Combining sieving and washing, a way to treat MSWI boiler fly ash

Waste Management xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Waste Management journal homepage: www.elsevier.com/locate/wasman

Combining sieving and washing, a way to treat MSWI boiler fly ash Aurore De Boom ⇑, Marc Degrez 4MAT Department, Brussels School of Engineering, Université libre de Bruxelles (ULB), Avenue F.D. Roosevelt 50, CP165/63, 1050 Brussels, Belgium

a r t i c l e

i n f o

Article history: Received 14 August 2014 Accepted 29 January 2015 Available online xxxx Keywords: Municipal Solid Waste Incineration (MSWI) Fly ash Washing Sieving Valorisation

a b s t r a c t Municipal Solid Waste Incineration (MSWI) fly ashes contain some compounds that could be extracted and valorised. A process based on wet sieving and washing steps has been developed aiming to reach this objective. Such unique combination in MSWI fly ash treatment led to a non-hazardous fraction from incineration fly ashes. More specifically, MSWI Boiler Fly Ash (BFA) was separately sampled and treated. The BFA finer particles (13 wt%) were found to be more contaminated in Pb and Zn than the coarser fractions. After three washing steps, the coarser fractions presented leaching concentrations acceptable to landfill for non-hazardous materials so that an eventual subsequent valorisation may be foreseen. At the contrary, too much Pb leached from the finest particles and this fraction should be further treated. Wet sieving and washing permit thus to reduce the leachability of MSWI BFA and to concentrate the Pb and Zn contamination in a small (in particle size and volume) fraction. Such combination would therefore constitute a straightforward and efficient basis to valorise coarse particles from MSWI fly ashes. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction In Europe, the 2008/98/EC directive (EC, 2008) dictates the management system to apply to waste, based on Lansink’s scale and known as the waste hierarchy. The hierarchy promotes prevention, reuse and material recovery, before energy recovery and elimination processes. Considering the waste hierarchy for waste treatment residues such as Municipal Solid Waste Incineration (MSWI) fly ashes and Air Pollution Control (APC) residues, material recovery could be a promising technique. Indeed, such ashes and residues contain compounds (heavy metal ones for example) that could be valorised if they are extracted and purified. Recent projects over landfill mining aiming to valorise energy and material from landfills (ELFM, 2010; Van Gerven et al., 2010) prove the current need of an intensive and effective waste management to avoid wasting resources. Whenever such a project exists, material recovery, if feasible, should preferably take place before landfilling, to avoid material weathering and excavation process. In Belgium, waste landfilling is extremely limited (Eurostat, 2012) but concerns the incineration residues. Before being landfilled as hazardous waste MSWI fly ashes and APC residues have to be treated essentially because of their salt and lead content (Van Gerven et al., 2005). Different treatments have been developed to decrease the leachability of the solids; the treatments

⇑ Corresponding author. Tel.: +32 26503982. E-mail address: [email protected] (A. De Boom).

are generally classified as separation, stabilisation/solidification and thermal techniques (Chandler et al., 1997; Quina et al., 2008a). Effectively, MSWI fly ashes and APC residues contain high chloride amounts, too high for landfilling or for any valorisation option. Chlorides are mostly present as NaCl and KCl in MSWI fly ashes (Le Forestier and Libourel, 1998; Mangialardi, 2003), while they are also found as CaCl2 in APC residues (Le Forestier and Libourel, 1998). Several studies recommend washing (only with water) the residues to remove chlorides before applying another treatment (Chandler et al., 1997; Nagib and Inoue, 2000; Mangialardi, 2003; Piantone et al., 2003; Hyks et al., 2009; Jiang et al., 2009; Liu et al., 2009; Wang et al., 2010). A single washing, followed by displacement washing of the filtration cake, removes from 30% up to 95% of the chlorides, depending on the liquid-to-solid (L/S) ratio (Wang et al., 2001; Jiang et al., 2009). Washing fly ashes produces hydrate phases that form protective layers around the fly ash grains and reduce the inhibition action on cement hydration observed in case of cement stabilisation (Mangialardi, 2003). Used before acid leaching, washing reduces the acid consumption (Nagib and Inoue, 2000; Zhang and Itoh, 2006). When fly ashes are washed before a thermal treatment, the amount of volatised heavy metal compounds decreases, because the chlorides have been removed and calcium-containing aluminosilicates have formed (Wang et al., 2001; Jiang et al., 2009). However, other leachable compounds than chlorides dissolve when fly ashes or APC residues are in contact with water. Table 1 presents the extracted amounts of several elements generally found in MSWI fly ashes and APC residues. The extracted amounts

http://dx.doi.org/10.1016/j.wasman.2015.01.040 0956-053X/Ó 2015 Elsevier Ltd. All rights reserved.

Please cite this article in press as: De Boom, A., Degrez, M. Combining sieving and washing, a way to treat MSWI boiler fly ash. Waste Management (2015), http://dx.doi.org/10.1016/j.wasman.2015.01.040

2

A. De Boom, M. Degrez / Waste Management xxx (2015) xxx–xxx

are firstly expressed as a weight ratio comparing the leached element weight to the total solid weight (the ratio may also represent the concentration of the soluble form of the element) and secondly, as the percentage of the leached element amount compared to its concentration. All washing experiments presented in Table 1 are batch tests, without pH control. L/S ratio and time vary from test to test, from 1 to 100 and from 15 min to 20 h, respectively. Although both parameters may influence the compound extraction, the comparison of the results given in Table 1 brings useful information. From this table, it can be noticed that the electrofilter residues are generally composed of less soluble compounds, compared to scrubber and bag filter ones. This could be due to the fact that electrofilter residues may be composed mainly by fly ashes, while scrubber and bag filter residues also contain reaction products from the APC system (CaCl2 in case of lime injection, NaCl if soda is used). Table 1 also shows that, while heavy metal compounds may leach during the washing step and achieve too high concentrations in water, the extraction yield from the solid is low and the majority of the element remains in the solid. In case of Pb, maximum 1.1% of the total Pb was extracted while the leached amount may achieve up to 160 mg/kg. Washing could then be considered as an extraction step for high soluble compounds such as chlorides. However, it implies a pollution transfer to the resulting solution, which has to be treated before discharge. Furthermore, using large amounts of water would raise treatment costs and environmental impacts, because of the necessary wastewater treatment. An optimal washing should thus consume a minimum of water and dissolve a maximum of chlorides. Mixing time may influence the compound dissolution, for kinetic reasons. The dissolution appears to be rather rapid for chloride salts. 5 min were sufficient to obtain the maximal and steady dissolution of K and Na (Wilewska-Bien et al., 2007). However, a too long mixing time may also reduce the dissolution or extraction rate; during leaching tests performed on APC residues, the maximum chloride extraction was obtained after 2 h mixing and was lower when the time exceeded 4 h (Chimenos et al., 2005). Some other elements, such as Cr, take several days to achieve an equilibrium concentration (Astrup et al., 2005). A mixing time of 2 h was found to be acceptable for washing (Chimenos et al., 2005; Jiang et al., 2009). To set up an efficient washing treatment, the water consumption should also be minimised. This can be made by reducing the amount of water used per mass of solid (corresponding to the L/S ratio) of each step and of the entire process. The L/S ratio of each step may not be too low to permit to handle the suspension as a

