Fractions of Zinc, Chromium and Cobalt in Municipal Solid Waste Incineration Bottom Ash

In the fresh samples of municipal solid waste incineration (MSWI) bottom ash, the fractional composition of Zn, Cr and Co was investigated. The BCR method was utilised to evaluate the acid soluble and exchangeable fraction (F1), reducible fraction (F2), oxidizable fraction (F3) and residual fraction (F4). Physico-chemical parameters such as pH, dry mass, and pseudo-total metal content were also determined. The percentage of zinc in fractions was in the following order: F3 (31.8%) > F1 (30.6%) > F2 (27.8%) > F4 (9.8%), for chromium F4 (88.7%) > F3 (5.4%) > F1 (3.2%) > F2 (2.7%), and in the case of cobalt F4 (59.8%) > F1 (14.3%) > F2 (13.6%) > F3 (12.3%). Mobile metal pool (F1–F3) gathered the most of zinc (90.2%) and the least of chromium (11.3%).


INTRODUCTION
The amount of municipal solid waste will be increasing in the coming years, mainly due to the progressive urbanization and living standards. Therefore, the waste disposal problems will be also aggravated (Vishnu and Singh 2021).
One of the widely used procedures of disposal is incineration, which reduces the solid waste volume by approx. 90%, which translates into the mass reduction by 70% (Vaitkus et al. 2019). This process is currently the main approach to municipal solid waste management in Europe (Minane et al. 2017). MSWI bottom ash is the main solid residue from the waste incineration. It comprises even 20-25% of the waste introduced into the incinerator (Huber et al. 2021). It is a potential source of environmental threat, due to the content of heavy metals and other harmful pollutants (Huang et al. 2020). MSWI bottom ash is utilized as concrete or loose aggregates and raw material in the production of cement and ceramic materials (Verbinnen et al. 2017). Nowadays, in Europe about 50% of countries are utilizing the processed bottom ash as a construction material (Vateva and Laner 2020).
The aim of this study was determination of fractional composition of zinc, chromium and cobalt in particle size fraction ≤2mm, since this fraction accumulates the most of heavy metals contained in MSWI bottom ash (Chimenos et al., 1999). In order to evaluate the fractional composition of metals in bottom ash, the sequential extraction method is widely used (Bruder-Hubscher et al., 2002;Kuokkanen et al., 2006;Haberl and Schuster, 2019). The main advantage of the BCR method used in this study is harmonization (Salam et al. 2019), which has a major impact on the comparability of results at international level. This extraction scheme illustrates the processes which occur in the environment, related to metal fractionation and thus its solubility (Rodgers et al. 2015).

Sample collection
Twenty samples of the fresh bottom ash, approx. 1 kg each, were derived from MSWI plant in Białystok, Poland (53°09'30" N, 23°10'07" E). They were collected during 7 weeks in 2016, when 11650t of waste with average heating value of 8.95 MJ·kg -1 were incinerated. Incineration of mixed MSW and sorted waste ran in the stokerfired furnace at 700 °C on the grate.
Water was used to cool down the bottom ash. Ferrous metals were magnetically recovered before sampling. The samples were ground in an agate mortar, after previous sifting through a 2 mm sieve.

Physico-chemical properties
Dry mass was determined by drying at 105°C to the constant mass. Flame atomic absorption spectrometry (FAAS) was used to evaluate the pseudo-total Zn, Cr and Co content, after closed vessel microwave digestion (0.5 g sample with 3 cm 3 HClO 4 and 10 cm 3 HNO 3 ). The pH was measured in suspension (5 g of ash + 25 cm 3 of deionized water), after 24 hours using the potentiometric method.

Assessment of metal fractions
In the dry MSWI bottom ash, the fractional composition of zinc, chromium and cobalt was evaluated by the ultrasound accelerated BCR method. The extraction comprised three stages ( Figure 1).
The content of Zn, Cr and Co in fractions was determined by flame atomic absorption spectrometry (FAAS). The percentage of zinc, chromium and cobalt in fractions and the mobile pool (F1-F3) were computed. The amount of studied metals in the residual fraction (F4) was calculated as a difference between pseudo-total content and content in fractions F1-F3.

