Controls on Metal Leaching from Secondary Pb Smelter Air-Pollution

Oct 7, 2008 - Waste Management (CWM), T. G. Masaryk Water Research. Institute, Podbabská 30, 160 62, Praha 6, Czech Republic. Received May 7, 2008...
0 downloads 0 Views 5MB Size
Environ. Sci. Technol. 2008, 42, 7878–7884

Controls on Metal Leaching from Secondary Pb Smelter Air-Pollution-Control Residues ˇ C H E T T L E R , * ,† O N D R ˇ E J Sˇ E B E K , ‡ VOJTE ´ Sˇ G R Y G A R , § TOMA ´ ,§ MARIANA KLEMENTOVA ˇ KA,§ AND PETR BEZDIC ´ †,| H A L K A S L A V ´I K O V A Institute of Geochemistry, Mineralogy and Mineral Resources, Charles University, Albertov 6, 128 43 Prague 2, Czech Republic, Laboratories of Geological Institutes, Charles University, Albertov 6, 128 43, Prague 2, Czech Republic, Institute of Inorganic Chemistry, Academy of Science of the ˇ zˇ , Czech Republic, and Centre for Czech Republic, 250 68 Re Waste Management (CWM), T. G. Masaryk Water Research Institute, Podbabska´ 30, 160 62, Praha 6, Czech Republic

Received May 7, 2008. Revised manuscript received August 4, 2008. Accepted August 18, 2008.

Air-pollution-control (APC) residues are among the most toxic waste materials from secondary Pb metallurgy. Two distinct APC residues collected from bag-type filters of one Czech secondary Pb smelter were subjected to leaching experiments to determine the mineralogical and geochemical controls on leaching of metallic contaminants (Cd, Cu, Pb, Zn). A kinetic 720-h leaching test in deionized water at a liquid-tosolid (L/S) ratio of 10 L kg-1 and a 48-h leaching test at various L/S ratios (0.5, 1, 5, 10, 50, 100, 500, and 1000 L kg-1) were performed and coupled with the mineralogical investigation of solid residues (TEM/EDS, XRD) and PHREEQC-2 speciationsolubility modeling. The rapid release of contaminants into the solution at all the L/S ratios showed the rapid dissolution of primary phases. The leaching at high L/S ratios, representing longterm predictions of leaching behavior, showed that primary alkaline sulfates and chlorides (Na3Pb2(SO4)3Cl and KCl · 2PbCl2) were dissolved and anglesite (PbSO4) was formed as their final and stable alteration product. Primary amorphous PbSO3 partly crystallized during leaching and oxidized to anglesite. These results are consistent with the mineralogical investigation of soils exposed for decades to Pb smelter emissions, where only anglesite was detected. The leaching experiments showed that washing residues at L/S > 50 accompanied by spontaneous anglesite precipitation can be an alternative for improved technological treatment of these residues. Although this process would require further treatment of contaminated effluent, the newly precipitated anglesite is more favorable than the primary APC residue phases for an efficient metallurgical recovery of Pb.

* Corresponding author phone: +420 221 951 493; fax: +420 221 951 496; e-mail:[email protected]. † Institute of Geochemistry, Mineralogy and Mineral Resources, Charles University. ‡ Laboratories of Geological Institutes, Charles University. § Academy of Science of the Czech Republic. | T. G. Masaryk Water Research Institute. 7878

