Assessment of Potential Environmental Risks Posed by Soils of a Deactivated Coal Mining Area in Northern Portugal—Impact of Arsenic and Antimony

Abstract

Active and abandoned mining sites are significant sources of heavy metals and metalloid pollution, leading to serious environmental issues. This study assessed the environmental risks posed by potentially toxic elements (PTEs), specifically arsenic (As) and antimony (Sb), in the Technosols (mining residues) of the former Pejão coal mine complex in Northern Portugal, a site impacted by forest wildfires in October 2017 that triggered underground combustion within the waste heaps. Our methodology involved determining the “pseudo-total” concentrations of As and Sb in the collected heap samples using microwave digestion with aqua regia (ISO 12914), followed by analysis using hydride generation-atomic absorption spectroscopy (HG-AAS). The concentrations of As an Sb ranging from 31.0 to 68.6 mg kg⁻¹ and 4.8 to 8.3 mg kg⁻¹, respectively, were found to be above the European background values reported in project FOREGS (11.6 mg kg⁻¹ for As and 1.04 mg kg⁻¹ for Sb) and Portuguese Environment Agency (APA) reference values for agricultural soils (11 mg kg⁻¹ for As and 7.5 mg kg⁻¹ for Sb), indicating significant enrichment of these PTEs. Based on average I_geo values, As contamination overall was classified as “unpolluted to moderately polluted” while Sb contamination was classified as “moderately polluted” in the waste pile samples and “unpolluted to moderately polluted” in the downhill soil samples. However, total PTE content alone is insufficient for a comprehensive environmental risk assessment. Therefore, further studies on As and Sb fractionation and speciation were conducted using the Shiowatana sequential extraction procedure (SEP). The results showed that As and Sb levels in the more mobile fractions were not significant. This suggests that the enrichment in the burned (BCW) and unburned (UCW) coal waste areas of the mine is likely due to the stockpiling of lithic fragments, primarily coals hosting arsenian pyrites and stibnite which largely traps these elements within its crystalline structure. The observed enrichment in downhill soils (DS) is attributed to mechanical weathering, rock fragment erosion, and transport processes. Given the strong association of these elements with solid phases, the risk of leaching into surface waters and aquifers is considered low. This work underscores the importance of a holistic approach to environmental risk assessment at former mining sites, contributing to the development of sustainable remediation strategies for long-term environmental protection.

1. Introduction

Active and abandoned mining sites are the primary anthropogenic sources of heavy metals in the environment [1,2,3,4]. Mining waste often contains elevated concentrations of heavy metals [5,6] that can be readily mobilized due to acid mine drainage, contaminating surrounding soils and groundwater, ultimately disrupting ecosystems and posing risks to living organisms.

Total metal concentrations in soil, while indicative of both the geological origin of soils and anthropogenic contributions, are poor predictors of element mobility or bioavailability, making selective extraction methods a very useful approach to studying diffuse metal contamination in soils [4,7,8,9,10].

A sequential extraction procedure (SEP) method consists of subjecting a sample to a consecutive sequence of reagents with increasing strength to extract specific species of the desired analyte in each extraction step. Since the required reagent strength increases through the steps, the PTE species extracted in the early steps of the procedure (easily extracted from the solid phase due to weak interactions) are the most concerning due to being the more labile species [4,7,8,9,10]. Many SEP methods have been implemented for the speciation of heavy metals. For example, the modified three-step procedure proposed by the Community Bureau of Reference (BCR) method has been commonly used to assess the potential environmental contamination of soils with Cu, Cd, Zn, Pb, etc. [11,12,13].

Relatively to the target potentially toxic elements (PTEs) analysis, although atomic absorption spectroscopy (AAS) is commonly used for the determination of several metals (Na⁺, Ca⁺, K⁺, etc.), for other metals and metalloids, more sensitive techniques such as hydride generation (HG-AAS) and cold vapor atomic absorption spectroscopy (CV-AAS) are frequently used [4,14,15,16,17,18]. Some elements, such as arsenic (As), antimony (Sb), and selenium (Se), absorb radiation with wavelengths below 200 nm. In this range, the absorption of radiation by the flame gases is substantial and, thus, complicates the correct analysis of such elements. An alternative approach is then to take advantage of the properties of some elements to form volatile hydrides [14,18,19,20].