solution and not as sludge, while the L/S ratio of the entire process should be high enough to remove a maximum of soluble compounds. The L/S ratio influences the compound dissolution, due to their solubility but the influence is generally notable only for L/S ratio lower than 10 L/kg in case of high soluble compounds such as chlorides. Cl, K and Na dissolution remains constant for L/S ratios from 10 to 100 L/kg (Nagib and Inoue, 2000; Wang et al., 2001), while it decreases for L/S ratios lower than 10 L/kg (Wang et al., 2001; Jiang et al., 2009). The influence of the L/S ratio on the heavy metal dissolution depends on the metal: Cr, Pb and Zn dissolution increases for L/S ratio going from 2 to 100 L/kg while Cd and Cu dissolution remains constant (Wang et al., 2001; Jiang et al., 2009). A L/S ratio lower than 10 L/kg should thus permit to dissolve a major part of chloride compounds and minimise the heavy metal (Cr, Pb and Zn) dissolution. However, a L/S ratio of 10 L/kg still represents huge amounts of water to treat MSWI ashes. It is therefore advantageous to use a L/S ratio high enough per washing step and to assure a low L/S ratio for the whole process (global consumed water volume compared to the treated solid mass), by re-using the same solution for several washing steps in a counter-current or in a recirculating process. Chimenos et al. developed a counter-current batch washing process, with two washing steps and one rinse on the final filter (Chimenos et al., 2005). They found acceptable residues, according to the local regulation, after recirculating the solution for 4 cycles, using a L/S ratio of 3 L/kg for each step and a mixing time of 1 h (the real water consumption is not mentioned but the solution is counter-current recirculated and fresh water is only added to keep the same L/S ratio at each stage). Wilewska-Bien et al. investigated a washing treatment of fly ash at a pilot scale and found that the leachate resulting from the water washing step might be recirculated for chloride (NaCl, KCl) extraction; the input of fresh water represented only a L/S ratio of 0.5 L/kg (Wilewska-Bien et al., 2007). A recirculating washing process seems thus efficient to remove chlorides from MSWI fly ashes and APC residues and permits to reduce the water consumption, compared to washing with only fresh water. Besides the chloride salt removal, some other compounds, such as metal ones, could also be extracted from MSWI fly ashes and APC residues. Physical separation processes, such as magnetic, Eddy current or size-based separation, are usually applied to bottom ash (Chandler et al., 1997; Vandecasteele et al., 2007; Grosso et al., 2011) which globally presents coarser solid fractions than fly ashes and APC residues. Moreover, the processes are rarely developed for these fine residues; no eddy current application for MSWI fly ashes or APC residues has been found. Magnetic

Table 1 Extraction yield ranges of some elements during MSWI fly ashes and APC residues washing (Katsuura et al., 1996; Nagib and Inoue, 2000; Wang et al., 2001; Mangialardi, 2003; Chimenos et al., 2005; Jiang et al., 2009).

Al Ca Cd Cl Cr Cu Fe K Mg Na Ni P Pb S Si Zn

g/kg g/kg mg/kg g/kg mg/kg mg/kg g/kg g/kg g/kg g/kg mg/kg g/kg mg/kg g/kg g/kg mg/kg

Electrofilter

Scrubber

Bag filter

0.7–0.8 10–34 0.7–650 22–27 2.8–4 2.0–2.3 0–0.01 15–125 0.02–2 13–85 2–5 – 6.25–160 13–20 – 18–22

2–7 25–75 0.7–0.9 34–56 4.5–44 0.4–0.6 1–2 25–28 1–2 20–24 – 1–5 2.1–19 – 1 0.3–4.4

0.9–1.5 21–50 – 72–205 4.1–28 0.3–0.5 0.3 19–50 – 32–79 – – 1.6–5.1 1–2 1.7–3.4 0.2–0.3

Al Ca Cd Cl Cr Cu Fe K Mg Na Ni P Pb S Si Zn

% extracted

Electrofilter

Scrubber

Bag filter

1.3–1.5 3.8–8.1 3.0–3.3 31–40 1.0–1.4 0.16–0.22 0–0.06 33–47 0.12–0.20 24–45 2.2–2.3 – 0.12–0.16 30–31 – 0.12–0.28

4.8–17 21–63 0.7–0.9 59–97 1.4–13 0.04–0.06 5.3–11 63–70 5.9–12 47–56 – 3.8–19 0.1–1.1 – 0.89 0–0.10

2.4–4.0 13–31 – 37–77 1.8–12 0.10 0.8–0.9 24–63 – 27–68 – – 0.10–0.20 3.0–5.0 1.6–3.2 –

Please cite this article in press as: De Boom, A., Degrez, M. Combining sieving and washing, a way to treat MSWI boiler fly ash. Waste Management (2015), http://dx.doi.org/10.1016/j.wasman.2015.01.040

A. De Boom, M. Degrez / Waste Management xxx (2015) xxx–xxx

separation has been applied to fly ashes and APC residues but a maximum of only 12% was extracted from the residues while the extracted particles were contaminated by several non-magnetic compounds (De Boom et al., 2011). Concerning the size-based separation, heavy metal particle distribution generally shows a higher contamination of the particles smaller than 125 lm (Alba et al., 1997; Chandler et al., 1997; Wang et al., 2002; Hammy et al., 2005), while no size effect has also been reported (Hammy et al., 2005; Chiang et al., 2008), depending partly on the type of residue. Considering that size-based separation could concentrate some contaminants in a specific size range, the remaining fraction would therefore be decontaminated and could join a valorisation scheme, while the heavy metals could be concentrated enough to be efficiently extracted by hydrometallurgical processes. Moreover, heavy metal extraction from MSWI fly ashes and APC residues has been investigated via alkaline or acid leaching (Nagib and Inoue, 2000; Levasseur et al., 2005; Zhang and Itoh, 2006), as well as via chelating or complexing agents (Hong et al., 2000a,b; Janos et al., 2002; Youcai et al., 2002; Pedersen, 2002). However, some of the reagents are toxic and their use implies important precautions and further substantial wastewater treatment. Moreover, the extraction yield depends on the target metal compounds. An efficient treatment should thus comprise a washing step to remove chlorides, completed by a decontamination step aiming to remove or to stabilise heavy metals. In addition to chlorides and heavy metals, the question of high organic compound levels could also raise (Derie, 1996; Quina et al., 2008a) but the present study was limited to mineral contaminants. To improve MSWI fly ash and APC residue treatment, the particularities of each kind of residue should be exploited, by collecting them separately and applying a specific treatment to each solid (Chandler et al., 1997; De Boom and Degrez, 2007). Among the different sorts of MSWI fly ashes and APC residues, boiler and electrostatic precipitator (ESP) (if there is no reactive injection before this device) fly ashes appear to be not directly influenced by the APC system and may be considered as ‘‘pure fly ashes’’. Therefore, boiler and ESP fly ashes constitute residues that can be the most easily generalised. However, a deNOX system may likely be placed on top of the furnace, being therefore before the boiler. Different systems exist and some imply ammonia injection, which could adsorb on fly ash; well-performing deNOX systems should only generate gaseous N2 and H2O, which should not influence the fly ash characteristics. The present paper constitutes the first part of a large MSWI fly ash valorisation study, from the treatment development until the valorisation of the treated residues. It presents the development of an entire treatment of MSWI fly ashes in order to valorise them. The composition of 14 fractions obtained by screening has been analysed. Based on the analyses, the number of fractions has been reduced and further washing has been applied on the different fractions, whose leaching behaviour has been investigated. The second part of the study will consider the incorporation of the treated residues into cement-based materials (De Boom et al., in preparation).

2. Materials and methods 2.1. Material Boiler Fly Ash (BFA) has been sampled at a Belgian MSWI plant (capacity: 13 t/h), directly at the boiler. After the boiler, the APC system is composed of a scrubber (lime milk injection), an electrostatic precipitator, a second lime milk addition, an activated carbon injection and finally a bag filter. BFA may be considered as pure fly

