Remediation potential of mining, agro-industrial, and urban wastes against acid mine drainage

Acid mine drainage (AMD) poses serious consequences for human health and ecosystems. Novel strategies for its treatment involve the use of wastes. This paper evaluates the remediation potential of wastes from urban, mining and agro-industrial activities to address acidity and high concentrations of potentially toxic elements (PTE) in AMD. Samples of these waste products were spiked with an artificially prepared AMD, then pH, electrical conductivity (EC), and PTE concentrations in the leachates were measured. The artificial AMD obtained through oxidation of Aznalcóllar’s tailing showed an ultra-acid character (pH − 2.89 ± 0.03) and extreme high electrical conductivity (EC − 3.76 ± 0.14 dS m−1). Moreover, most PTE were above maximum regulatory levels in natural and irrigation waters. Wastes studied had a very high acid neutralising capacity, as well as a strong capacity to immobilise PTE. Inorganic wastes, together with vermicompost from pruning, reduced most PTE concentrations by over 95%, while organic wastes retained between 50 and 95%. Thus, a wide range of urban, mining, and agro-industrial wastes have a high potential to be used in the treatment of AMD. This study provides valuable input for the development of new eco-technologies based on the combination of wastes (eg. Technosols, permeable reactive barriers) to remediate degraded environments.

www.nature.com/scientificreports/ UV-vis spectrophotometer (Thermo Fisher Scientific Inc., Waltham, MA, USA) using a solution of ammonium molybdate and ascorbic acid 38 . Furthermore, heterotrophic respiration was measured by determining the CO 2 flux from waste material with a microbiological analyser μ-Trac 4200 SY-LAB model (SY-LAB Geräte GmbH, Neupurkersdorf, Austria) according to ISO 17,155 39 and the results were expressed as the basal respiration rate (BR) in μg CO 2 day −1 kg soil −1 . Total concentrations of PTE (As, Cd, Co, Cr, Cu, Fe, Ni, Pb, Sb, V, and Zn) were analysed in triplicate by inductively coupled plasma optical emission spectrometry (ICP-OES) in a spectrometer PerkinElmer Avio 500 (PerkinElmer, Inc., Waltham, MA, USA) after acid digestion (HNO 3 + HF 3:1 V:V for inorganic wastes and HNO 3 + H 2 0 2 1:1 V:V for organic wastes) in a Mars XP1500 Plus microwave (CEM Corporation, Matthews, CN, USA). The precision and accuracy of this method were assessed by measurement (three replicates) of a certified reference material (CRM BCR-482 EC-JRC-IRMM, Geel, Belgium). For all elements of interest, measured values were within the prediction interval of the certified value.
Preparation of acid mine drainage. An artificial AMD was prepared in the laboratory following a method-based on the oxidation of pyritic tailings with hydrogen peroxide (H 2 O 2 ) 40 . In detail, this pollutant solution used as AMD was prepared by progressive addition of 1 L H 2 O 2 (33%) + 1 L H 2 O to 42.85 g of pyrite tailing and after three days, the solution was extracted by discarding the precipitated sediment, and then pH and EC (2.89 and 3.76 dS m −1 , respectively) were measured. The pyrite tailing used comes from the Aznalcóllar mine (Seville, Spain), and belongs to the 0.9 × 10 6 m 3 of toxic tailings discharged into the Agrio and Guadiamar river basin, in one of the biggest mining accidents in Europe, the Aznalcóllar's environmental disaster in 1998 [41][42][43] . The PTE concentrations in the toxic tailings immediately after the accident (Table S1) were measured in previous studies 42,44 . Acid mine drainage treatment using waste materials. All waste materials were spiked with the acid mine drainage (AMD) prepared from the oxidation of pyritic tailings. This experience was made by the addition of 50 mL of AMD to 10 g of each waste material in triplicate to check the first impact of the AMD on different www.nature.com/scientificreports/ waste materials. Afterward, they were stirred for 24 h and filtered (Filter-Lab n°1250, pore size: 10-13 µm), separating the waste material (solid phase) from the leachate (liquid phase). In the leachate, which is the AMD treated, pH (L) and EC (L) were measured with a pH/conductivity-meter 914 Metrohm (Metrohm AG, Herisau, Switzerland) and an Eutech CON700 conductivity-meter (Oakton Instruments, Vernon-Hills, IL, Waltham, MA,  USA), respectively, and PTE concentrations in solution were determined by inductively coupled plasma optical mass spectrometry (ICP-MS) in a spectrometer PerkinElmer NexION 300D (PerkinElmer, Inc., Waltham, MA, USA). The precision and accuracy of this method were assessed by measurement (three replicates) of a certified reference material (CRM BCR-482 EC-JRC-IRMM, Geel, Belgium). For all elements of interest, measured values were within the prediction interval of the certified value.