The Fojo waste pile of the deactivated Pejão coal mine complex, located at Castelo de Paiva municipality, was identified as potentially harmful to surrounding soils and aquifers [4,21,22]. In October 2017, forest wildfires triggered underground combustion within the waste heaps, further exacerbating potential environmental hazards due to technical intervention in the mine area (see Figure 1a). The waste pile was originally composed of darker mining residues, that comprised the lithologic remaining coal and carbonous shists (Figure 1b), and highlighted accessory minerals, such as Chlorite, Siderite, and Pyrite [22]. The self-combustion process changed the mineralogic composition of the residues and resulted in red-tone residues highly enriched in iron oxides and sulfates such as Hematite, Jarosite, and Alunite [22] (Figure 1c).

Previous studies performed by our research team in the Fojo mine region indicated significant mercury (Hg) enrichment in soil samples compared to Portuguese and international reference values for soils [4]. Regarding Hg availability and mobility, while it predominantly resides in the semi-mobile fraction, materials from the waste pile exposed to combustion showed a worrisome increase in Hg levels within the mobile fraction. It is important to highlight that this fraction contains the more labile and bioavailable Hg species, making it a substantial source of environmental contamination [4]. Thus, based on this finding, we have decided to further explore potential environmental threats imposed by other PTEs.

This work focused on determining arsenic (As) and antimony (Sb) content in soils from the Fojo mine area since post-mining tailings and waste rock, when exposed to surface conditions, promote their leaching from minerals by precipitation/dissolution and adsorption/desorption processes (strongly governed by metalloid oxidation state and form, and medium alkalinity and salinity) [23,24,25], resulting in environmental contamination that can severely impact the adjacent ecosystems and even the human health. Arsenic and antimony are recognized pollutants, considered of priority interest by the USEPA [26], with well-characterized mechanisms of action and established toxicological profiles. For example, As, particularly in its inorganic forms, is a known human carcinogen, has a high degree of genotoxicity and cytotoxicity [27], being also toxic to both plants and living organisms [28]. Moreover, Sb can easily be accumulated in organs and tissues, inhibits enzymatic activity, disrupts cellular ion homeostasis, and interferes with protein and carbohydrate metabolism, ultimately leading to damage in the heart, lungs, liver, kidneys, and nervous system [23,24].

In this work, the potential environmental impact of As and Sb was investigated by first digesting sub-samples of the collected soil (see Figure 1b,c) through a microwave-assisted procedure using aqua regia to assess the total PTE content. Furthermore, the Shiowatana SEP was used to evaluate metal fractionation and mobility and get a more realistic perspective of the environmental risk posed by As and Sb in the mining area. The As and Sb content in sample extracts were analyzed by HG-AAS.

2. Experimental Part

2.1. Soil Sampling

A total of 25 surface soil samples (0–20 cm) were collected from the waste pile and surrounding soils (see Figure 2). From the 25 total samples, 10 were collected from the burned coal waste (BCW samples; Technosol), 5 from the unburned coal waste (UCW samples; Technosol), 5 from the soil located downhill from the coal waste (downhill soil—DS samples; Regosol), to study the spread of the contamination, and 5 from the wood located uphill from the coal waste (uphill soil—US samples; Regosol). Each sample was collected with a stainless-steel shovel, weighed approximately 1.5 kg, and was stored in a polyethylene bag.

The samples were air-dried at room temperature for several days. They were then sieved through a 2 mm mesh to remove inorganic and organic coarse elements. Subsampling was conducted using the coning and quartering method to ensure homogeneity.

2.2. Chemicals, Solvents and Instruments

The chemicals and solvents used in this work were: sodium borohydride (99%, ReagentPlus, Sigma-Aldrich, Taufkirchen, Germany), sodium hydroxide (≥98%, Analytical grade, Sigma-Aldrich, Taufkirchen, Germany), ethanol (≥99.8%, HPLC grade, Fisher-chemical, Dreieich, Germany), hydrochloric acid (37% (w/v), Laboratory grade, Fisher-chemical, Dreieich, Germany), nitric acid (69% (w/v), ACS grade, Fisher-chemical, Dreieich, Germany) and As and Sb standard solutions (Certified solution, Thermo Scientific, Dreieich, Germany).