3

ash, as all the chemical injections related to the APC system occur after the boiler. A daily sample (5–10 kg) has been sampled during 9 days. Afterwards, equivalent volumes of each daily sample have been mixed together and the final mix constitutes the working sample. 2.2. Analyses Sample composition was analysed by X-ray fluorescence (XRF) (Bruker RS 3000, Be source, Rh-anti-cathode, OVO 55 analyser crystal), using a standard-less method (semi-quantitative analysis) and the Spectra Plus software. Main crystalline phases were identified by a powder X-ray diffractometer (XRD) (Bruker D5000, Cu source – Ka = 1.5406 Å, graphite monochromator, 1.2 s/step, step size: 0.02°, data base: ICDD PDF-2 2004). Ba, Ca, Cd, Cr, Cu, K, Mo, Na, Ni, Sb, Se, Pb and Zn concentrations in solution were analysed by Inductively Coupled Plasma Optical Emission Spectroscopy (ICP-OES, Varian VISTA-MPX CCD Simultaneous). Ba, Cd, Cr, Cu, Mo, Pb, Sb, Se and Zn were identified as potential pollutants, while Ca represented the low soluble matrix and K and Na, supposed to be present mainly as chlorides, the high soluble compounds. Chlorides were estimated by chloride titrators (Quatab, Hach, Environmental Test Systems). Titrators are graduated test strips and work as one-use chromatography test. The chloride concentration is given by the height of a white peak on the strip, thanks to a conversion table. The concentration is given by a graduation; the results correspond thus to a chloride range and not to an accurate concentration. Titrators may detect chloride concentration above 30 ppm. The water content is defined as the ratio of the mass of water to the mass of wet matter. The mass of water is determined by drying the wet matter at 105 °C during at least 3 h. 2.3. BFA treatment The BFA treatment developed in the present study comprised a sieving step followed by washing ones. The sieving consisted in sorting BFA in several fractions of different size ranges by using superposed sieves placed on a sieve shaker. Since the BFA presented small particles (De Boom and Degrez, 2012), wet sieving was applied to increase the separation efficiency (Fuerstenau and Han, 2003). Additionally, from a safety point of view, BFA wet sieving would advantageously limit the inhalation risk. However, a wet sieving implies higher water consumption and the dissolution of the soluble fraction. To overcome both drawbacks, the wet sieving step was carried out with a ‘‘recirculated solution’’: the solution recovered from a wet sieving step was used for the next one (see Fig. 1). For the first sieving step, a synthetic recirculated solution was made on the basis of the major compounds found in the BFA solid sample and leachate, namely chlorides (NaCl, KCl); some calcium (CaSO4
2H2O et Ca(OH)2) salts were also added because the pH is reputed to be controlled by calcium compounds (Chimenos et al., 2005; Hyks et al., 2009). The equilibrium reactions occurring in the synthetic solution have not been further studied. Sieving was applied on BFA to separate more contaminated fractions (richer in heavy metals, especially Pb compounds) from less contaminated ones. The ideal sieving should guarantee at the same time a minimum volume of the most contaminated fraction and a sufficient decontamination of the remaining fraction. Therefore, the best mesh sieve sizes were determined on the basis of the elementary composition and mass distribution of 14 BFA fractions (see Section 2.3.1.).

Please cite this article in press as: De Boom, A., Degrez, M. Combining sieving and washing, a way to treat MSWI boiler fly ash. Waste Management (2015), http://dx.doi.org/10.1016/j.wasman.2015.01.040

4

A. De Boom, M. Degrez / Waste Management xxx (2015) xxx–xxx

The separation was repeated three times, by recirculating the outflow: the filtrate obtained at the end of the first separation was mixed with a second BFA sample and the filtrate obtained at the end of the second separation was mixed with a third BFA sample. The sample weights were adapted to the available volume of solution.

Fig. 1. Boiler Fly Ash (BFA) treatment flow-sheet (X and Y: lowest and uppermost sieving sizes, respectively).

Then, the separation was restricted to 3 fractions obtained by using 2 sizes chosen among the previous ones (see Section 2.3.2.). After the wet sieving, the washing steps (see Section 2.3.3.) aimed to remove the recirculated solution and the high soluble compounds, particularly chlorides (NaCl, KCl) whose concentration is strictly limited for landfilling (EC, 2003) as well as in building applications. Each mixing step was performed on an orbital shaker (Stuart S 150) at ambient temperature (20–25 °C) and lasted 2 h. Fig. 1 represents the treatment scheme. If the treatment were carried out at an industrial scale, the washing steps themselves would be also processed in a counter-current way to reduce the global water consumption of the process. However, in this work, the washing steps were performed with demineralised water, for practical and investigation reasons. 2.3.1. Sieving size determination A BFA sample was mixed with recirculated solution (L/S = 6 L/ kg). Afterwards, the suspension was poured on a 38 lm-sieve, placed on a collector with outlet that allows recovering the solution passing through the sieves (the outflow). The sieve and the collector were fixed on a sieve shaker (AS 200 control). The outflow was recovered and filtered to obtain the lowest fraction (<38 lm). Both fractions (bigger and lower than 38 lm) were dried at 105 °C. The bigger fraction was further sieved by dry process on 13 sieves (38, 53, 75, 106, 150, 212, 300, 425, 600, 850 lm and 1.18, 1.70 and 2.36 mm) placed on the sieve shaker. Each of the 14 fractions was recovered and analysed by XRF. The test was repeated three times. 2.3.2. Wet sieving in recirculated solution Based on particle mass and composition distribution over the 14 size-fractions obtained in Section 2.3.1., two sieving sizes were chosen: the first one was obviously 38 lm (see Section 3.1.), to separate the most contaminated fraction; the second one appeared quite less evident and was set at 150 lm to check if a second sieve was efficient or not. The sieving was achieved as explained in Section 2.3.1., but the suspension was here poured on two sieves (the uppermost had a sieve range of 150 lm, the lowest of 38 lm), placed on a collector with outlet. The mixing vessel was rinsed with extra recirculated solution and the rinsing solution was poured on the humid sizefractions, so that the final L/S achieved 7 L/kg. The upper size-fractions (38–150 lm and > 150 lm) were recovered on the corresponding sieves; the outflow, containing the particles smaller than 38 lm, was filtrated (Whatmann n°1 filter). The resulting cake constituted the finest fraction and was recovered.

2.3.3. Size-fraction washing Each of the size-fractions obtained by wet sieving was mixed with a volume of demineralised water equivalent to twice the dry solid mass, by assuming that wet solids contained 25% of water, which was checked during the experiments. After mixing, solutions were filtered on Millipore 0.45 lm membrane for all fractions but the fractions bigger than 38 lm were firstly filtered on a Whatmann n°1 filter. Wet solids were recovered and washed three times following the same procedure. After each washing, solutions were sampled and checked for some elements (see Section 2.3). The solid was also sampled to measure the water content. 3. Results 3.1. BFA characterisation BFA elementary composition analysed by XRF is reported in Table 2. Ca is the major element present in BFA, although no lime is injected at the boiler. BFA also contain large quantities of S, which, along with the Ca concentration, is related to the anhydrite abundance in the ashes. A lot of minor elements are found, especially heavy metals. Normally, non-volatile species are found in BFA because of their mechanical entrainment by the flue gases while volatile species evaporate at the furnace and condensate on fine particles (Chandler et al., 1997; Belevi and Moench, 2000). XRD analyses reveal that BFA mainly contain anhydrite (CaSO4) and quartz (SiO2), while the other crystalline phases are halite (NaCl), sylvite (KCl), calcite (CaCO3) and gehlenite (Ca2Al2SiO7) (see Fig. 2). The minerals are usually found in MSWI fly ashes (Le Forestier and Libourel, 1998; Bodénan and Deniard, 2003; Quina et al., 2008b; Mahieux et al., 2010). Portlandite (Ca(OH)2) and pervoskite (CaTiO3) could also be present but their identification is not univocal. 3.2. Sieving size determination Fig. 3 shows the mass distribution of BFA. The mass value for the fraction below 38 lm is the sum of the fractions obtained by wet and dry sieving. The lower and the bigger size fractions are distinguishable from the others by their higher relative mass that achieves 16%. The mass fraction of the particles larger than

Table 2 Elementary composition of BFA (average ± Standard deviation) (wt%). Element

Concentration (wt%)

Element

Concentration (wt%)

Al Ba Bi Br Ca Cl Cr Cu F Fe K Mg Mn

2.87 ± 0.038 0.120 ± 0.010 0.062 ± 0.002 0.022 ± 0.001 20.1 ± 0.012 4.59 ± 0.050 0.056 ± 0.002 0.066 ± 0.002 0.970 ± 0.130 1.46 ± 0.040 2.90 ± 0.012 1.21 ± 0.006 0.051 ± 0.002

Mo Na Ni P Pb S Sb Si Sn Sr Ti Zn Zr

0.008 ± 0.001 3.88 ± 0.057 0.013 ± 0.001 1.10 ± 0.012 0.319 ± 0.007 11.1 ± 0.040 0.036 ± 0.002 6.56 ± 0.041 0.037 ± 0.004 0.051 ± 0.001 1.06 ± 0.021 0.911 ± 0.008 0.013 ± 0.001

Please cite this article in press as: De Boom, A., Degrez, M. Combining sieving and washing, a way to treat MSWI boiler fly ash. Waste Management (2015), http://dx.doi.org/10.1016/j.wasman.2015.01.040

A. De Boom, M. Degrez / Waste Management xxx (2015) xxx–xxx

5

Fig. 2. Raw BFA – XRD pattern.