Statistical analysis.
A preliminary analysis of descriptive statistics was made. Non-parametric Kruskal-Wallis and Dunn tests (p < 0.05) for the analysis of mean comparison in the waste materials characterisation were chosen due to the sample size 45 . To analyse the results of the treatment of AMD by wastes, normality was checked with the Shapiro-Wilk test and homoscedasticity with the Levene test. As none of these conditions were met, even after transforming the variables, non-parametric Kruskal-Wallis and Dunn tests (p < 0.05) for multiple comparisons were applied. Furthermore, to analyse the influence of waste properties on their capacity of acid neutralisation and removal of PTE in polluted waters, significant bivariate Spearman's correlations were also performed. All analyses were made with a confidence level of 95% by using RStudio software (RStudio Inc., 250 Northern Ave, Boston).
Ethical approval. This study did not use any kind of human participants or human data, which require any kind of ethical approval or consent to participate.

Consent to publish.
Our study did not use any kind of individual data such as video and images.
Organic wastes showed significant differences in relation to the inorganic ones, mainly by the higher content in OC, CEC, exchangeable bases, total N, and available P (Table 1). Otherwise, differences between the organic wastes were also important. Organic carbon ranged between 10.5% in vermicompost from gardening (VC) and 28% in bio-stabilised material of municipal solid waste (BM) and composted solid olive-mill irrigated with olive leachate (OL); CEC varied between 36 cmol + kg −1 in VC and 91 cmol + kg −1 in OL; N T was between 0.6% in VC and 3.1% in composted sewage sludge (WS); and P A ranged between 134 mg kg −1 in BM and 403 mg kg -1 in compost of agricultural greenhouse (GW). For the other properties, no significant differences were observed with respect to the inorganic wastes, although among the organic wastes there were. In this way, pH ranged from 6.5 in BM and 9.5 in GW; EC was low for VC (< 0.4 dS m −1 ), very high for composted solid olive-mill irrigated with water (OW) and OL (2-4 dS m −1 ), and extremely high (> 7 dS m -1 ) for the rest; and CaCO 3 was also detected in all cases, ranging from 7.7% in BM to 24.9% in VC. Basal respiration (BR) presented a wide range of values without significant differences between inorganic and organic wastes, with maximum of 124 µg CO 2 day −1 kg −1 in CW and minimum of 14 in WS µg CO 2 day −1 kg −1 .
