Introduction
The development of the social economy and improvement of living standards in China has resulted in increasing amounts of municipal solid waste (MSW). Incineration, due to its primary advantages of hygienic control, volume reduction, mass reduction, and energy recovery, has become an attractive MSW treatment method [
1]. In an incineration process, different solid residues, including bottom ash, fly ash, and air pollution control residues, are generated. Bottom ash, including grate siftings, is the main waste stream, accounting for approximately 80% of total solid residues [
2]. Nowadays, MSWI bottom ash is reused as a secondary construction material in China [
3]. However, it contains a high level of heavy metals, resulting in its leaching potential [
4-
6]. Leaching of heavy metals from the MSWI bottom ash is of concern in many countries, and may inhibit its potential beneficial reuse as a secondary construction material, such as in road coffering or brickmaking. Among the heavy metals, copper (Cu) is of particular concern due to its high potential leaching ability and high biotoxicity [
7]. Therefore, it is important to understand the underlying leaching behavior of Cu from MSWI bottom ash for a proper evaluation of the environmental impact of MSWI bottom ash reutilization or disposal.
Distinct from other heavy metals, Cu is known to have high affinity for organic ligands [
8,
9]. Studies related to Cu leaching from MSWI bottom ash suggested that Cu leaching was significantly enhanced by dissolved organic carbon (DOC) [
10-
12]. To describe the complexation of Cu with DOC, some geochemical speciation models were established, such as the NICA-Donnan model [
13], the Gaussian DOM model [
14], and the Stockholm humic model (SHM) [
15]. The role of different types of organic substances in Cu leaching behavior from MSWI bottom ash was investigated using these models. Hydrophilic organic substance, notably fulvic acid, was found to be the main contributor in enhancing Cu leaching behavior [
11]. Moreover, leached Cu was mostly bound to DOC in the leachate [
16]. Besides the relatively well-known enhancing effect of DOC, other geochemical characteristics of MSWI bottom ash might also be important. For example, the acid-neutralizing capacity (ANC) of MSWI bottom ash, which controls the leachate pH, might influence Cu leaching behavior [
17,
18]. Furthermore, the speciation of Cu in MSWI bottom ash, which reveals the specific chemical forms, binding state, and mobility of Cu in various environmental conditions, could affect the leaching of Cu. Unfortunately, to date, most studies are focused on the role of DOC with respect to its enhancing effect on the leaching behavior of Cu. Few studies consider the ANC of MSWI bottom ash and Cu speciation. However, the leaching of Cu from MSWI bottom ash can only be discerned through a multi-analysis of these factors together.
In this research, the DOC and ANC of MSWI bottom ash and Cu speciation were considered together to discern the leaching behavior of Cu from MSWI bottom ash. X-ray diffraction (XRD) was used to characterize surface mineral composition. Batch titration procedure was used to identify ANC and Cu leaching behavior as a function of pH. The sequential extraction procedure (SEP) originated by Tessier et al. [
19] was adopted to analyze the speciation of Cu in MSWI bottom ash. The geochemical model Visual MINTEQ was used to evaluate the equilibrium of leachates with relative minerals, and to determine the speciation of the aqueous Cu in the leachate. The comprehensive analysis provided an important insight into the long-term leaching behavior of Cu from MSWI bottom ash as well as its impact on the environment.
Materials and methods
Sampling and pretreatment
Fresh MSWI bottom ash sample was collected from the Green Energy MSWI Plant in Zhejiang Province. The plant consists of three parallel stoker incinerators with MSW treatment capacity of 650 t/d. The operating temperature of the incinerators is 850°C-1100°C, and the retention time of MSW in the incinerator is approximately 50 min. The MSWI bottom ash was water quenched and separated magnetically to recover iron and ferrous metals before sampling. The sampling period lasted for five days. Approximately 5 kg of fresh MSWI bottom ash sample was collected from the plant every day to obtain a 25 kg sample. It was mingled and air dried. Part of the sample was grounded into less than 154 µm with a grinder for composition analysis.