All reagents and solvents used were of analytical grade adequate to accurate trace metal determination by HG-AAS technique and were used without further purification. All aqueous solutions used in this work for soil digestion and HG-AAS analysis were freshly prepared prior to use to guarantee the best reaction performance. Water purified with a Milli-Q purification system (resistivity ≥ 18 MΩ·cm) was used.

The instruments used in this work were: a Pro-Analytical centrifuge (Centurion Scientific), a Vortex stirrer (LBX instruments; V05 series), an Analytical balance (A&D; GR-202), a Hot plate magnetic stirrer (IKA; C-MAG HS 7) and an Ultrasonic bath (Bandelin Sonorex; Digitec DT 100 H).

2.3. Soil Sample Digestion with Aqua Regia

For each collected sample, the aqua regia soluble As and Sb were determined by a microwave-assisted digestion procedure, based on protocols described in ISO 12914 [29] and USEPA 3051A [30]. The microwave dissolution program consisted of maintaining the temperature of 175 °C inside vessels for around 20 min.

To perform soil digestion, only a small sample was used to minimize the formation of gases and avoid the rupture of the safety membrane due to excessive pressure inside the vessel. Therefore, about 0.25 g of homogenized soil sub-samples weighed to 100 mL PTFE vessels and placed in a fume hood. Then, 6.0 mL of concentrated HCl and 2.0 mL of concentrated HNO₃ were added to the pre-cleaned digestion vessels.

The vessels were placed inside the microwave digestion rotor, sealed tight using the multifunctional lid of the system, which served as a rack for up to 6 vessels and protection shield during the reaction, and finally submitted to the microwave dissolution in an Anton Paar MULTIWAVE 1000W microwave oven. After digestion and cooling of the samples, the final solutions were filtered through Whatman filter papers, transferred into 25 mL or 50 mL volumetric flasks, and the volume was made up with pure water prior to analysis by HG-AAS.

2.4. Arsenic and Antimony Analysis by HG-AAS

The analysis of certain metals and metalloids, such as As and Sb, using traditional flame AAS can be challenging due to interference from the flame gas absorption. Coupling AAS with hydride generation (HG) minimizes these limitations while reducing interference from the sample matrix (by isolating the analyte from the sample), therefore enhancing sensitivity [18,31]. The HG-AAS method also benefits from easy automation, mild atomization conditions, and simple analyte preconcentration [18,31].

HG-AAS analysis was performed using a Thermo Scientific iCE 3000 Series double-beam Atomic Absorption Spectrometer combined with the VP100 accessory, a continuous flow vapor generation (hydride) system. The AAS system and the VP100 were both fully controlled by Thermo Scientific software SOLAAR.

For the metalloid measurements, the standard “T” cell was used for hydride analysis. Hydrides were generated by the reaction of the acidified samples with sodium borohydride (NaBH4) [14]. The formed hydrides were carried out by an N₂ stream to the absorption cell mounted in the burner for atomization and the absorption was measured. For hydride generation at the VP100 continuous flow vapor generator system, solutions of diluted hydrochloric acid (10%, w/v) and sodium borohydride (1%, w/v; stabilized in 0.5% (w/v) sodium hydroxide solution), as reducing agent, were selected. The HG-AAS/VP100 operating parameters adopted for the measurements can be consulted in

Table 1. Operating conditions of the HG-AAS system.

AAS and VP100 Operating Conditions
Slit width	0.5 nm
Burner height	15 mm
Hollow lamp wavelength (As)	217.6 nm
Hollow lamp wavelength (Sb)	193.7 nm
Lamp current/voltage	75%/6 V
Signal measurement	Transient Height
Signal type	Background correction
Total measurement time	100 s
Sample aspiration time	25 s
Carrier gas flow	100 mL·min−1
Pump speed	40 rpm
Sample uptake rate	7.5 mL·min−1
Reducing agent uptake rate	1.6 mL·min−1
Acid uptake rate	0.7 mL·min−1

.

