Treatment trains for improving the leaching behavior of the mineral fraction of different types of waste incineration bottom ash in view of recycling under unrestricted conditions

Waste incineration is generally classified as an energy recovery strategy. Nevertheless, its role in terms of material recovery/recycling has been also recently recognized. Strategies for valorizing the main by-product of this process¸ bottom ash, which represents about 20% of the mass of the treated fuel (such as Municipal Solid Waste MSW, the dry fraction of MSW, or Refuse Derived Fuel RDF) are being increasing investigated and implemented. Lately, in Europe, bottom ash (BA) treatment plants have begun operation with the main aim of recovering valuable fractions such as ferrous and non-ferrous metals. However, adequate management strategies for the mineral fraction (over 80% by weight of the BA) are still lacking. This material presents similar characteristics to natural materials used in construction applications and could be employed as a filler or aggregate in road subbases or in concrete. The main barrier for bottom ash reuse is the leaching of heavy metals and/or metalloids. This issue has been significantly investigated employing several types of both laboratory and field tests. However, the leaching behaviour of the residual fraction after pre-treatment for metals removal has been significantly less studied up to now, as well as the properties of BA resulting from dry extraction processes, which have been shown to facilitate the recovery of ferrous and non-ferrous metals. Another interesting aspect for the management of the mineral fraction of waste incineration BA is that in view of the EU’s circular economy strategy some Countries have already started to set recovery targets for this material that require its application under free conditions, i.e. as natural aggregates; this implies that in order to avoid potential environmental pollution, the material should meet restrictive leaching guidelines. Depending on the characteristics of the feed waste treated in the incineration plant, but also on the type of BA extraction method and pre-treatment processes applied, several different constituents may prove critical. Therefore, a single treatment may not be sufficient to improve the behavior of the material in view of meeting the requirements for free use in terms of leaching of potential environmental contaminants. In addition, also the types of treatments required and their effects may depend on the type of waste treated in the incineration plant (e.g. RDF or MSW) and on the BA extraction method (wet or dry) employed. The aim of this doctoral thesis was to evaluate treatment trains for improving the leaching behavior of the mineral fraction in view of allowing its reuse under unrestricted conditions. In particular, a characterization study was carried out to examine the leaching behavior of the selected samples (A - the mineral fraction of BA from the incineration of the dry fraction of MSW, B - the mineral fraction of BA from MSW incineration, C – dry extracted BA from RDF incineration in a rotary kiln), all three of which have been significantly less investigated compared to MSW incineration BA. To this end, geochemical modeling was used in order to identify the main mechanisms that control the leaching of major constituents and trace metals of the examined samples comparing them to the data reported for other types of BA. Experimental leaching tests were conducted on the as received material. Specifically, sample A showed a high release of Cr, Chlorides and an alkaline pH (pH= 12.45). The results of geochemical modelling indicated that the release of Cr was probably controlled by a Ba sulfate phase Ba[SCr]O4[96%SO4]; chlorides release appeared to be controlled by Calcium aluminate complex salts. Sample B showed a high release of Al, Cu, Mg, Sb, Chlorides and Sulphates and lower pH value compared to the previous sample (i.e. pH=10.33). The high release of Cu could have depended on the presence of dissolved reactive organic carbon (resulted much higher than that of the other two samples). The results of the geochemical modelling indicated that the release of Al was probably controlled by Al[OH]2; Mg release appeared to be controlled by Brucite; while sulphates release was probably controlled by Anhydrite. Sample C showed a lower release of Al, Cu, Sb, Si and SO4 compared to the other samples and higher release of Ba, Pb, Zn and an alkaline pH (pH= 12.83). The concentration of Al for pH values higher than 10 was one order of magnitude lower compared to samples A and B, probably due to the presence of different phases, such as Al[OH]3, controlling the release in that pH region. Wollastonite controlled the release of Si; while, Pb[OH]2 and Zincite controlled the release of Pb and Zn. Finally, even if the slag’s Sb content showed a higher content in the solid matrix