Bottom ash composition analysis
The content of the individual elements in MSWI bottom ash was analyzed after the sample was digested according to the method described by Yamasaki [
20]. First, 0.5 g air-dried sample was weighed into a Teflon beaker. Then, 2.5 mL HNO
3 and 2.5 mL HClO
4 were added, and the mixture was heated at 150°C for 2-3 h. After cooling, 2.5 mL HClO
4 and 5.0 mL HF were added, and the mixture heated at 150°C for 15 min. Then, 5.0 mL of HF was added until the residue became almost dry. The residue was dissolved using 5.0 mL HNO
3, and diluted to 100 mL with distilled deionized water. The element concentrations in the solution were determined by inductively coupled plasma-optical emission spectroscopy (ICP-OES) (Thermo Electron Corporation IRIS/AP, USA).
The mineralogical investigation of MSWI bottom ash sample was conducted by X-ray powder diffraction (Rigaku D/max-r B, Japan).
Physical and chemical properties, namely, moisture content, bulk density, loss on ignition at 600°C (LOI600°C), and pH, were also determined. The measurements followed the ASTM D2216 for water content, ASTM C29 for bulk density, and GB7876-87 for LOI600°C. The pH was determined after a 24 h equilibration period with a liquid-to-solid ratio of 5. The value of LOI600°C indirectly reflected the organic matter content in MSWI bottom ash.
Batch titration procedure
ANC and Cu leaching as a function of pH were determined by the batch titration procedure as suggested by Johnson et al. [
18]. Each 2.5 g of MSWI bottom ash sample was placed in 25 previously acid washed polyethylene bottles, and thoroughly rinsed with deionized water. Acidic solutions (250 mL) were produced from degassed deionized water and 1.0 mol·L
-1 of HNO
3, and were added to the samples ranging from 0-4.8 mmol H
+/g MSWI bottom ash. The solutions were continually flushed with N
2 to avoid contact with the atmosphere, and shaken for 24 h at 25°C. The solution pH values were determined immediately. The remaining solution was filtered through a 0.45 μm membrane filter. The resulting solutions were analyzed for cations and anions by ICP-OES and IC (Dionex ICS-2000, USA), respectively. The DOC was determined using a total organic carbon (TOC) analyzer (Shimadzu TOC-V CPH, Japan).
SEP
The chemical speciation of Cu in the MSWI bottom ash sample was determined using SEP as suggested by Tessier et al. [
19]. It was performed according the following procedure.
Exchangeable (F1): A 2.0 g portion of the MSWI bottom ash was added with 16 mL magnesium chloride solution (1 mol·L-1 MgCl2, pH 7.0), and shaken for 1 h at room temperature.
Bound to carbonate (F2): The residue from the exchangeable fraction was shaken with 16 mL 1 mol·L-1 CH3COONa (pH 5.0) for 5 h at room temperature.
Bound to Fe-Mn oxides (F3): A aliquot of 20 mL 0.04 mol·L-1 hydroxylamine hydrochloride (NH2OH·HCl) in 25% (v/v) CH3COOH was added to the residue from the carbonate fraction, and heated at 96±3°C for 5 h with occasional agitation.
Bound to organic matters (F4): A 6 mL aliquot of 0.02 M HNO3 and 10 mL 30% hydrogen peroxide (H2O2) (pH 2.0) were added to the residue from the Fe-Mn oxides fraction, and heated at 85±2°C for 2 h with occasional agitation. Another aliquot of 6 mL 30% H2O2 (pH 2.0) was added, and heated again at 85±2°C for 3 h with intermittent agitation. After cooling to room temperature, 10 mL 3.2 M ammonium acetate (CH3COONH4) in 20% HNO3 was added, and the solution was agitated continuously for 30 min.
Residual fraction (F5): The residual fraction was determined by digestion of the residue from the fraction bound to organic matters, as described in Section 2.2.
