Astrup, T.; Rosenblad, C.; Trapp, S.; Christensen, T. H. EnViron. Sci. Technol. 2005, 39, 3321-3329.

reported to be capable of controlling Cr leaching by reducing Cr(VI) released from the material into the less soluble Cr(III).31 It may be argued that calcite precipitated as a result of carbonation forms a protective rim that coats the particles of metallic aluminum, preventing it from reacting with Cr in solution, which is thus kept in the more soluble oxidized forms. Summary and Conclusions The carbonation of APC residues was shown to be a viable process for the storage of CO2, yielding almost 60% conversion of calcium available as calcium oxide or hydroxide and an uptake of around 120 g of CO2/kg of residue. Considering that the estimated production of APC residues in the 25 countries of the European Union (EU 25) from municipal, hazardous, and hospital solid-waste incinerators is approximately 1260 kilotons/ year,34 the CO2 storage capacity would amount to about 0.15 megatons of CO2/year. This figure will be compared to the CO2 emission reduction expected from the application of the Kyoto (32) International Ash Working Group (IAWG: Chandler, A. J.; Eighmy, T. T.; Hartleń, J.; Hjelmar, O.; Kosson, D.; Sawell, S. E.; van der Sloot, H. A.; Vehlow, J.), Municipal Solid Waste Incinerator Residues, Studies in Environmental Sciences 67, Elsevier Science: Amsterdam, The Netherlands, 1997. (33) Sabbas, T.; Polettini, A.; Pomi, R.; Astrup, T.; Hjelmar, O.; Mostbauer, P.; Cappai, G.; Magel, G.; Salhofer, S.; Speiser, C.; HeussAssbichler, S.; Klein, R.; Lechner, P. Waste Manage. 2003, 23, 61-88.


protocol, about 8% of the 1990 CO2 emissions, i.e., about 350 megatons of CO2/year, or to the more realistic 19.7% reduction effort with respect to the actual 2010 emission baseline, i.e., about 800 megatons of CO2/year. Thus, the APC residues produced from the existing incineration plants would cover only 0.02-0.05% of the total CO2 European storage capacity required to comply with the Kyoto protocol objectives. Nevertheless, APC residues represent just a small fraction of alkaline residues available in the European Union, and they could be effective (34) European Commission. Study To Facilitate the Implementation of Certain Waste Related ProVisions of the Regulation on Persistent Organic Pollutants (POPs). Synthesis Report ENV.A.2/ETU/2004/0044, Brussels, Belgium, 2005.

Baciocchi et al.

as additional CO2 storage capacity. Furthermore, they could be suitable for niche applications, such as at steel plants and municipal solid-waste incinerators, where both the residues and the CO2 are present.2 Even though the ANC results did not show a major effect of the carbonation process on the leaching behavior of the APC residues, Cd, Cu, and Cr concentrations in the leachate of the ENV 12547 leaching test were below the limits imposed by the Italian regulation for disposal in nonhazardous waste landfills, whereas only the Pb concentration still exceeded the corresponding limit value. Thus, carbonation may also represent an effective way to address the issues related to the disposal of these and other alkaline residues. EF060135B

CO2 Sequestration by Direct GasâSolid Carbonation of Air Pollution

Recommend Documents

CO2 Sequestration by Direct GasâSolid Carbonation of Air Pollution