Total concentrations of PTE showed significant differences among wastes (Table 2). However, in general, between organic and inorganic wastes there were no clear differences, although the concentrations of Cr, Cu, Ni, and Zn were usually higher in organic wastes than in inorganic ones. Within the inorganics, IO and GS had the highest concentrations of most PTE, especially IO with concentrations of As, Pb, and Sb close to 24, 29, and 21 mg kg −1 , respectively. Otherwise, MS presented very low concentrations for As, Pb, V, and Zn; while CW showed very low concentrations for Co, Cr, Ni, Sb, and V. The organic wastes presented low concentrations of As, Cd, Co, and Sb, with values below 5.3, 2, 5, and 0.4 mg kg −1 , respectively. Lead showed differences between wastes, ranging between 2.4 mg kg −1 in OW and 52.6 mg kg −1 in BM; while V ranged between 12 mg kg −1 in BM and 27 mg kg −1 in WS. The elements with higher concentrations in relation to inorganic wastes also presented significant differences between organics; Cr ranged between 16 mg kg −1 in OW and 45 mg kg −1 in BM; Cu varied between 25 mg kg −1 in VC and 365 mg kg −1 in GW; Ni was between 9.7 mg kg −1 in OW and 21.5 mg kg −1 in BM; and Zn oscillated between 49 mg kg −1 in OW and 517 mg kg −1 in WS.
The artificial AMD prepared by oxidation of the Aznalcóllar's toxic tailings discharged in the accident showed both the typical ultra-acid character (pH (L)− 2.89 ± 0.03) and the extremely high EC (L) (3.76 ± 0.14 dS m −1 ). Moreover, most of the PTE were present in high concentrations in AMD (Table 3). Below 100 µg L −1 were Ba, Be, In, Mo, Sc, Th, Tl, U, V, and Y; between 100 and 500 µg L −1 were Bi, Cd, Co, Cr, Ni, and Sn; between 500 and 1000 µg L −1 were Pb and Sb; and above 1000 µg L −1 were As, Cu, Mn, and Zn. Acid mine drainage treatment using waste materials. All leachates obtained after waste treatment showed a pH (L) close to slightly acidic-neutral values (6-7.25), although with statistically significant differences www.nature.com/scientificreports/ Table 1 IO 7.27 ± 0.08 b 0.04 ± 0.01 a n.d 13.08 ± 0.22 b 6.34 ± 0.38 a 4.44 ± 0.39 a n.d www.nature.com/scientificreports/   www.nature.com/scientificreports/ among wastes (Fig. 2a). Whereas changes in EC (L) due to waste treatment were quite heterogenous among the waste material used (EC (L) : 2-24 dS m −1 ). Some of them (IO, MS, GS, and VC) reduced the EC (L) of the AMD; however, other wastes cause a significant increase in EC (L) (GW, WS, BM, OW, and OL) between 2-and sixfold the EC measured in the AMD (Fig. 2b). Most PTE concentrations in the soluble fraction decreased significantly after waste treatments, although with large differences in removal effectiveness between organic and inorganic wastes (Table S2). Inorganic wastes showed a higher removal effectiveness of PTE than organic wastes, excluding VC which had similar removal rates to inorganic ones (Table 4). For the main PTE (As, Cd, Cr, Cu, Pb, Sb, and Zn), the retention rate of all tested inorganic wastes (IO, MS, CW, GS) as well as VC was above 95% in most cases and close to 100% for many of them. Thus, reducing the concentration of these elements to values below the regulatory levels in most cases. Similarly, the retention rate of other uncommon PTE such as In, Sc, Sn, Th, Tl, V and Y had also been outstanding, almost 100% in all inorganic wastes and VC. Furthermore, there were other less significant PTE for which the variability in retention rate is very high, such as Ba, Be, Bi, Co, Mn, Mo, Ni, and U. Among the inorganic wastes, dry sludge rich in iron oxyhydroxides (IO) had the highest capacity to retain PTE, followed by wastes with a high calcium carbonate content (MS: dry marble sludge, CW: carbonated waste from a peat exploitation). The gypsum spoil (GS) was not effective for Ba, Co, Mn, Mo, and Ni retention, but for other PTE it was as affective as the other inorganic wastes. On the other hand, most organic wastes demonstrated an overall good removal effectivity for these PTE, although lower than for inorganic wastes with the exception of VC. The wastes with lowest retention capacity for most PTE were BM and GW.