Fig. 3. Mass distribution (%) of BFA in function of the size (lm).

2.36 mm might be biased by the fact the solid was dried before sieving, which might imply particle agglomeration. An important mass fraction is also noticed between 150 and 300 lm. The composition of each size fraction is available in Table 3. The distribution of Pb and Zn is particularly interesting since these heavy metals are more concentrated in the finest fraction. Up to 39% of Pb was recovered in the fraction lower than 38 lm obtained during the wet sieving step. 3.3. Wet screening with recirculated solution and washing When dividing BFA in three size fractions, the mass distribution differed from the previous results, especially for the 38–150 lm fraction, probably because there was more solid on each sieve, which hindered a complete separation. On average, the BFA <38 lm, 38–150 lm and the >150 lm mass fractions respectively represented 13%, 28% and 59% of the initial solid. However, some large deviations were noticed from test to test, especially for the finest fraction (see Table 4). The mean water content of the fractions reached 23% for the finest fraction and 25% for two other fractions (see Table 4) but the difference is not significant, because of the standard deviation values. This confirms the hypothesis of a water content of 25%. The initial solution presented a pH value of 10.8 and a density of 1.23 kg/L, as indicated in Table 5. It contained high concentrations

in Cl, K and Na while it was almost free of heavy metals. After one wet screening, the solution pH and density increased to a slight extent; the Cl concentration remained almost constant, whereas Ca, Cr, K, Na, Pb and Zn concentrations rose. In the first filtrate, the sum of the K and Na molar equivalent values overpassed the Cl molarity; other anions, such as sulphates, would also have dissolved during the wet screening but were not analysed. When the solution was recirculated for the 2nd and 3rd wet screening, pH and density values remained stable; most of the element concentrations fell, reaching probably an equilibrium value. However, the Cr concentration kept on increasing, almost linearly. Recirculating the solution for the wet sieving would therefore not cause an additional contamination of the solids, except in Cr whose concentration may be worth monitoring. 3.4. Size-fraction washing The solid fractions were impregnated with recirculated solution after the sieving step. The first washing would thus dilute the recirculated solution remaining in the solids, in addition to dissolving some compounds. The solution density was still higher than 1 kg/L after the first washing but decreased at the second washing, except for the finest solid washing (see Table 6). The pH values of the washing solutions from the larger fractions (38–150 lm and >150 lm) were very similar and slowly decreased

Please cite this article in press as: De Boom, A., Degrez, M. Combining sieving and washing, a way to treat MSWI boiler fly ash. Waste Management (2015), http://dx.doi.org/10.1016/j.wasman.2015.01.040

(+2360) ( 2360 + 1700)

1.7 ± 0.13 0.068 ± 0.005 0.02 ± 0.02 10.4 ± 0.33 23.6 ± 0.31 0.033 ± 0.003 0.039 ± 0.002 0.84 ± 0.05 8.2 ± 0.5 0.69 ± 0.02 0.033 ± 0.003 12.6 ± 0.3 0.018 ± 0.001 0.6 ± 0.03 0.09 ± 0.02 6.5 ± 0.55 n.d. 4.4 ± 0.26 0.013 ± 0.001 0.029 ± 0.001 0.56 ± 0.05 0.48 ± 0.01 n.d. 1.7 ± 0.1 0.069 ± 0.004 0.01 ± 0.02 10.3 ± 0.21 23.5 ± 0.22 0.028 ± 0.007 0.039 ± 0.002 0.94 ± 0.09 8.1 ± 0.13 0.69 ± 0.01 0.032 ± 0.002 12.7 ± 0.23 0.02 ± 0.01 0.59 ± 0.02 0.16 ± 0.02 6.5 ± 0.29 0.005 ± 0.008 4.5 ± 0.46 0.009 ± 0.009 0.029 ± 0.001 0.57 ± 0.05 0.50 ± 0.03 0.003 ± 0.004

( 1700 + 1180) ( 1180 + 850)

2.1 ± 0.15 0.078 ± 0.003 0.04 ± 0.007 11.7 ± 0.25 20.7 ± 0.32 0.032 ± 0.003 0.043 ± 0.001 1.02 ± 0.09 6.8 ± 0.12 0.87 ± 0.04 0.034 ± 0.002 11.9 ± 0.3 0.016 ± 0.004 0.65 ± 0.04 0.175 ± 0.005 6.8 ± 0.32 0.010 ± 0.009 5.0 ± 0.27 0.017 ± 0.003 0.031 ± 0.001 0.66 ± 0.06 0.55 ± 0.01 0.006 ± 0.005 2.6 ± 0.19 0.094 ± 0.006 0.04 ± 0.005 13.3 ± 0.13 17.3 ± 0.49 0.038 ± 0.002 0.048 ± 0.001 1.18 ± 0.04 5.28 ± 0.08 1.06 ± 0.04 0.038 ± 0.003 10.7 ± 0.2 0.017 ± 0.002 0.78 ± 0.03 0.167 ± 0.009 7.0 ± 0.42 0.015 ± 0.002 6.0 ± 0.27 0.019 ± 0.004 0.037 ± 0.001 0.78 ± 0.04 0.554 ± 0.006 0.003 ± 0.004

( 850 + 600) ( 600 + 425)

3.0 ± 0.17 0.113 ± 0.006 0.043 ± 0.007 14.3 ± 0.13 14.3 ± 0.31 0.042 ± 0.001 0.054 ± 0.001 1.33 ± 0.04 4.6 ± 0.16 1.18 ± 0.05 0.042 ± 0.003 9.1 ± 0.38 0.019 ± 0.004 0.91 ± 0.06 0.167 ± 0.003 7.0 ± 0.49 0.01 ± 0.01 7.2 ± 0.59 0.018 ± 0.003 0.041 ± 0.001 0.90 ± 0.03 0.573 ± 0.001 0.011 ± 0.001 3.2 ± 0.22 0.125 ± 0.008 0.041 ± 0.006 15.3 ± 0.51 10.1 ± 0.12 0.045 ± 0.003 0.056 ± 0.002 1.57 ± 0.08 4.1 ± 0.18 1.21 ± 0.05 0.049 ± 0.003 6.3 ± 0.32 0.023 ± 0.002 1.08 ± 0.06 0.157 ± 0.008 7.3 ± 0.13 0.005 ± 0.009 9.7 ± 0.6 0.011 ± 0.009 0.044 ± 0.001 0.97 ± 0.07 0.57 ± 0.01 0.013 ± 0.001

( 425 + 300) ( 300 + 212)

3.1 ± 0.15 0.13 ± 0.01 0.042 ± 0.007 16.6 ± 0.35 7.7 ± 0.29 0.047 ± 0.001 0.061 ± 0.001 1.57 ± 0.07 3.8 ± 0.13 1.17 ± 0.02 0.049 ± 0.002 4.5 ± 0.24 0.022 ± 0.001 1.12 ± 0.04 0.15 ± 0.01 8.5 ± 0.22 0.011 ± 0.009 9.7 ± 0.33 0.017 ± 0.001 0.047 ± 0.001 1.00 ± 0.04 0.61 ± 0.02 0.013 ± 0.001 3.0 ± 0.14 0.14 ± 0.02 0.047 ± 0.007 18.3 ± 0.36 7.0 ± 0.44 0.051 ± 0.001 0.062 ± 0.001 1.53 ± 0.04 3.8 ± 0.22 1.12 ± 0.02 0.051 ± 0.002 3.9 ± 0.15 0.02 ± 0.002 1.13 ± 0.03 0.158 ± 0.009 10.0 ± 0.46 0.01 ± 0.01 7.6 ± 0.53 0.024 ± 0.003 0.049 ± 0.001 1.03 ± 0.04 0.7 ± 0.02 0.015 ± 0.001

( 212 + 150) ( 150 + 106)