Visual MINTEQ model
Visual MINTEQ was used for geochemical modeling [
21]. The leachate composition in equilibrium with the selected minerals and hydroxides was calculated. The calculation enabled the presentation of the model predictions together with the analytical data in a graph of log concentration versus pH, which assisted in determining the degree of saturation of the selected minerals in the element concentrations in the solution. The distribution of aqueous Cu in the leachate was investigated with this model. The input files were composed of the concentrations of DOC and inorganic elements measured in the leachate, the pH, which was fixed at the measured value, and the minerals selected as the solid phase.
Results and discussion
Composition analysis
The physi-chemical properties and bulk chemical composition of the MSWI bottom ash sample are presented in Tables 1 and 2. The LOI600°C, indirectly reflecting the organic matter content in the sample, was 2.2%. Si was the major element in the sample, accounting for more than 20% of the sample. The alkali metals, including Na, K, Al, Ca, and Mg were abundant. Thus, high ANC was expected. The Cu content was high, which was 14 times higher than that of the soils in China (22.6 mg/kg in the soils of China). It greatly exceeded the set standard for the “environmental quality standard for soil” (GB15618-1995, Grade II for soil pH<6.5∶Cu≤50 mg/kg indicating a pollution warning threshold), and a high potential threat to surrounding ecosystem.
Figure 1 illustrates the X-ray diffractogram of the MSWI bottom ash sample. The principal minerals identified were quartz (SiO
2), calcite (CaCO
3), maghemite (Fe
2O
3), and bauxite (Al
2O
3). No significant amount of porlandlite [Ca(OH)
2] was identified, contrary to expectation. Similar results were obtained by Polettini et al. [
22]. The previous phenomenon may be ascribed to the fast reaction of atmosphere CO
2 with Ca(OH)
2 during the water quenching treatment.
The results suggest that the MSWI bottom ash sample was mainly characterized by the high content of SiO
2, alkali metals (especially Ca), and heavy metals, which were often observed for MSWI bottom ash [
18,
23,
24]. Furthermore, its high Cu content indicates its potential environmental risk.
ANC of MSWI bottom ash
Leaching of Cu from the MSWI bottom ash is strongly dependent on the pH. Given that ANC controls the leachate pH, Cu leaching is therefore related to the ANC of MSWI bottom ash [
12]. The pH titration curve and the ANC of the MSWI bottom ash, which could be drawn by the corresponding target pH to the acid addition, are exhibited in Fig. 2. ANC
7.5 was 1 mequiv/g, and the pH was 10.3 when no acid was added to the solution, which decreased gradually with the addition of the acid. When 1 mmol H
+/g MSWI bottom ash was added, the pH value reached 7.5, at which the hydroxides, soluble basic silicate hydrates, and carbonates were thought to have been consumed [
18]. The leachate pH decreased to 3.3 when 4.0 mmol H
+/g MSWI bottom ash was added, which was the lowest in this study.
MSWI bottom ash was mainly buffered by hydroxides, silicates, and carbonates [
18,
23,
24]. Cation and anion concentrations in the leachate increased with the addition of the acid. Figure 3 demonstrates the dissolution of various species of cations and anions as a function of acid addition. This suggests that the MSWI bottom ash leachate pH was not only buffered by the Ca minerals in the alkaline condition, but also by other alkali metals (Na, K, Mg, and Al), by SiO
2 as well as metal oxide in an acidic condition. These minerals were the main contributors for the ANC of the MSWI bottom ash, which was in good accordance with the bulk composition of the latter.
Cu leaching from MSWI bottom ash
Cu leaching as a function of pH and Visual MINTEQ prediction, assuming equilibrium with the selected minerals, are shown in Fig. 4. Cu leaching was significantly dependent on leachate pH. The leaching was relatively low when the leachate was in basic or neutral condition (pH>6), with a concentration of 0.054-0.071 mg/L in the leachates. However, Cu was remarkably released when the leachate pH further decreased to the acid range. It peaked when 4 mmol H+/g MSWI bottom ash was added, and the leachate pH reached 3.3. Afterward, no significant further release was observed.