Discussion
Physical, chemical, and biological characteristics of the waste materials reflect considerable differences in their composition. There are wastes with a strong carbonate character (CW and MS), others that are highly organic (VC, GW, OL, WS, OW, and BM), and also waste with high iron oxyhydroxide content (IO). These characteristics are selected by their important role in the immobilisation of PTE and the acid neutralisation [49][50][51] . For example, organic matter has a high affinity for some PTE because of the presence of ligands or functional groups 52 , in this order: Cu 2+ > Hg 2+ > Cd 2+ > Fe 2+ > Pb 2+ > Ni 2+ > Co 2+ > Mn 2+ > Zn 2+ > As 5+ > As 3+53,54 . Thus, organic matter together with total humic extract and humic and fulvic acids provide an important content of reactive colloidal fractions that allow the complexation of the different chemical forms of PTE 55,56 . Carbonates also exert a strong control over pH, which is considered a key property in controlling the immobilisation of most PTE because of its influence on the electrical charge of colloidal components 57 . In addition, it is a key component to neutralise acid solutions 40 . Likewise, iron oxyhydroxides content is another constituent to consider for the retention of some PTE, especially As, for which they exert a strong control on speciation and bioavailability 58,59 . In fact, the results of AMD treatment test indicate that many of the wastes tested show considerable acid neutralisation and PTE immobilisation capacity. The concentration of most PTE in AMD was very high, exceeding the guideline values established by different legislations for As, Cd, Co, Cr, Cu, Mn, and Zn: (i) environmental quality standard for surface water in Spain 46 ; (ii) legal regime for the reuse of treated water for irrigation in Spain 47 and (iii) guidelines for water reuse in the USA 48  www.nature.com/scientificreports/ also considered relevant in relation to their high concentrations also exceeding these regulatory levels. Other elements like Pb, Sb, and Tl presented potentially concerning concentrations, although their guideline values are not included in the previous references. In addition, most of the PTE in this acid mine drainage were at much higher concentrations than those found in the acidic water discharged in the Aznalcóllar mine accident 42 , as well as the concentrations in AMD generated in metal mines in Australia 60 or in other mining areas around the world 8 . Thus, the results in this study can be extrapolated to most acid mine waters treatment situations around the world; moreover, the use of the wastes tested in this study to treat real AMD worldwide would most likely produce a better quality treated water than that achieved for the artificial AMD used in this study. The treatment of AMD with wastes has been effective in neutralising the acidity in all cases. The pH in treated water increases from pH < 3 to values above 6 and close to neutrality depending on the waste used. In this sense, although the role of carbonates in the neutralisation of acid mine drainage has already been widely demonstrated 61,62 , no statistical correlation was found between pH in leachate (pH (L) ) and the CaCO 3 concentration in the different wastes (Table S3). Nevertheless, carbonates are not the only buffering components controlling pH; there are other constituents in the wastes (e.g., organic matter, exchange bases, Fe and Al oxides, silicates) with relevant influence in this process capacity 63,64 . Likewise, the concentrations of several PTE in AMD after treatment with wastes have been significantly reduced. Indeed, the removal efficiencies of PTE obtained with these wastes have been much higher than those achieved in other studies 8,9,65 . Among the wastes used, inorganic wastes were much more effective in retaining PTE than organic ones. The decreasing order of effectiveness was as follows: IO > CW ≥ MS ≥ VC > GS > OW > OL > WS > GW > BM; where wastes rich in iron oxyhydroxides and carbonates are more effective in the retention of PTE than wastes rich in organic matter. The removal rates for wastes dominated by carbonates (CW and MS) or iron oxyhydroxides (IO) are above 95% for most PTE present in AMD, while for organic wastes the removal rate was below 95% in most cases, with values as low as 15% in the case of bio-stabilised material of municipal solid wastes (BM). In other studies, for similar wastes the removal rates achieved were similar or even lower. For example, water filters partly made of iron-rich materials achieved removal rates of 50% for As 66 . However, other studies that also