2.5. Arsenic and Antimony Determination

The calibration curve method was used to quantify As and Sb in sample extracts. To build the calibration curve, standard solutions of 3.00, 8.00, 14.0, 20.0, and 25.0 µg L⁻¹ were prepared from a stock solution (C = 1004 ± 7 mg L⁻¹) by adding to different volumetric flasks appropriate aliquots of a 2.00 mg L⁻¹ intermediate solution and making up the volumes with 10% w/v HCl.

In the HG-AAS technique, the signal was continuously measured for 100: during the first 10 s, no sample was aspirated; from 10 s to 35 s, the sample was aspirated (consuming ~4 mL of the sample) and; from 35 s to the end of the measurement, no sample was aspirated. After each measurement, a 10 % (w/v) hydrochloric acid solution was aspirated to clean the sampling tube and the cell, thus minimizing “memory effects” between measurements.

2.6. Sequential Extraction Procedure for As and Sb

In this work, the four-step procedure described by Shiowatana et al. [32,33,34] was used for arsenic and antimony fractioning in mine soils. Comparative studies [32,33] using distinct methods for arsenic fractionation, namely the Tessier SEP, the Rauret SEP, the Wenzel SEP, the Larios SEP, and the Shiowatana SEP, revealed the best performance for the latest two methods; however, the Shiowatana SEP demonstrated greater effectiveness in extracting potentially bioavailable arsenic fractions and in identifying their sources, suggesting its superior suitability for risk assessment of arsenic-contaminated soils. Consequently, the Shiowatana SEP was selected for this study, not only for As fractioning but also to assess Sb mobility in the same soil samples. Both metalloids belong to Group 15 of the periodic table, having chemical similarities, such as being frequently encountered with the same oxidation states in the environment (+3 to +5) and both usually occur as oxides, hydroxides, or oxoanions [25,27]. Similarly to this work, many studies reported in the literature performed parallel As and Sb analysis in environmental soil samples [26,35,36]. The details of the experimental procedure are given in

Table 2. Extraction procedure developed by Shiowatana et al. and used in this work for As and Sb fractioning.

Step	Defined Fraction	Extracting Solution	Experimental Conditions
1	Water soluble	30 mL of ultrapure water	Shake for 16 h at 25 °C
2	Surface adsorbed	30 mL of NaHCO3 0.5 M (pH = 9)	Shake for 16 h at 25 °C
3	Associated with Fe/Al	30 mL of NaOH 0.1 M (pH = 13)	Shake for 16 h at 25 °C
4	Acid extractable (a)	30 mL of HCl 1 M	Shake for 16 h at 25 °C

^(a) Associated with carbonates.

.

3. Results and Discussion

3.1. Arsenic and Antimony Quantification by HG-AAS

The main advantage of HG-AAS relative to flame AAS is the higher sensitivity due to the longer residence time of the analyte in the absorption cell and the fact that flame gas absorption does not interfere with the measured signal.

In this technique, hydrides are usually generated by reacting the acidified sample with sodium borohydride (NaBH₄) [14], as shown in the representative equation below for As(III):

$${As}_{ \left(aq\right)}^{3+}+{{6BH}_4^{-}}_{ \left(aq\right)}+{3H}_{ \left(aq\right)}^{+}\to {{3B}_2{H}_6}_{ \left(aq\right)}+{{3H}_2}_{\left(g\right)}\uparrow +{{AsH}_3}_{\left(g\right)}\uparrow$$

(1)

The formed hydride is carried out by an inert gas (i.e., N₂) stream to the absorption cell mounted in the burner where it will be atomized and the absorption measured.

The obtained graphic profile of absorbance versus time had a peak shape (see Figure S1). The peak maximum corresponds to the higher quantity of volatile species accumulated within the absorption cell and was used analytical signal to build the daily calibration curves, after measuring five standard solutions with concentrations in the range from 3.00 to 25.0 µg L⁻¹. The representative plots of the measured peak height versus the corresponding concentration are shown in Figure 3 for As (Figure 3a) and Sb (Figure 3b).