Physico-chemical properties of MSWI bottom ash
The dry mass content and pH of the investigated MSWI bottom ash were described in the previous paper (Łukowski and Olejniczak, 2020).
The studied samples contained the most of pseudo-total zinc (669.6 mgˑkg -1 ) and the least of cobalt (14.72 mgˑkg -1 DM on average) ( Table 1). Zinc content was in good accordance with the results presented by Yao et al. (2010), who noted 304.5-1922 mgˑkg -1 of Zn in bottom ash from six incinerators. The amount of cobalt was a little higher (17.4-28.07 mgˑkg -1 ) than in the present study. Vateva and Laner (2020) emphasized that the mineral part of fine fraction, which was predominant, contained 13047 mgˑkg -1 of Zn and 815 mgˑkg -1 of Cr, while the Co content was below the limit of determination. Chimenos et al. (2003) found 11096 mgˑkg -1 (48.3%) of Zn and 332 mgˑkg -1 (46.5%) of Cr in the discussed grain size fraction. Dou et al. (2017) reported, according to the results from many regions in the world, that the content of Zn as well as Cr and Co in the MSWI bottom ash is in approximate range 100-10000 and 10-1000 mgˑkg -1 , respectively.

Fractionation of metals in MSWI bottom ash
The most of zinc, 215.6 mgˑkg -1 on average, was found in F3 fraction (Table 1). It was 31.8% of pseudo-total content (Figure 2), in the range of 8.0 to 43.8%. Comparable result was achieved by Yao et al. (2013), since the authors noted about 28% of zinc in the oxidizable fraction. The metals in this fraction are bound to organic matter and/or sulfides (Pöykiö et al., 2016) and can be released under oxidizing conditions. Microbial activity could lead to such a process as well, because biodegradation of organic matter is a source of organic acids and carbon dioxide. These compounds The metals from this fraction can be released under reductive conditions, but the probability of such a process is lower, than in the case of acid soluble and exchangeable fraction (Zhao et al. 2021). The least of zinc (9.8%) was found in F4 fraction. It was also the lowest share among all the studied elements, which probably arose from its volatility. It seems that this attribute also influenced the near-even Zn distribution in F1-F3 fractions. Zinc is regarded as semi-volatile element, whereas Cr and Co are non-volatile (Yang et al., 2016), which is confirmed in the present study by their high content in residual fraction. Zn mobile pool, 90.2% at the range from 74 to 98%, was the highest among studied metals, which means that almost entire metal can be potentially mobilized, as a result of changing environmental conditions.
The highest amount of chromium, 300.1 mgˑkg -1 on average, was found in the F4 fraction. It constituted 88.7% of pseudo-total content on average, at the range from 83 to 95%. This was the largest percentage among investigated elements. It means that Cr was mainly bound to the vitreous and silicate phases (Gonzales et al., 2019) and indicates its very low leachability. Similar result, about 96% in residual fraction, was stated by Gonzales et al. (2019). According to the authors, such a high share is obvious, since this element is a glass component. They also emphasised that only small part of chromium (4%), bound to oxyhydroxides and organic matter, was mobilized under redox conditions. The present investigations confirmed such a claim, because F2 and F3 fractions gathered merely 8.1% of Cr. Additionally, the entire Cr mobile pool amounted to 11.3%, within the range of 5-17%, and was the lowest among studied metals. Thus, potential chromium release to the environment was also the lowest. It should be borne in mind that chromium is considered as very harmful and even small amounts pose a significant hazard for the ecosystem (Nilanjana and Lazar 2011;Coetzee et al., 2020).
Cobalt, similarly to Cr, was accumulated mainly in F4 fraction (14.72 mgˑkg -1 on average). It was 59.8% of pseudo-total content,