explore As retention capacity of water filters with iron oxide-rich materials reached rates of 90% 67 and 99% 68 . The latter study concerned not only filters made from iron-rich waste, but also marble slurry filters for which As removal rate is 95% 68 . Furthermore, the success of Table 4. Retention effectiveness of potentially toxic elements (PTE) of the inorganic and organic wastes expressed in %. IO-Dry sludge rich in iron oxyhydroxides, MS-Dry marble sludge, CW-Carbonated waste of a peat exploitation, GS-Gypsum mining spoil, WS-Composted sewage sludge, BM-Bio-stabilised material of municipal solid wastes, VC-Vermicompost from pruning and gardening, OW-Composted solid olive-mill by-product irrigated with drinking water, OL-Composted solid olive-mill by-product irrigated with leachates of the olive-mill, GW-Composted greenhouse plant waste. Letters represent significant differences among different waste materials for a same element (Kruskal-Wallis and Dunn tests, p < 0.05). Very high retention > 95% (bold), High retention > 50% (italic), Low retention < 50% (bold italic), No retention (-).  0.00 b 100.00 ± 0.00 b 100.00 ± 0.00 b 100.00 ± 0.00 b 83.52 ± 2.65 b  16.57 ± 4.36 a 100.00 ± 0.00  www.nature.com/scientificreports/ these materials is not limited to As; for example, along with near 100% As retention in groundwater affected by an abandoned gold mine when treated with various mixtures composed of organic carbon, zero-valent iron and limestone, a strong decrease in the concentration of Al, Cd, Co, Cu and Ni has been demonstrated 69 ; although the concentrations of these elements in the groundwaters are much lower than in our study. On the other hand, although less studied, the capacity of some organic wastes has also been assessed; for example, it has been reported a 70% reduction of some PTE (Al, As, Cd, Cu, Fe, Ni, Mn, Pb, and Zn) present in sulfide mine leachates by the addition of aqueous organic wastes from domestic wastewater 16 . Agricultural wastes have also been used to remove pollutants; for example, solid-olive mill by-products have a great capacity to remove Cr, Mn, Cu, Zn, Ni, and Pb from mining wastewater 70 . Similarly, there is an extensive list of agricultural waste (agave, bananas, wheat, rice, citrus fruits) that have been used for the immobilisation of different PTE (Cd, Pb, Zn) with uncertain results 71 . Particularly noteworthy is the case of vermicompost (VC), which shows retention rates of PTE close to those of carbonated and iron-rich wastes. This may be due to the higher content of calcium carbonate and total iron compared to other organic wastes, and, to a lesser extent, its considerable high OC content. In this sense, vermicompost can be a very effective material for the treatment of AMD. A similar study for the treatment of AMD 72 using vermicompost and other agricultural by-products (sheep, cow, and rabbit manure) reported retention rates of 90% for As, Cd, Cu, and Zn in AMD. Similarly, gypsum spoil (GS) also has a high retention capacity for PTE similar to that of the other inorganic wastes, although for some, such as Ni and Co, was very low. The high retention capacity of GS is related to high CaCO 3 and FeT contents. Equally, it should not be overlooked that the content of PTE in some wastes may pose a potential risk. In relation to the initial concentration of PTE in the wastes, sludge rich in iron oxyhydroxide and gypsum spoil presented slightly high concentrations of As, Pb and Sb. However, they do not exceed the guideline values to declare a soil polluted according to the regional regulations 73 or the maximum levels that a compound must have in order to be used as a fertiliser product in Spain 74 . The rest of the inorganic wastes have low concentrations of most PTE. The same applies to organic wastes, although some of them show high concentrations of certain PTE (Cr, Cu, V and Zn), they do not exceed the guideline values. In particular, the organic wastes with the highest concentrations are compost from greenhouse waste (GW), composted sewage sludge (WS), and bio-stabilised material from municipal solid waste treatment (BM); which are also the wastes with lowest retention capacity. The presence of PTE in waste related to urban activities is common 75,76 , although in our case they do not exceed the guidelines values and, therefore, pose a low risk of PTE pollution. Anyway, concern should be raised about their use due to the very high salinity reflected in their high EC values. In fact, most of the organic wastes except vermicompost cause an increase in EC in the leachates resulting from the treatment with respect to AMD.