To explore the controlling mechanism of Cu leaching, the measured Cu concentration was compared with the Visual MINTEQ predicted value. Since Cu leaching is significantly enhanced by DOC according to previous works [
16,
25-
27], DOC concentration was incorporated into the Visual MINTEQ with the SHM, NICA-Donnan model, and Gaussian DOM model, respectively. The results are shown in Figs. 4(a), (b), and (c). When the pH value was higher than 9.1, the measured Cu concentrations were saturated with respect to Cu(OH)
2, and tenorite according to the Gaussian DOM model [Fig. 4(c)], but were 2-3 orders of magnitude too low to be in equilibrium with the selected minerals for the SHM and the NICA-Donnan models [Figs. 4(a) and (b)]. However, when the pH was lower than 9.1, the three models exhibited the same result that Cu concentration was far from saturation with respect to all the selected minerals, namely, Cu(OH)
2, azurite [Cu
3(OH)
2(CO
3)
2], tenorite (CuO), and malachite [Cu
2(OH)
2CO
3]. In practice, the results of the SHM and the NICA-Donnan model were close, and the patterns of predicted leaching curves by the two models were in good accordance with the measured one. This indicated that the two models seemed to be more suitable for the prediction.
The undersaturation of the Cu concentration might be caused by either: (1) the Cu leaching being controlled by unknown, less-soluble minerals, or (2) all the leaching available faction being leached out. To determine the specific reason, further batch titrations with different liquid-to-solid ratios were performed. The liquid-to-solid ratios were 200, 500, and 1000 respectively, and 4.0 mmol H+/g MSWI bottom ash was added. The leachate pH were 3.2, 3.3, and 3.3 for liquid-to-solid ratio of 200, 500, and 1000 respectively, which were close to the pH of the liquid-to-solid ratio of 100 (pH 3.3). Thus, the influence of pH difference among the leachates with different liquid-to-solid ratios was negligible. The leaching results are shown in Fig. 5. The Cu concentration decreased proportionally with the increase in liquid-to-solid ratio, suggesting that Cu leaching was not controlled by unknown less-soluble minerals. As in that scenario, Cu concentration would be relatively steady to keep the equilibrium with the controlling minerals. This indicated that Cu leaching was limited by the amount of leaching available, which was determined by its speciation in the MSWI bottom ash.
Cu speciation in MSWI bottom ash and its relationship with leaching behavior
To explore the leaching ability of Cu, SEP was used to analyze the speciation of Cu in the MSWI bottom ash sample. The results are shown in Fig. 6. The amount of each fraction followed the sequence of organic matter-bound fraction>residue fraction>carbonate-bound fraction>Fe-Mn oxide-bound fraction>exchangeable fraction. Organic matter-bound fraction was the major component, accounting for 45.0%. The high percentage of Cu occupied in organic matter-bound fraction was also observed for other solid phase materials [
28], which may be ascribed to the tendency of Cu to form strong compound complexes with organic ligands, such as fluvic acid. For MSWI bottom ash, some organic matters in the input MSW were not destroyed during incineration (which could be reflected by the value of LOI
600°C in Table 1), and were bound with Cu to form the organic matter-bound Cu complex. Besides the organic matter-bound fraction, the residue fraction accounted for 27.6%, while the carbonate-bound fraction accounted for 23.7%. Moreover, the Fe-Mn oxide-bound fraction and exchangeable fraction accounted for 2.5% and 1.2% respectively, which were relatively low.
Generally, exchangeable, carbonate, and Fe-Mn oxide fractions are considered unstable with potential leaching abilities and bioavailability [
1]. However, for Cu, the organic matter-bound fraction could be referred to as unstable because Cu leaching can be significantly enhanced by DOC, as mentioned before. According to this assumption, the total unstable amount of Cu in the MSWI bottom ash was calculated to be 227.8 mg/kg.