Considering the calibration curves employed for the metalloid trace analysis, the limits of detection (LOD) and quantification (LOQ) and the sensitivity of the HG-AAS method were estimated and depicted in

Table 3. Analytical parameters estimated from the representative calibration curves.

	Arsenic		Antimony
Analytical Parameter	Statistical Parameter	Value (µg mL−1)	Statistical Parameter	Value (µg mL−1)
LOD	Sy/x = 0.003	0.50	Sy/x = 0.003	0.68
LOQ		2.8		2.1
Sensitivity	Abs (8 µg L−1) = 0.076	0.50	Abs (8 µg L−1) = 0.076	0.20

. The sensitivity was considered as the concentration that gives an absorbance of 0.0044, therefore, it was easily estimated by analyzing a sample that returned a signal of about 0.1. The LOD and LOQ were estimated as the corresponding concentration for a signal of 3 and 10 times the standard deviation of the residuals (S_y/x) [37], respectively.

3.2. Aqua Regia Extractable Content of As and Sb in Soil Samples

Various acids and mixtures can be used for microwave digestion of soil samples. Still, due to the hazardous potential of other acid mixtures, a microwave-assisted digestion procedure using aqua regia was adopted in this work. This method is less aggressive than other procedures that use mixtures of stronger (and extremely hazardous) acids, such as HF and HClO₄, and the obtained extracts are commonly considered to represent “pseudo-total” concentrations of the PTE in the sample. Nonetheless, recent studies reported that digestion procedures with aqua regia could yield, for several metals, statistically similar performance to procedures using more aggressive acid mixtures [38,39].

In this work, the aqua regia soluble As and Sb in soil samples were first determined to assess the contamination of the mine region by these elements considering the worst possible scenario, that all the species in the soil would be bioavailable for contamination of surrounding soils, aquifers, and ecosystems. After the soil samples digestion, the As and Sb contents in the extracts were determined by HG-AAS. Table S1 (SI) compiles the pseudo-total concentration of As e Sb in all samples collected from the deactivated Pejão coal mining complex. For clarity, the data obtained is also graphically represented in Figure 4a and Figure 4b for As and Sb, respectively.

The metalloid concentration levels in the analyzed soil samples, with values ranging from 10 to 69 mg kg⁻¹ and 0.7 to 8.3 mg kg⁻¹ for As and Sb, respectively, are above the European background values reported in project FOREGS (11.6 mg kg⁻¹ for As and 1.04 mg kg⁻¹ for Sb), with exception of sample US1 for As and samples US1-3 for Sb, showing a significant enrichment of these PTEs, thus being a potential source of contamination in the waste pile area.

Arsenic concentrations obtained in this study are compatible with the ones previously reported for Pejão coal waste piles [40]. Furthermore, comparing to reference values provided by the Portuguese Environment Agency (APA) for agricultural soils, it is important to notice that (with the exception of sample US1) all samples showed concentration values for As significantly above the threshold value (11 mg kg⁻¹), including the samples from uphill soils (without influence from the waste pile). This observation is in line with previously reported research work reporting that the As enrichment is related to arsenian-rich pyrites in the coal from the Douro Carboniferous Basin (DCB) and its respective mining wastes [41]. DCB coals have shown As enrichment when compared with world hard coals. Arsenic showed affinities with Fe, S, and Sb [40,41], possibly related to sulfide enrichment from the contiguous Au-Sb and Ag, Pb, and Zn mineralizations along the Dúrico-Beirão Belt [21]. Regarding Sb, samples BCW1, BCW4, BCW9, and BCW10, taken in the waste pile area that was affected by combustion, revealed values either very close to or above the reference value (7.5 mg kg⁻¹).

The geoaccumulation index (I_geo) [42] values were estimated (see Tables S2–S4, SI) to categorize the extent of the PTE contamination in the collected soil samples from the studied mine regions (BCW, UCW, and DS). Based on the I_geo classification chart, and considering the average I_geo values, the three mine regions were classified as “unpolluted to moderately polluted” relative to As contamination. For Sb, samples from the waste pile regions (both BCW and UCW) were classified as “moderately polluted” while samples from the downhill soil (DS) were classified as “unpolluted to moderately polluted”.