The main PTE (As, Cd, Cr, Cu, Pb, and Zn) have been successfully removed (close to 100%) from AMD by waste treatment. Especially inorganic wastes and vermicompost have the highest capacity, leaving the concentrations of most of them in the treated water below the regulatory levels for irrigation and surface water in Spain 46-48 . In contrast, in the treatment with the organic wastes, although significantly reduced the PTE concentrations, the values were above the regulatory levels in most cases. However, the retention of other less studied PTE such as In, Sc, Sn, Th, Tl, V and Y is also remarkable. Promising results are obtained for specific elements, as in the case of V, where previous studies with commercial iron products and a ferric residue from groundwater treatment obtained 85% of removal of this element from mining water 77 , compared to values close to 100% removal in our study for inorganic and vermicompost wastes. Thallium is another highly toxic element and quite understudied 78 ; and the treatment and removal in wastewater is one of the major challenges in the coming years 79 . In our study, the removal rate of Tl in AMD is above 75% for all wastes analysed and for some wastes such as IO, CW, WS and VC above 90%, whereas in other studies included in 78 , the reduction of Tl in wastewaters after treatment with lime is between 21 and 49%. Antimony is also considered a concern element due to the potential toxicity in surface and groundwater; and the use of commercial coagulants such as iron salts have proven to be effective in remediating Sb-polluted waters; in this case, the ferric chloride coagulant presented removal rates higher than 80% across a broad pH range 80 . The efficiency of Sb removal from AMD in our study is higher than 95% for inorganic and vermicompost wastes, which shows the high potential application of the wastes that we have analysed.
Nowadays, the demand of many elements is projected to be high to achieve the energy transition and mining is an essential activity which is reactivating. The production and availability of technology-critical elements is also a current concern. In this scenario, the potential pollution and widespread of PTE into the environment is predicted to rise in the short-term, together with the production of waste related to the different human activities. This study is in line with both problems (increased input of pollutants into the environment and increased production of waste), so the promising results obtained may contribute to the environmental protection and human safety.

Conclusions
This study tests the effectiveness of various wastes as a potential treatment of acid mine drainage to promote mine restoration and environmental protection by the sustainable management of urban, mining, and agro-industrial wastes in a circular economy scenario. Our results conclude that the waste materials studied have a very high acid neutralising capacity, as well as a strong capacity to retain potentially toxic elements. Inorganic wastes, together with vermicompost from pruning and gardening, reduced by more than 95% the concentrations of most PTE in a highly polluted simulated AMD, while organic wastes retain between 50 and 95%. The potential effectiveness followed this order: IO > CW ≥ MS ≥ VC > GS > OW > OL > WS > GW > BM. Thus, a wide range of mining, urban, and agro-industrial wastes could be recovered for use in the treatment of AMD. The use of these wastes as AMD treatment technique showed promising results to be applied in the decontamination of polluted waters and as a control technique on tailing deposits to prevent the AMD generation. This study is the first step in the development of green technologies based on the different combinations of wastes with contrasting characteristics, www.nature.com/scientificreports/ to create solution (e.g.: Technosols, permeable reactive barriers, etc.) with a higher capacity to retain a greater variety of PTE and reduce acidity in polluted environments. The use of waste to remediate AMD will decrease the cost of the water treatment. This is especially relevant for the rehabilitation of areas with historical or abandoned mines, where the decrease in cost by replacing commonly used and expensive reagents for worthless waste will increase the affordability of water treatments. Nevertheless, additional site-specific studies should be conducted to include the cost of waste transport, as well as to evaluate the in-situ effectiveness of waste combinations under real field conditions.