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 21, 2008

Introduction Fly ash and air-pollution-control (APC) residues from high temperature industrial processes (waste incineration, coal burning, metal smelting) are environmentally hazardous materials due to (i) the elevated concentrations of volatile metals and metalloids, (ii) the soluble forms of these inorganic contaminants, and (iii) the small size of the particles (PM2.5) that can easily enter pulmonary systems (1-3). In particular, lead (Pb) smelters are important local sources of metallic pollution, as shown in numerous recent studies on aerosols and fly ashes (3-6). The particles emitted by smelter stacks were found to influence the quality of soils in the vicinity of large metallurgical complexes, as has been observed in the soil concentration profiles (7-9) and through isotopic tracing of the contamination (10, 11). Leaching tests help to evaluate possible environmental effects related to the release of contaminants from smelter emissions, fly ash, or air-pollution-control (APC) residues. Many recent papers, which were focused especially on APC residues from municipal solid waste incinerators (MSWI), showed the importance of an experimental leaching protocol when interpreting the results (2, 12, 13). The studies of contaminant leaching from APC residues from MSWI underlined the effect of the liquid-to-solid (L/S) ratio for the investigation of the long-term leaching behavior of waste materials, where a value of L/S can be related to time in a “disposal scenario” and corresponds to a cumulative ratio of liquid that can be in contact with waste over time (12, 13). Whereas low L/S ratios (2 or 10 L kg-1) generally used in the regulatory tests (e.g., European batch leaching test EN 12457 (15)) may show the short-term leachability in the horizon of hundreds of years, high L/S ratios can be used for long-term predictions of waste behavior up to hundreds of thousands of years in a specific “disposal scenario” (12, 14). This paper follows up the previous investigations of mineralogy of smelter APC residues (6) and their short-term leaching behavior using regulatory single extractions tests (16). The present study included kinetic batch leaching in a time frame of 720 h and leaching at various L/S ratios, performed on two contrasting APC residues. The leaching results were coupled with thermodynamic modeling and a thorough investigation of the newly formed phases to determine the key solubility-controlling mechanisms related to the release of contaminants from these materials, in particular when deposited in the environment (soils) or reused for further metal recovery in technical processes in the smelter.

Experimental Section APC Residues. Two APC residues corresponding to two technologically distinct samples were collected at the bottom of the bag-type filters in the secondary Pb smelter at Prˇ´ıbram, Czech Republic. The technology of flue gas cleaning is explained in detail in the Supporting Information. Residue A corresponded to a solid trapped by bag-type filters at 200 °C after previous cooling of the flue gas by pure water. Residue B corresponded to a solid trapped under the identical conditions after previous cooling of the flue gas by alkaline water. The chemical and mineralogical compositions of both residues are given in Table 1. The methodology of chemical and mineralogical analyses is given in the Supporting Information. Leaching Experiments. Kinetic Leaching. The batch leaching experiment was performed at a liquid-to-solid (L/ S) ratio of 10 L kg-1 in deionized water as a leaching medium 10.1021/es801246c CCC: $40.75

 2008 American Chemical Society

Published on Web 10/07/2008

TABLE 1. Selected Chemical and Mineralogical Parameters of Fresh APC Residues residue Aa

element

residue B a

Cu (mg kg-1) Cd (mg kg-1) Pb (mg kg-1) Zn (mg kg-1) Na (mg kg-1) K (mg kg-1) Cl (%) S (%) TICb (%)

Chemical Composition 149 ( 3 274 ( 8 9950 ( 640 4460 ( 240 539300 ( 20510 248300 ( 18200 4900 ( 90 5010 ( 210 26400 ( 550 204800 ( 2630 60370 ( 850 32770 ( 770 19.7 ( 0.4 19.9 ( 0.4 2.91 ( 0.044 6.45 ( 0.014 1 µm crystal size. The material contains ∼50% of XRD amorphous component. Comparison of the Rietveld and chemical analyses indicated that the amorphous or nanocrystalline components are potassium lead chloride and/or caracolite. In contrast, the crystalline fraction of residue B is composed of predominant caracolite with 30 nm crystal size, soluble salts (NaCl and Na2SO4) with minor amounts of potassium lead chloride with 70 nm crystal size, lead sulfite (PbSO3) with 50 nm crystal size, galena (PbS), and phosgenite (PbCl2 · PbCO3). The solid also contains ∼50% of amorphous component, probably amorphous PbSO3 as follows from the comparison between the Rietveld and chemical analysis and indirectly also from the formation of crystalline PbSO3 during the leaching. Kinetic Leaching Test. The time-dependent release of metals into the leachate, the saturation indices of major phases predicted by PHREEQC-2, and the amounts of the phases in the fresh and leached residues are reported in Figure 1. Residue A exhibits rapid release of metals, achieving a steady state within several hours. The most important contaminants released are Pb (up to 1713 mg L-1) and Cd (up to 889 mg L-1). Significantly lower concentrations were observed for Zn (up to 297 mg L-1) and Cu (up to 8.7 mg L-1). The pH of the solution also yields a steady-state value of ∼4.3 (Figure 1). Lead and Cd were mainly present in the leachates as chlorocomplexes, whereas Zn and Cu are mainly present as the free ionic forms (Table S1 in the Supporting Information). The PHREEQC-2 calculation of the saturation indices showed that the leachates of residue A are oversaturated or VOL. 42, NO. 21, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