2.9 ± 0.1 0.12 ± 0.01 0.054 ± 0.008 19.34 ± 0.09 6.9 ± 0.57 0.053 ± 0.001 0.062 ± 0.001 1.42 ± 0.06 3.9 ± 0.15 1.08 ± 0.02 0.054 ± 0.003 3.6 ± 0.18 0.016 ± 0.002 1.11 ± 0.02 0.168 ± 0.009 11.0 ± 0.41 n.d. 6.0 ± 0.18 0.025 ± 0.005 0.051 ± 0.001 1.02 ± 0.02 0.778 ± 0.003 0.016 ± 0.001 2.78 ± 0.08 0.117 ± 0.006 0.06 ± 0.01 19.9 ± 0.11 7.1 ± 0.72 0.051 ± 0.001 0.062 ± 0.002 1.31 ± 0.06 4.1 ± 0.2 1.06 ± 0.03 0.054 ± 0.004 3.6 ± 0.16 0.014 ± 0.002 1.05 ± 0.04 0.184 ± 0.006 11.4 ± 0.49 0.02 ± 0.02 5.1 ± 0.19 0.03 ± 0.002 0.051 ± 0.001 1.02 ± 0.01 0.871 ± 0.008 0.015 ± 0.001

( 106 + 75) ( 75 + 53)

2.64 ± 0.08 0.113 ± 0.006 0.07 ± 0.01 19.87 ± 0.09 7.3 ± 0.78 0.053 ± 0.002 0.060 ± 0.002 1.31 ± 0.03 4.2 ± 0.11 1.06 ± 0.01 0.059 ± 0.002 3.7 ± 0.25 0.015 ± 0.001 1.06 ± 0.01 0.192 ± 0.005 11.5 ± 0.62 0.01 ± 0.02 4.88 ± 0.07 0.031 ± 0.005 0.050 ± 0.001 1.01 ± 0.04 0.94 ± 0.01 0.014 ± 0.001 2.75 ± 0.09 0.113 ± 0.006 0.07 ± 0.01 19.89 ± 0.07 7.5 ± 0.65 0.054 ± 0.002 0.061 ± 0.001 1.24 ± 0.02 4.2 ± 0.19 1.09 ± 0.04 0.057 ± 0.003 3.7 ± 0.19 0.013 ± 0.002 1.02 ± 0.02 0.211 ± 0.006 11.2 ± 0.61 0.01 ± 0.02 4.9 ± 0.15 0.039 ± 0.004 0.050 ± 0.001 1.04 ± 0.03 0.98 ± 0.01 0.014 ± 0.001

( 53 + 38) ( 38) dry ( 38) wet

1.2 ± 0.71 0.07 ± 0.02 0.06 ± 0.05 14 ± 4.0 18 ± 7.1 0.04 ± 0.01 0.06 ± 0.01 0.5 ± 0.11 7 ± 1.0 0.6 ± 0.39 0.028 ± 0.008 12 ± 8.3 n.d. 0.5 ± 0.27 0.59 ± 0.05 9 ± 2.2 0.03 ± 0.03 3 ± 1.3 0.06 ± 0.02 0.033 ± 0.001 0.5 ± 0.17 1.1 ± 0.17 n.d.

Fractions (lm)

Al Ba Bi Ca Cl Cr Cu Fe K Mg Mn Na Ni P Pb S Sb Si Sn Sr Ti Zn Zr

Table 3 BFA size fractions – elementary composition (average ± standard deviation) (%) – n.d. = non detected.

1.7 ± 0.13 0.072 ± 0.005 0.01 ± 0.02 11.2 ± 0.33 23.0 ± 0.31 0.029 ± 0.003 0.038 ± 0.002 0.80 ± 0.05 9.1 ± 0.5 0.65 ± 0.02 0.031 ± 0.003 11.8 ± 0.3 0.013 ± 0.001 0.58 ± 0.03 0.16 ± 0.02 7.1 ± 0.55 0.017 ± 0.002 3.6 ± 0.26 0.018 ± 0.001 0.030 ± 0.001 0.56 ± 0.05 0.49 ± 0.01 0.003 ± 0.005

A. De Boom, M. Degrez / Waste Management xxx (2015) xxx–xxx

3.03 ± 0.03 0.095 ± 0.002 0.07 ± 0.01 19.5 ± 0.37 9±1 0.057 ± 0.004 0.065 ± 0.002 0.86 ± 0.08 0.96 ± 0.04 5.1 ± 0.25 1.05 ± 0.01 0.047 ± 0.001 4.8 ± 0.53 0.009 ± 0.002 0.9 ± 0.02 0.28 ± 0.02 10.4 ± 0.63 0.03 ± 0.03 4.5 ± 0.15 0.044 ± 0.005 0.047 ± 0.001 0.95 ± 0.02 1.02 ± 0.01

6

from step to step (10.8 after the third washing), while the pH value of the washing solution from the finest fraction was somewhat higher and remained at this higher value (12.4) (see Table 6). Several elements were dissolved during the washing; the extracted amounts are indicated in Table 6. In addition to the listed results, Cd, Cu, Ni, Sb and Se concentrations were also analysed but their presence was not detected. Solutions after the first washing contained high quantities of chlorides, K and Na, resulting from the recirculated solution dilution and from NaCl and KCl dissolution. The extraction of those elements decreased at each washing step but was still effective after the third washing. The extracted quantities of chlorides achieved a range between 2 and 5 g/kg dry solid at the treatment end, which is acceptable for a landfill for non-hazardous waste, according to the 2003/33/EC decision (EC, 2003). Ca compound extraction did not follow the same trend since the extracted Ca amounts slightly arose in solutions from the finest fraction washing and barely diminished in the solutions from the larger fractions. Some other elements limited for waste landfilling (EC, 2003) were also detected: Ba, Cr, Cu, Mo, Pb and Zn. Extracted amounts of Ba decreased from the first to the second washing and remained more or less stable at the third step. Cr quantities were very irregular and no exact trend might be proposed; Cr dissolution has been shown to be kinetically controlled and to depend on redox conditions and solid carbonation (Astrup et al., 2005), parameters which were not controlled in the present experiments. Cu was not detected during the two first washings, regardless of the fraction size; Cu dissolution was only noticed at the third washing: the smallest fraction, the highest concentration. Mo was found in each solution; the higher extracted amounts were found in the finest fraction. Pb behaviour was particularly interesting; indeed, in spite of large standard deviations, the extracted amount of Pb from the finest fraction was obviously huge compared to the larger fractions. Moreover, the extracted quantity of Pb from the finest fraction increased at the second washing. Extracted amounts of Zn were also highest for the finest fraction. By comparing the analysed elements of the last washing solution with the leaching limits given by the European decision 2003/33/EC (EC, 2003), both larger fractions (38–150 lm and >150 lm) might be acceptable for a landfill for non-hazardous waste (this should be confirmed by standardised leaching tests), while only the finest fraction (<38 lm) should be landfilled as hazardous waste, according to the limits into force in Europe. The final residues were composed of the same principal elements and compounds as the raw BFA (see Table 2 for the raw BFA composition, Table 7 for the final residues composition and Fig. 4 for the mineralogical compounds). Anhydrite, calcite, gehlenite and quartz were not removed during the treatment. The XRD pattern of the particles bigger than 150 lm presents unusual high diffraction peaks of quartz, especially at a range of 68–68.5 on the 2-Theta scale (identified by ‘‘Q?’’ on the Fig. 4). The peaks were not detected for the other residues and are generally not found in other works on MSWI fly ashes (Le Forestier and Libourel, 1998; Jiang et al., 2009; Mahieux et al., 2010) but have already been measured for coal fly ash (Chancey et al., 2010). Gypsum, most probably resulting from the hydration of anhydrite, was observed in the treated residues but not in the raw sample. The final residues contained low concentrations of Cl, K, and Na, due to the sylvite and halite dissolution, as confirmed by the XRD analyses. At the contrary, some elements, namely Ca, Cu, and Zn, presented higher concentrations in the final residues than in the raw BFA; those elements were thus supposed to be present in low soluble or insoluble compounds. The smallest fraction presents syngenite (K2Ca(SO4)2
H2O). Syngenite has already been identified in MSWI fly ashes and characterised naturally aged or washed fly ash (Fermo et al., 1998; Mangialardi et al., 1999; Bayuseno and Schmahl, 2011).