A more concise analysis of the mean pseudo-total concentration of these metalloids in the burned (BCW1-10; As = 43 mg kg⁻¹; Sb = 6.2 mg kg⁻¹) and unburned (UCW1-5, As = 41 mg kg⁻¹; Sb = 4.1 mg kg⁻¹) waste piles revealed a statistically significant (Student t-test, p < 0.05) enrichment of 240% and 227% for As and of 582% and 391% for Sb, respectively, in relation to the reference samples (US1-5; As = 18 mg kg⁻¹; Sb = 1.1 mg kg⁻¹), highlighting the contribution of anthropogenic activities for the contamination of the soil.

Arsenic and antimony can occur in coal mining residues, particularly associated with sulfides, adsorbed and in organically bound forms. In fact, coal combustion is one of the major atmospheric emission sources of these elements around the world [43,44]. The combustion process can contribute to the enrichment of more mobile species in mine soil samples due to carbon consumption in the wastes, concentrating the inorganic fractions. At the same time, during coal combustion, the volatilization of organically bound arsenic predominates at temperatures below 600 °C. Between 800 and 900 °C, arsenic bound in sulfide forms undergoes decomposition or oxidation, resulting in its release. At temperatures exceeding 1000 °C, the decomposition of arsenate compounds may occur [45,46]. The resulting fly ashes particles can be enriched in As and Sb which may vaporize during combustion and, in some instances, recondense and/or adsorb on ash surfaces, further contributing to the enrichment of these PTEs [46].

Moreover, the mean values found in the downstream soil samples (DS1-4; As = 40 mg kg⁻¹; Sb = 2.2 mg kg⁻¹) were also significantly higher (Student t-test, p < 0.05), 224% and 205% for As and Sb. This is in line with the results presented by Espinha Marques et al. [22] who observed a clear enrichment in PTEs of the DS soil resulting from the particle transport due to coal waste erosion, as well as from the input of dissolved chemical components resulting from acidic interflow in the burned waste pile.

Therefore, considering that the As and Sb aqua regia extractable content in mine samples, fractioning and speciation of both PTEs, using sequential extraction procedures (SEP), is highly recommended to assess their mobility, which can be regarded as potential bioavailable for surrounding soils and aquifers.

3.3. Fractionating and Mobility of Arsenic and Antimony

A four-step procedure described by Shiowatana et al. [32,33,34] was reported to be most adequate for arsenic (As) fractioning in soils and sediments. The method can be extended to study antimony (Sb) mobility due to similar reactivity to As species [25,26,27,35,36] and allows the classification of the species in soils into mobile and stable fractions. Mobile As and Sb species are those that, subjected to specific environmental conditions, can be converted to a free state (or that already exist freely in soils). In this context, the fractions correlated with potentially bioavailable arsenic (high potential for contamination) are those that do not rely on the destruction of the crystalline network of the soil matrix, i.e., fractions 1 (soluble in water), 2 (adsorbed to the surface) and 4 (extracted by acid/associated with carbonates). The remaining species extracted in step 3 (associated with Fe and Al oxides and hydroxides) are classified as stable fractions and therefore they contain less mobile and potentially bioavailable species.

In this work, this SEP procedure was applied to samples from the mine affected by combustion (BCW samples) since it is the area with the highest potential for harmful environmental and health impacts. Data obtained from the application of the SEP in the soil samples (summarized in Tables S5 and S6) are represented graphically in Figure 5a,b for As and Sb, respectively. A clear shortfall in the sum of the PTEs in the SEP 4-fractions compared to the obtained by microwave-assisted digestion with aqua regia was observed. This may indicate that most As and Sb species are strongly incorporated into the soil matrix and may be only released as residual under harsh conditions (e.g., aqua regia extraction).