compared to other two samples, its release is was lower. This could be due to the presence of a Fe[OH]3 phase, found in the geochemical modeling results, not present in the other two samples. Subsequently, different treatments were tested to improve the leaching behaviour of each type of tested sample. In particular, accelerated carbonation treatments, performed via two routes, wet and slurry phase were tested on all three samples but in particular on sample A, that presented the highest reactivity towards CO2. It was shown that accelerated carbonation reduced the pH of the eluate from 12 to 10 and decreased the leaching of Cr, Cu, Pb and Zn. Moreover, significant differences were achieved as a function of the type of carbonation route employed. Indeed, for sample A, wet carbonation at T=50 °C and p=3 bar allowed to reduce the leaching of Ba and Zn; while, slurry carbonation at T=50 °C and 100°C and p=3 bar, allowed to reduce Cr, Cu, Mo and Cl release. On the basis of XRD and acid neutralization capacity analysis, for Sample B a limited content of potentially reactive phases with CO2, such as Ca hydroxides or silicates species, was found. This suggested that the carbonation treatment may not be an efficient treatment for this type of residue and that, instead, just washing could improve its leaching behavior. However, in order to better investigate this issue, both washing, at different temperature conditions, and carbonation treatments were performed. The washing treatment decreased the release of all constituents, except for Mg and Cr. While, the product of slurry phase carbonation performed at T=100 °C, displayed a lower release of Ba, Cr and Mo. Moreover, only for the slurry phase process, also the release of Cl decreased, especially at T=50 °C. Finally, the products of carbonation processes applied to sample C showed a decrease of the release of Ba, Pb and Zn, but an increase of Sb, Si and SO4 mobility. Chlorides release, instead, increased after wet carbonation, but decreased after slurry phase treatment, similarly to what found for sample A. Overall, slurry phase carbonation showed to be promising for bottom ash, as it could also reduce the release of salts; however, it can lead to the mobilization of elements, such as Sb. Hence, the addition of Fe (hydr)oxide phases synthesized from FeSO4*7H2O, which XRD analysis showed to be goethite and magnetite after three washing steps, was applied following two methodologies: a stepwise or direct procedure. This treatment allowed to reduce the release of antimony and to further improve the leaching behavior of other contaminants. For sample A, this treatment was applied on the products of both wet route and slurry phase carbonation. The results showed that, employing step-wise procedure with an amount of synthesized (hydr)oxides corresponding to the addition of 2% of Fe by weight of the sample, allowed to decrease the release of Cu, Mo and Cl from the product of the wet carbonation route. The application of the direct procedure with the same amount of Fe addition, to the product of the wet carbonation route reduced the leaching of Ba, Cr and Sb. Regarding the slurry phase carbonation treatment, only the direct procedure was applied, after or during the carbonation treatment. The results showed how the addition of Fe (hydr)oxides, in both cases improved the release of Ba, Cu, Cl, and also of Sb. For sample B, instead, this treatment was applied to the as received sample. The stepwise procedure showed to decrease the release of Mo, Zn, Cl and SO4, while through the direct procedure the release of Ba, Cr and Cu decreased the most. Both procedures lead to a lower release of Sb, probably due to binding with magnetite, detected by XRD in the solid product. Finally, for the direct procedure, the influence of the addition of different amounts of Fe (hydr)oxides was examined. The results showed that increasing the amount of Fe (hydr)oxide suspension added decreased the release of Sb, until around a 5% Fe addition by weight of the sample. In conclusion, the overall results of this study showed that the environmental behaviour of the mineral fraction generated from the treatment of BA from different waste to energy plants may vary significantly depending on the origin of the sample and in particular on the characteristics of the feed waste. In particular, for mineral fractions or BA generated from RDF or dry fraction incineration and which present an alkaline pH (samples A and C), slurry phase carbonation with subsequent addition of Fe (hydr)oxides appears to be a particularly suitable treatment. While, for the mineral fraction deriving from the incineration of residual MSW (sample B), presenting a lower pH but high salts and DOC release, a washing treatment is necessary. If, as in this case, the material presents a high Sb release, treatment with the addition of Fe (hydr)oxides may be particularly appropriate.