Please cite this article in press as: De Boom, A., Degrez, M. Combining sieving and washing, a way to treat MSWI boiler fly ash. Waste Management (2015), http://dx.doi.org/10.1016/j.wasman.2015.01.040

7

A. De Boom, M. Degrez / Waste Management xxx (2015) xxx–xxx Table 4 Size-fractions obtained by wet screening – mass distribution (%) and water content (g water/g humid matter). Screening

1st

Fractions (lm) Dry mass (%) Water content (g/g)

<38 6 0.27

2nd 38–150 32 0.26

>150 62 0.25

3rd

<38 20 0.21

38–150 23 0.27

>150 57 0.26

Mean

<38 14 0.21

38–150 29 0.23

>150 57 0.25

<38 13 ± 7 0.23 ± 0.03

38–150 28 ± 5 0.25 ± 0.02

>150 59 ± 3 0.252 ± 0.006

Table 5 Solutions before and after wet screening – pH, density, concentration in Ca, Cl , K, Na (g/kg solution) and Cr, Pb, Zn (mg/kg solution). Units

Initial solution

1st filtrate

2nd filtrate

3rd filtrate

pH Density

– kg/L

10.8 1.23

11.0 1.24

11.1 1.24

11.0 1.24

Ca Cl Na K

g/kg solution

0.40 208 83 62

0.91 206 100 71

0.68 206 88 60

0.56 205 81 55

Cr Pb Zn

mg/kg solution

n.d. 0.7 n.d.

0.9 12.9 0.43

1.9 5.5 0.08

3.0 5.6 n.d.

n.d. = non detected.

Table 6 pH, density (kg/L) and extracted quantities (mg or g/kg dry matter) of some elements in solutions after washings of the size-fractions obtained by wet screening (average ± standard deviation). Fractions (lm)

1st washing <38

2nd washing 38–150

>150

<38

3rd washing 38–150

>150

<38

38–150

>150

Landfilling limits (EC, 2003)

pH – Density kg/L

12.1 ± 0.02 11.3 ± 0.3 11.5 ± 0.5 12.5 ± 0.2 10.9 ± 0.3 10.9 ± 0.2 12.4 ± 0.2 10.8 ± 0.3 10.8 ± 0.3 Class 1 Class 2 Class 3 1.059 ± 0.003 1.050 ± 0.005 1.052 ± 0.002 1.018 ± 0.008 1.000 ± 0.004 1.006 ± 0.003 1.00 ± 0.01 0.999 ± 0.004 0.997 ± 0.003

Ca Cl K Na

g/kg

4±2 140 ± 39 60 ± 7 52 ± 5

2±1 87 ± 4 37 ± 3 38 ± 4

2±1 93 ± 34 32 ± 2 35 ± 3

4±2 36 ± 27 27 ± 15 9±2

1.6 ± 0.5 14 ± 3 6±3 5±3

1.4 ± 0.4 12 ± 4 7±3 6±2

5±4 5.3 ± 0.4 7±5 5±6

1.5 ± 0.6 4±3 1.1 ± 0.5 0.9 ± 0.6

1.1 ± 0.3 1.6 ± 0.6 1.0 ± 0.4 0.6 ± 0.3

– 17 – –

– 10 – –

– 0.55 – –

Ba Cr Cu Mo Pb Zn

mg/kg

0.8 ± 0.8 1±2 n.d. 5±2 258 ± 160 5±5

0.8 ± 0.8 2±2 n.d. 4±2 11 ± 19 0.8 ± 1.3

3±1 0.3 ± 0.2 n.d. 3±1 11 ± 11 2±3

0.3 ± 0.5 1±1 n.d. 4±5 673 ± 266 12 ± 6

0.4 ± 0.7 1.5 ± 1.5 n.d. 2±2 2±3 n.d.

0.4 ± 0.7 1±1 n.d. 2±2 2±4 n.d.

0.5 ± 0.9 1±2 0.4 ± 0.8 3±2 443 ± 355 4±2

0.4 ± 0.7 1±1 0.3 ± 0.5 1.6 ± 1.6 2±3 n.d.

0.4 ± 0.7 1±1 0.2 ± 0.3 1.6 ± 1.4 2±4 n.d.

100 25 50 20 25 90

30 4 25 5 5 25

7 0.2 0.9 0.3 0.2 2

n.d. = non detected.

3.5. Mass balance The raw solid heterogeneity may partly explain the large standard deviations of the results. The experiment mass balance allows checking the solid heterogeneity. To estimate the mass balance, the inflow of the wet screening was calculated on the basis of the raw BFA and initial recirculated solution composition, while the outflow was estimated by the three humid solid fractions and the final solution. However, the solid fraction composition has not been analysed after the wet screening but at the end of the washing steps, since those fractions were entirely used for the subsequent washings. The solid fraction contribution to the outflow was thus calculated on the basis of the composition of the final residue and of the different washing solutions, taking into account the masses and volumes used for each step. Doing so, the wet screening mass balance can be estimated (see Table 8). Ca, Cl, K and Na were initially present in the synthetic recycling solution in large quantities; their inflow values were thus quite high and exceeded 1 g/g of the initial BFA. The mass balance shows that the more soluble elements (Cl, K, Na) presented some deficits; the recovery of Pb was

also not complete. On the contrary, mass balance of trace elements achieved more than 100%; this could be due to the solution analysis, which was more accurate than the solid analysis precision. Anyway, the outflow values were comparable between the three tests, in a smaller extent for K and Na. So, the deviations observed during the washings might not be related to the solid heterogeneity, since almost the same amount of each analysed element was obtained for each wet screening. Further investigations are thus needed to explain those large deviations. 3.6. Flow-sheet proposal Combining sieving and washing appears thus to be a solution to obtain non-hazardous residues (according to the limits of the 2003/33/CE directive) from MSWI fly ashes. However, such a treatment consumes huge amounts of water. The treatment may only be interesting if, at the contrary, it limits the water consumption but also, the energy and reagent one. So, a flow-sheet is proposed, based on sieving and washing but integrating the recirculation of the solution (see Fig. 5). Based on the sieving results, the sieving was limited to one size, leading to two fractions, as only the

Please cite this article in press as: De Boom, A., Degrez, M. Combining sieving and washing, a way to treat MSWI boiler fly ash. Waste Management (2015), http://dx.doi.org/10.1016/j.wasman.2015.01.040

8

A. De Boom, M. Degrez / Waste Management xxx (2015) xxx–xxx

Table 7 Final residues – composition (average ± standard deviation) (wt%). Fractions (lm)

<38

38–150

>150

Concentration (wt%) Al Ba Bi Ca Cd Cl Cr Cu Fe K Mg Mn Mo Na Ni P Pb S Sb Si Sn Sr Ti Zn Zr

2.7 ± 0.37 0.16 ± 0.01 0.16 ± 0.05 26 ± 0.13 0.02 ± 0.002 0.75 ± 0.005 0.08 ± 0.02 0.12 ± 0.02 1.1 ± 0.18 0.65 ± 0.07 1.5 ± 0.36 0.06 ± 0.006 0.005 ± 0.001 0.26 ± 0.03 0.01 ± 0.002 1.3 ± 0.17 0.79 ± 0.06 12 ± 2 0.06 ± 0.007 5.7 ± 1 0.07 ± 0.003 0.06 ± 0.003 1.1 ± 0.2 2.2 ± 0.3 n.d.

3.8 ± 0.17 0.15 ± 0.006 0.09 ± 0.02 24 ± 0.8 n.d. 0.70 ± 0.007 0.07 ± 0.001 0.09 ± 0.005 1.6 ± 0.08 0.54 ± 0.02 1.4 ± 0.04 0.07 ± 0.001 0.007 ± 0.001 0.31 ± 0.07 0.02 ± 0.001 1.4 ± 0.03 0.27 ± 0.09 12 ± 0.05 0.06 ± 0.01 5.9 ± 0.23 0.05 ± 0.01 0.06 ± 0.002 1.3 ± 0.08 1.2 ± 0.16 0.02 ± 0.0004

4.6 ± 0.09 0.16 ± 0.009 0.06 ± 0.01 21 ± 0.44 n.d. 0.59 ± 0.04 0.06 ± 0.002 0.08 ± 0.001 2.0 ± 0.07 0.78 ± 0.01 1.6 ± 0.08 0.07 ± 0.003 0.01 ± 0.001 0.62 ± 0.01 0.02 ± 0.002 1.5 ± 0.06 0.25 ± 0.04 10 ± 0.27 0.04 ± 0.004 8.6 ± 0.32 0.03 ± 0.001 0.06 ± 0.001 1.4 ± 0.01 0.96 ± 0.03 0.02 ± 0.001

n.d. = non detected.