This also agrees with the fractioning data obtained for As. The average concentration found for the more mobile and bioavailable species (sum of fractions 1, 2, and 4), of 2.8 ± 0.3 mg kg⁻¹, ranging from 1.2 ± 0.5 to 5.1 ± 0.5 mg kg⁻¹, was well below the APA’s reference value (of 11 mg kg⁻¹, for agriculture use soil). Furthermore, for all samples, the most dominant fraction was fraction 3, with an average concentration value of 10 ± 1 mg kg⁻¹ (ranging from 7 ± 1 to 12 ± 1 mg kg⁻¹), thus of lower mobility since it contains the more stable arsenic species. Still, samples BCW2, BCW3, BCW5, BCW6, BCW8, and BCW9 presented concentration values higher or very close to APA’s reference value, emphasizing the need for periodic monitoring the levels of As in the mine soil as over time species found in the less bioavailable fraction may convert into more bioavailable forms.

Relatively to Sb, the concentration levels of all factions are far below the APA’s reference value (of 7.5 mg kg⁻¹) and even the sum of fractions 1, 2, and 4 was only 0.94 ± 0.05 mg kg⁻¹ (ranging from 0.56 to 1.12 ± 0.04 mg kg⁻¹), thus below the established threshold.

Overall, the soil enrichment with (not readily mobile) As and Sb species may be primarily due to the significant amount of mining wastes hosting sulfides, including pyrite and stibnite (BCW and UCW) deposited in the waste piles. Furthermore, the results indicate a low leaching mobility of As and Sb present in the BCW area, raising the strong possibility that the higher (~2-fold) DS:US concentration ratios were mainly due to non-leaching processes such as mechanical and physical erosion of the wastes followed by particle transportation by gravity or aerial spreading. The high content of As present in the waste pile rocks seems to be preserved and retained within mineral crystalline forms and therefore does not present significant mobility (nor bioavailability).

4. Conclusions

In this work, the total amount of As an Sb in the Fojo coal waste pile area was estimated by determining the aqua regia soluble content using a microwave-assisted digestion method, aiming to assess the scale of soil contamination with these PTEs. The pseudo-concentration values obtained revealed enrichment of As and Sb in mine waste soil samples relative to the reported background values for Portugal in project FOREGS. Furthermore, based on the calculated average I_geo values, As contamination in soil samples was classified as “unpolluted to moderately polluted” while Sb levels were classified as “moderately polluted” for the waste pile samples and “unpolluted to moderately polluted” in downhill soils (DS).

However, PTEs’ bioavailability and toxicity to organisms are strongly influenced by the chemical forms in which they are found in the environment (speciation). Thus, a four-step sequential extraction procedure was applied to the mine waste soil samples to fractionate As and Sb into mobile and stable fractions and evaluate the PTEs’ potential bioavailability to the surrounding soils and aquifers. The results indicate that As and Sb levels in the more mobile fractions were not significant, suggesting that PTE enrichment in the burned (BCW) and unburned (UCW) waste areas of the mine is likely due to the stockpiling of lithic fragments rich in sulfides, and these elements may be trapped in the crystalline structure of primary minerals such as arsenian pyrites and stibnite and/or secondary minerals, whose subsequent erosion and transport processes have contributed to the increased As and Sb concentrations in DS samples.

These less mobile fractions present a lower risk to surficial water courses and aquifers, as well as a low plant absorption. Nonetheless, the arsenic concentration in the less mobile fraction was close to or even higher than APA’s threshold, indicating that remediation may be recommended to avoid the conversion of As species into more bioavailable ones if the geochemical conditions change. To further address this problem, the research team is currently developing new sensing technologies for the in situ detection of PTEs [47,48,49,50] and the mitigation of their effects using sustainable remediation strategies [51].

This research highlights the critical need for continued vigilance and proactive measures at deactivated mining sites to safeguard environmental health and ensure the well-being of affected communities. Understanding the complex behavior and mobility of PTEs in these environments is paramount for mitigating potential impacts on delicate ecosystems, water resources, and human health. Our findings underscore the importance of robust environmental monitoring programs and the implementation of innovative, sustainable remediation strategies. By focusing on sustainable solutions, we can effectively manage and prevent the long-term spread of contamination, contributing to the ecological integrity and overall resilience of regions impacted by historical mining activities.