When no acid was added, the pH of the leachate was 10.3 and 5.4 mg/kg MSWI bottom ash of Cu was released to the leachate (Fig. 4), a little higher than the exchangeable fraction of Cu (3.7 mg/kg MSWI bottom ash). This may be ascribed to the partial hydrolysis of the carbonate-bound fraction of Cu, which would occur if the pH was under 10.3, according to the Visual MINTEQ calculation. The slow hydrolysis of the carbonate-bound fraction continued until the leachate pH decreased to an acidic condition (pH<6). Afterward, most of the carbonate fraction was hydrolyzed, and Cu was significantly released. When the leachate pH further decreased, the Fe-Mn oxide-bound and organic matter fractions of Cu were also released. The leaching peaked and became steady when the leachate pH decreased to 3.3 with 4 mmol H+/g MSWI bottom ash addition, which was 204.6 mg/kg MSWI bottom ash. This value was close to the unstable amount calculated by the SEP results (227.8 mg/kg MSWI bottom ash), indicating that almost all leaching available Cu had been released. Thus, no further release was observed with the increasing addition of the acid. The result confirms the assumption that Cu leaching was limited by the speciation of Cu in the MSWI bottom ash.
Environmental impact assessment
The Cu leaching behavior from the MSWI bottom ash could be predicted from the results. The leaching of Cu was expected to be relatively low when the leachate was in alkaline or neutral condition, but when enough acid was imposed to the MSWI bottom ash, and the leachate was in an acidic condition (2.4 mmol H+/g MSWI bottom ash addition in this study), Cu would be greatly released. The total leaching quantity of Cu can be referred to as the sum of the exchangeable, carbonate-bound, Fe-Mn oxide-bound and organic matter-bound fractions in the MSWI bottom ash. Generally, decreasing the leachate pH to the acid condition merely by acid precipitation is a long process because of the high ANC of MSWI bottom ash. However, if the MSWI bottom ash was exposed in an acidic environment, the result would be different. For example, in South China, the pH of acid rain is usually under 5.0. If the MSWI bottom ash was reused for road coffering in flood-prone areas in South China, Cu would be greatly leached.
To evaluate the environmental impact of the long-term leaching of Cu, both the quantity and speciation of the aqueous Cu should be considered. In this study, Cu speciation in the leachate was investigated with Visual MINTEQ by using SHM and the NICA-Donnan model, as shown in Table 3. In a basic or neutral condition, Cu was mainly bound to organic matter, such as fluvic and humic acid, but in an acidic condition, Cu mainly exist as Cu2+ in the leachate. Metal-organic complexion reduces the Cu toxicity, and the environmental impact of a low leaching amount of Cu is relatively small in basic or neutral conditions. However, when enough acid is imposed to overcome ANC, and the leachate pH reaches the acidic range, the environmental impact would become profound and worrisome. Unfortunately, most MSWI plants in China are located in South China, and most MSWI bottom ash is generated and reused there, which is a severe acid rain area. Thus, the environmental impact of the reutilization of the MSWI bottom ash should be strictly monitored and assessed in South China.
Conclusions
Leaching from the MSWI bottom ash was low in basic and neutral conditions. However, it was significantly released when the MSWI bottom ash was in an acidic condition. Cu leaching was limited by the amount of leaching available, as confirmed by the SEP results. According to these results, the exchangeable, carbonate-bound, Fe-Mn oxide-bound, organic matter-bound, and residual fractions of Cu accounted for 1.2%, 23.7%, 2.5%, 45.0%, and 27.6% in the MSWI bottom ash, respectively. The amount of leaching available was in agreement with the practical leaching amount. Furthermore, the modeling results of the Visual MINTEQ indicated that the aqueous Cu was almost 100% bound with organic matter in basic and neutral conditions, but mainly exists as Cu2+ in an acidic condition. Considering the leaching pattern of Cu and its speciation in the leachate, it is believed that a profound environmental impact might happen if the MSWI bottom ash was reused in an acid rain-prone area, such as South China.
Higher Education Press and Springer-Verlag Berlin Heidelberg