separation at 38 lm was interesting. In the flow-sheet, raw BFA are mixed with a recirculated water solution and sieved at a mesh size of ‘‘X’’ lm in two fractions. The mesh size is chosen in order to minimise the weight of the most contaminated fraction. In the studied case, X was fixed at 38 lm. Both fractions are washed. The washing in itself may comprise several steps and is preferably designed in a counter-current way, with L/S ratio higher than 2 L/ kg for each step. The low-contaminated fraction should be acceptable for landfilling after the treatment. The high-contaminated fraction is supposed to be further treated, especially to remove heavy metal compounds such as Pb and Zn ones. The water should also be treated before being recirculated. Salts (essentially chlorides) should be extracted to avoid precipitation during the ash treatment. A modelling gave a global L/S ratio of 0.3 (De Boom, 2009). The current used treatment costs from 150 to 200 € per ton (prices communicated by MSWI managers). To be economically acceptable, a new treatment should thus cost less, by taking into account the cost of goods sold but also the investment for a new treatment plant. To treat 8 kt of MSWI fly ashes per year, a capital expenditure of one million euros is suggested; cost of goods sold is evaluated to 500 k€ per year and human resources to 200 k€/year. A treatment price of 100 €/t will guarantee a positive net value after 4 years for the treatment plant (evaluated by considering a discount rate of 0.05 and an inflation rate of 1.02) and will represent a discount of minimum 45 €/t for the MSWI plant (360 k€/y).

Fig. 4. Final residues – XRD patterns.

Please cite this article in press as: De Boom, A., Degrez, M. Combining sieving and washing, a way to treat MSWI boiler fly ash. Waste Management (2015), http://dx.doi.org/10.1016/j.wasman.2015.01.040

9

A. De Boom, M. Degrez / Waste Management xxx (2015) xxx–xxx Table 8 Mass balance of the wet screenings (mg/g of initial BFA). In Wet screening

1st

Unit

mg/g ash

Ba Ca Cl Cr Cu K Mo Na Pb Zn

1.2 204 1826 0.56 0.66 558 0.08 749 3.2 9.1

Out 2nd

3rd

1st

Out/In 2nd

3rd

1st

2nd

3rd

>100 >100 80 >100 >100 70 >100 69 88 >100

>100 >100 80 >100 >100 74 84 72 89 >100

% 1.2 210 2209 0.56 0.66 773 0.09 1087 3.3 9.1

1.2 208 2228 0.58 0.66 661 0.10 969 3.3 9.1

1.4 201 1729 0.56 0.73 611 0.09 838 3.1 9.7

1.5 219 1757 0.60 0.78 539 0.11 752 2.9 10

1.4 209 1789 0.64 0.78 490 0.09 695 2.9 10

>100 98 95 >100 >100 >100 >100 >100 96 >100

Fig. 5. Flow-sheet proposal (X = sieving size).

4. Conclusion Boiler Fly Ashes (BFA) from Municipal Solid Waste Incinerator (MSWI) are mainly composed of Ca, S and Si; however, they also contain chlorides and heavy metals, which determine their hazardous character, according to the acceptance criteria for landfill (EC, 2003). BFA must thus be treated before landfill or before a potential valorisation. Chlorides are mostly easily soluble salts and can thus be removed by washing. However, washing produces contaminated wastewater that should be treated. In order to reduce the wastewater amount and contamination, a treatment has been developed, based on wet sieving in a recirculated solution, followed by washing steps. The wet sieving is aimed to get a less contaminated fraction and a more contaminated one; the washing is used to remove the soluble salts, essentially chlorides. In this study, BFA from a Belgian MSWI underwent the developed treatment. The wet sieving was performed into recirculated solution which stood for a recycled solution from a hypothetical counter-current process. After a whole granulochemistry, optimal sieving sizes have been determined and BFA were separated at 38 and 150 lm. Three wet screenings were performed, by recycling the solution. Major elements (Cl, K, Na) concentrations fell during the wet screenings to reach equilibrium. The same trend was

noticed for Pb and Zn, while the Cr concentration kept on increasing. After the wet screening, the solid fractions were washed three times with fresh water. The finest fraction appeared to be more contaminated in Pb since around 250 mg/kg were extracted at the first washing, while only 11 mg/kg came out from the larger fraction. Concerning the Pb contamination, separating the BFA in two size fractions seem to be a simple and efficient way to obtain a less loaded fraction (>38 lm) and a more contaminated fraction (<38 lm), accounting for 87% and 13% of the initial solid on average, respectively. The other analysed elements also suggested limiting the screening at one size, the smallest (38 lm). These results highlight the interest of a size-based separation in a treatment process for MSWI BFA, especially to remove the finest fraction. A one-size sieving would in that case be sufficient, since both larger fractions investigated in the study presented similar composition and leaching behaviour along the washing steps. As the finest fraction seemed to be more contaminated, a specific treatment should be applied on the different size fractions, for which different developments might be foreseen. Ideally, some metals would be extracted from the finest fraction, while the coarsest fraction would be used in some materials, like cementbased ones. Such an application is developed in the second part of the study.

Please cite this article in press as: De Boom, A., Degrez, M. Combining sieving and washing, a way to treat MSWI boiler fly ash. Waste Management (2015), http://dx.doi.org/10.1016/j.wasman.2015.01.040

10

A. De Boom, M. Degrez / Waste Management xxx (2015) xxx–xxx

Acknowledgements The authors would like to thank the former Chemicals and Materials department, now called 4MAT (ULB, Brussels, Belgium), especially Prof. M.P. Delplancke, for the analysis equipment, and the ‘‘Centre Terre et Pierre’’ (Tournai, Belgium) for their advice and sieving equipment. The Walloon Region and the European Social Funds funded this research work, as a ‘‘First Europe’’ project.

References Alba, N., Gassó, S., Lacorte, T., Baldasano, J.M., 1997. Characterization of municipal solid waste incineration residues from facilities with different air pollution control systems. J. Air Waste Manag. Assoc. 47, 1170–1179. Astrup, T., Rosenblad, C., Trapp, S., Christensen, T.H., 2005. Chromium release from waste incineration air-pollution-control residues. Environ. Sci. Technol. 39, 3321–3329. Bayuseno, A.P., Schmahl, W.W., 2011. Characterization of MSWI fly ash through mineralogy and water extraction. Resour. Conserv. Recycl. 55, 524–534. Belevi, H., Moench, H., 2000. Factors determining the element behavior in municipal solid waste incinerators. 1. Field studies. Environ. Sci. Technol. 34, 2501–2506. Bodénan, F., Deniard, P., 2003. Characterization of flue gas cleaning residues from European solid waste incinerators: assessment of various Ca-based sorbent processes. Chemosphere 51, 335–347. Chancey, R.T., Stutzman, P., Juenger, M.C.G., Fowler, D.W., 2010. Comprehensive phase characterization of crystalline and amorphous phases of a Class F fly ash. Cem. Concr. Res. 40 (1), 146–156. Chandler, A.J., Eighmy, T.T., Hartlén, J., Hjelmar, O., Kosson, D.S., Sawel, S.E., van der Sloot, H.A., Vehlow, J., 1997. Municipal Solid Waste Incinerator Residues – IAWG (International Ash Working Group). Elsevier. Chiang, K.-Y., Jih, J.-C., Chien, M.-D., 2008. The acid extraction of metals from municipal solid waste incinerator products. Hydrometallurgy 93, 16–22. Chimenos, J.M., Fernández, A.I., Cervantes, A., Miralles, L., Fernández, M.A., Espiell, F., 2005. Optimizing the APC residue washing process to minimize the release of chloride and heavy metals. Waste Manage. 25, 686–693. De Boom, A., 2009. Contribution à l’élaboration d’un procédé de valorisation des cendres volantes et de résidus d’épuration des fumées d’incinération d’ordures ménagères (in French). Ph.D. thesis. Université libre de Bruxelles (U.L.B.).. De Boom, A., Degrez, M., 2007. In: Characterisation of Different Fractions Obtained from Air Pollution Control Residues from Municipal Solid Waste Incineration with a View to Valorising them. Sardinia 2007 – Eleventh International Waste Management and Landfill Symposium, Cagliari, Italy.. De Boom, A., Degrez, M., 2012. Belgian MSWI fly ashes and APC residues: a characterisation study. Waste Manage. 32, 1163–1170. De Boom, A., Degrez, M., Hubaux, P., Lucion, C., 2011. MSWI boiler fly ashes: magnetic separation for material recovery. Waste Manage. 31, 1505–1513. De Boom, A., Fabien, A., Staquet, S., Degrez, M., 2015. Valorisation of MSWI fly ashes by combining sieving and washing: Part 2 – Incorporation of treated fly ashes into cement-based materials. (in preparation). Derie, R., 1996. A new way to stabilize fly ash from municipal incinerators. Waste Manage. 16 (8), 711–716. EC, 2003. Council decision 2003/33/EC of 19 December 2002 establishing criteria and procedures for the acceptance of waste at landfills pursuant to article 16 of and annex II to directive 1999/31/EC. Off. J. Euro. Commun. L11, 27–49. EC, 2008. Directive 2008/98/EC of the European parliament and of the council of 19 November 2008 on waste and repealing certain directives. Off. J. Euro. Union L312, 3–30. ELFM (Enhanced Landfill Mining) Symposium, 2010. (accessed on May 2011).. Eurostat, 2012. Eurostat news release 48/2012 – 27 March 2012, Environment in the EU27, Landfill Still Accounted for Nearly 40% of Municipal Waste Treated in the EU27 in 2010, pp. 3. (access on December 2012).. Fermo, P., Cariati, F., Pozzi, A., Demartin, F., Tettamanti, M., Collina, E., Lasagni, M., Pitea, D., Puglisi, O., Russo, U., 1998. The analytical characterization of municipal solid waste incinerator fly ash: methods and preliminary results. Fresenius J. Anal. Chem. 365, 666–673. Fuerstenau, M.C., Han, K.N. (Eds.), 2003. Principles of Mineral Processing, Society for Mining, Metallurgy, and Exploration (SME), pp. 584.. Grosso, M., Biganzoli, L., Rigamonti, L., 2011. A quantitative estimate of potential aluminium recovery from incineration bottom ashes. Resour. Conserv. Recycl. 55 (12), 1178–1184.