7879

FIGURE 1. Time-dependent leaching of Pb, Zn, Cu, and Cd at L/S of 10 L kg-1 for residues A and B, respectively, saturation indices of key solubility-controlling phases (calculated by PHREEQC-2), and mineralogical composition of leached residues (in % in crystalline fraction, calculated by the Rietveld refinement from the XRD patterns). Chemical compositions of phases: anglesite (PbSO4), orthorhombic lead sulfite (PbSO3), laurionite (Pb(OH)Cl), cotunnite (PbCl2), caracolite (Na3Pb2(SO4)3Cl), potassium lead chloride (KCl · 2PbCl2), phosgenite (PbCl2 · PbCO3). close to saturation with respect to anglesite (PbSO4), cotunnite (PbCl2), and laurionite (Pb(OH)Cl) (Figure 1, Table S3 in the Supporting Information). These predictions are consistent with the mineralogy of the leached solids. The XRD study showed that primary KCl · 2PbCl2 and caracolite are dissolved, and their amounts in the crystalline fraction of the residue at the end of experiment decrease to 10 for residue B. The leaching concentrations recalculated to the metal contents in the initial residues permitted calculation of the % amount of leached metals (Figure 2) and better indicates the leaching efficiency at a given L/S ratio. Cadmium is strongly leached at all L/S ratios: 85-96% (residue A) and 55-66% of total Cd (residue B). Similar leaching curves were observed for Zn in the case of residue A (47-63% of total Zn leached) and residue B (18-60% of the total Zn content) with a pronounced increase in the Zn leachability at L/S > 5. Whereas, in residue A, Cu exhibited a peak in the leachability at L/S of 50 (59%), only up to 5% of total Cu was leached in residue B. Significantly different leaching of Pb was observed. For residue A, the Pb leachability increased from 0.3 to 57% of the total Pb content at L/S > 10. In contrast, only 10% of the total Pb was leached from sample B at the highest L/S ratio of 1000 (Figure 2). The pH value is the most significant parameter influencing the leachability of metals, partially responsible for the differences in the leaching behavior of both residues: whereas residue A yields an equilibrium pH between 4.2 and 5.2, the equilibrium pH of residue B is shifted to more neutral conditions (6.1-7.2). The lower pH of the residue A leachates may be explained by the presence of free acids in the residue,

as suggested by Ettler et al. (16), and by the significant buffering capacity of residue B caused by flue gas cooling by alkaline water prior to residue trapping in bag-type filters. The mineralogical study showed a significant variation in the phase composition as a function of L/S, with more complex phase assemblages at low L/S. In residue A, primary caracolite and KCl · 2PbCl2 and newly formed palmierite (K2Pb(SO4)2) disappeared at L/S > 5, whereas newly formed cotunnite was still present at L/S ) 10. Laurionite still occurred at low amounts but disappeared at L/S ) 1000. The most important phase is anglesite, which becomes predominant at L/S > 10 (Figure 2). These data are in agreement with the PHREEQC-2 predictions that the leachates at L/S ) 1000 are oversaturated only with respect to anglesite, whereas other phases tend to dissolve (Figure 2). Distinct behavior was observed for residue B, where phases like laurionite, palmierite, and phosgenite dissolve with increasing L/S. Caracolite also tends to be dissolved at higher L/S ratios (with one peak at L/S ) 10), yielding fractions 50 and production of washed residue composed mainly of anglesite, which can be easily treated by reducing fusion in a shaft furnace (together with Pb scrap). The effluent produced by this washing could be further cleaned, e.g. by the Ferrox process (binding of contaminants to newly formed hydrous ferric oxides) (24) or recycled for cooling of flue gas prior to residue trapping in a bag-house facility. Environmental and Health Implications. Our study reveals that APC residues from secondary Pb smelter are highly soluble materials rapidly releasing high concentrations of metallic contaminants into solution. The fresh and leached residue particles are very small, commonly