Hammy, F., Mercier, G., Blais, J.-F., 2005. Caractérisation et traitement des résidus de contrôle de la pollution de l’air (RCPA) d’incinérateurs de déchets municipaux par procédé de lixiviation en milieu basique. J. Environ. Eng. Sci. 4, 287–298. Hong, K.-J., Tokunaga, S., Ishigami, Y., Kajiuchi, T., 2000a. Extraction of heavy metals from MSW incinerator fly ash using saponins. Chemosphere 41, 345–352. Hong, K.-J., Tokunaga, S., Kajiuchi, T., 2000b. Extraction of heavy metals from MSW incinerator fly ashes by chelating agents. J. Hazard. Mater. B75, 57–73. Hyks, J., Astrup, T., Christensen, T.H., 2009. Long-term leaching from MSWI airpollution-control residues: leaching characterization and modeling. J. Hazard. Mater. 162, 80–91. Janos, P., Wildnerová, M., Loucka, T., 2002. Leaching of metals from fly ashes in the presence of complexing agents. Waste Manage. 22, 783–789. Jiang, Y., Xi, B., Li, X., Zhang, L., Wei, Z., 2009. Effect of water-extraction on characteristics of melting and solidification of fly ash from municipal solid waste incinerator. J. Hazard. Mater. 161, 871–877. Katsuura, H., Inoue, T., Hiraoka, M., Sakai, S., 1996. Full-scale plant study on fly ash treatment by the acid extraction process. Waste Management – Cycle and Stabilization Technologies of MSW Incineration Residue 16, 491–499. Le Forestier, L., Libourel, G., 1998. Characterization of flue gas residues from municipal solid waste combustors. Environ. Sci. Technol. 32 (15), 2250–2256. Levasseur, B., Blais, J.-F., Mercier, G., 2005. Study of the metal precipitation from decontamination leachates of municipal wastes fly ash incinerators. Environ. Technol. 26, 421–432. Liu, Y., Zheng, L., Li, X., Xie, S., 2009. SEM/EDS and XRD characterization of raw and washed MSWI fly ash sintered at different temperatures. J. Hazard. Mater. 162, 161–173. Mahieux, P.Y., Aubert, J.E., Cyr, M., Coutand, M., Husson, B., 2010. Quantitative mineralogical composition of complex mineral wastes – contribution of the Rietveld method. Waste Manage. 30, 378–388. Mangialardi, T., 2003. Disposal of MSWI fly ash through a combined washingimmobilisation process. J. Hazard. Mater. 98, 225–240. Mangialardi, T., Paolini, A.E., Polettini, A., Sirini, P., 1999. Optimization of the solidification/stabilization process of MSW fly ash in cementitious matrices. J. Hazard. Mater. B70, 53–70. Nagib, S., Inoue, K., 2000. Recovery of lead and zinc from fly ash generated from municipal incineration plants by means of acid and/or alkaline leaching. Hydrometallurgy 56, 269–292. Pedersen, A.J., 2002. Evaluation of assisting agents for electrodialytic removal of Cd, Pb, Zn, Cu and Cr from MSWI fly ash. J. Hazard. Mater. 95, 185–198. Piantone, P., Bodénan, F., Derie, R., Depelsenaire, G., 2003. Monitoring the stabilization of municipal solid waste incineration fly ash by phosphation: mineralogical and balance approach. Waste Manage. 23, 225–243. Quina, M.J., Bordado, J.C., Quinta-Ferreira, R.M., 2008a. Treatment and use of air pollution control residues from MSW incineration: an overview. Waste Manage. 28, 2097–2121. Quina, M.J., Santos, R.C., Bordado, J.C., Quinta-Ferreira, R.M., 2008b. Characterization of air pollution control residues produced in a municipal solid waste incinerator in Portugal. J. Hazard. Mater. 152, 853–869. Van Gerven, T., Geysen, D., Stoffels, L., Jaspers, M., Wauters, G., Vandecasteele, C., 2005. Management of incinerator residues in Flanders (Belgium) and in neighbouring countries. A comparison. Waste Manage. 25, 75–87. Van Gerven, T., Geysen, D., Pontikes, Y., Cizer, O., Mertens, G., Elsen, J., Van Balen, K., Jones, P.T., Blanpain, B., 2010. An integrated materials valorisation scheme for Enhanced Landfill Mining. In: Symposium Proceedings – Enhanced Landfill Mining and the Transition to Sustainable Materials Management 4–6 October 2010, Molenheide (Houthalen-Helchteren, Belgium).. Vandecasteele, C., Wauters, G., Arickx, S., Jaspers, M., Van Gerven, T., 2007. Integrated municipal solid waste treatment using a grate furnace incinerator: In: The Indaver Case. Waste Management – Wascon 2006 6th International Conference: Developments in the Re-use of Mineral Waste 27, pp. 1366–1375.. Wang, K.-S., Chiang, K.-Y., Lin, K.-L., Sun, C.-J., 2001. Effects of a water-extraction process on heavy metal behavior in municipal solid waste incinerator fly ash. Hydrometallurgy 62, 73–81. Wang, K.-S., Sun, C.-J., Yeh, C.-C., 2002. The thermotreatment of MSW incinerator fly ash for use as an aggregate: a study of the characteristics of size-fractioning. Resour. Conserv. Recycl. 35, 177–190. Wang, L., Jin, Y., Nie, Y., Li, R., 2010. Recycling of municipal solid waste incineration fly ash for ordinary Portland cement production: a real-scale test. Resour. Conserv. Recycl. 54, 1428–1435. Wilewska-Bien, M., Lundberg, M., Steenari, B.M., Theliander, H., 2007. Treatment process for MSW combustion fly ash laboratory and pilot plant experiments. Waste Manage. 27, 1213–1224. Youcai, Z., Lijie, S., Guojian, L., 2002. Chemical stabilization of MSW incinerator fly ashes. J. Hazard. Mater. 95, 47–63. Zhang, F.-S., Itoh, H., 2006. Extraction of metals from municipal solid waste incinerator fly ash by hydrothermal process. J. Hazard. Mater. B136, 663–670.

Please cite this article in press as: De Boom, A., Degrez, M. Combining sieving and washing, a way to treat MSWI boiler fly ash. Waste Management (2015), http://dx.doi.org/10.1016/j.wasman.2015.01.040