Geochemical and Advanced Electron Microscopical Characterisations of Artisanal Gold Mining Rejects in Colombia

Abstract

Artisanal gold mining causes widespread health problems due to illegal exposure to hazardous inorganic compounds, such as arsenic (As) and mercury (Hg). The sources and prevalence of mining pollution are strongly influenced by topography, stream dynamics, soil type, and land use. In the present study, the potential hazardous elements (PHEs), absorption abilities of nanoparticles (NPs), and ultrafine particles (UFPs) were analysed from clandestine gold mining soils in Colombia. The proportions of PHEs including As, Hg, Cu, Cr, and Pb in carbonates, sulfides, clays, oxides, hydroxides, and sulfates were determined by field-emission scanning electron microscopy (FE-SEM), high-resolution transmission electron microscopy (HR-TEM), and selected area electron diffraction (SAED)/micro-beam diffraction (MBD)/energy dispersive X-ray spectroscopy (EDS). The results revealed that the concentrations of As, Hg, and Zn were significantly higher in clay particles when compared to the other soil samples. Furthermore, Al and Fe manifested excellent PHEs sorption abilities in the artisanal gold mining soils. The results presented will be useful for future mitigation measures in the gold mining areas.

1. Introduction

In the current scenario, nearly 15% of the world’s gold supply is sourced from the small-scale mines operated by 10–15 million people (or 90% of the total gold mining workforce worldwide) that annually work in the artisanal mine sectors. Likewise, these workers indirectly support over 100 million people by injecting cash into rural economies [1]. However, the poor mining practices in these regions create severe health, safety, and environmental problems due to the long-term exposure to Hg [2]. In recent years, the use of Hg and its harmful effects have attracted significant attention worldwide.

In the year 2017, the Minamata Convention on Mercury was established to protect humans and the environment from the harmful effects of Hg [3,4]). The Mercury Convention is an international treaty that promotes region-specific plans to eliminate unsafe practices in artisanal gold mining areas among its 140 signatory countries [5,6]. At many small-scale mining sites, illegal practices, including Hg amalgamation and open burning, are often observed [4,6]. The Minamata Convention has branded these inefficient techniques of gold extraction as “worst practices”. In comparison to the refined and modern techniques, these methods recover much less gold, cause serious health problems, and release large amounts of toxic materials into the environment ([3,5]. Therefore, artisanal gold mining has become one of the largest sources of global environmental pollution [7,8].

Furthermore, Colombian workers occasionally add cyanide to Hg amalgam waste to extract more gold, which exacerbates the risk of human health, safety, and environmental problems. The geological information of the targeted area can be evaluated in the previous study [1]. In addition, some regions have experienced “physical” disturbances, such as clearing and the excavation of mountain slopes. Other notable practices include open-pit, and underground mining, the construction of small processing plants and unregulated housing or structures. Lastly, the land, water bodies, and food sources in other mining areas have been chemically contaminated due to mining and mineral processing.

In many parts of Colombia, the guerrilla group National Liberation Army (Ejército de Liberación Nacional—ELN) and former members of the now-defunct Revolutionary Armed Forces of Colombia (Fuerzas Armadas Revolucionarias de Colombia—FARC), struggle to control illicit mining and trade in gold [9]. Furthermore, the significant improvements in road infrastructure have exacerbated the pollution levels experienced by people in the mining regions due to the increased traffic related to the mining by FARC. Moreover, the contamination of drinking water and atmosphere has soared, increasing the health, safety, and environmental concerns to the local people.

Despite Colombia’s efforts to control the illicit gold trade, approximately 80% of the precious metals extracted in the country are reportedly mined illegally or without the necessary environmental precautions. This scenario is exacerbated by poverty, unemployment, and unstable income, which have driven the local people toward illegal mining practices. Furthermore, the unregulated use of mercury (Hg) and cyanide (CN) to extract gold from the minerals has poisoned the waters and animals [10,11]. Consequently, some communities along the rivers in northern Colombia have stopped eating fish for fear of mercury poisoning [12]. Furthermore, substantial evidence of high concentrations of potentially hazardous elements (PHEs), such as Hg, As, Cd, Cu, Pb, Zn, Cr, and Sb, have been reported in the atmosphere, soil sediments, water, and biota in the vicinity of the mining areas [13,14]. Therefore, the elucidation of the mode of occurrence of PHEs in the mining environment is highly essential [15].

The absorption capacities of soils in mining areas depend on several factors, such as soil pH values and the presence of absorption sites in the soil components [16]. Typically, the proportion of clay minerals, hydroxides of Fe and Al, carbonates, and organic matter in soils control the absorption of PHEs [17]. The rising global demand for gold and other precious metals used as components for electronic devices has triggered numerous socio-economic, geo-political, and environmental challenges in Colombia [9]. Most notably, the incidence of deforestation, human trafficking, population displacement, and forced child labour have become rampant in remote locations, jungles, and mining communities [18,19].

In this context, Olivero-Verbel et al. [12] have examined the relationship between nanominerals and PHEs to develop sustainable practices for gold mining. Therefore, several analytical methods, such as time-of-flight secondary ion mass spectrometry (TOF-SIMS) and field-emission scanning electron microscopy (FE-SEM), have been used to understand the interactions between nanominerals and PHEs [20,21]. However, these procedures do not provide direct information about the nanomineralogy of the targeted compounds. Therefore, the present study presents a detailed investigation of the PHE adsorption abilities of nano- and ultra-fine minerals in the artisanal gold mining areas of south west Colombia.

2. Materials and Methods

2.1. Raw Samples

A total of thirty-nine (39) raw samples of gold cleaning rejects (GCRs) were received from the region with the geographical coordinates (6.138563556732871 N, −75.80905472476464 W) (Figure 1), similarly reported by Sánchez-Peña et al. [1]. However, the samples examined in this study were contaminated with mercury (Hg) and cyanide (Figure 1). Before the comprehensive characterisation, the collected samples were completely homogenised by the cone and quartering technique to obtain uniform and representative mixes. Next, precisely 100 g of samples were retrieved for the analytical tests described in Section 2.2, whereas the remaining fraction of the sample was stored for protection from light and moisture based on the guidelines reported in the literature [20].

2.2. Analytical Methods

The geochemical configurations of the collected GCRs were analysed using X-ray diffraction (XRD), field-emission scanning electron microscopy (FE-SEM), high-resolution transmission electron microscopy (HR-TEM; 200 kV) equipped with an efficient FE cathode and an energy omega-filter, and selected area electron diffraction (SAED) [22,23]. Furthermore, the fast Fourier transform (FFT) technique, scanning transmission electron microscopy (STEM), and energy-dispersive X-ray spectroscopy (EDS; Oxford Instruments INCA 4.09 software) were employed for comprehensive analyses of the samples [24]. The samples were examined by FE-SEM fitted with EDS under secondary electron (SE) and backscattered electron (BSE) modes [25,26,27]. The SEM was operated at 15 kV accelerating voltage and 0.1 nA beam current. The EDS is a semi-quantitative apparatus typically employed to detect the chemical configurations of sampled compounds at 0.1 wt.% detection limit. The EDS point-mode was carried out at less than 15 kV voltage, 2 nA beam current, and 10-mm working distance. After steadying the energy spectrums, each elemental atomic and weight percentage was converted by X-ray counts. Similar SEM procedures have been described in the literature [16,28,29] The elemental configurations of the NPs were examined by HR-TEM joined with EDS and FFT [30,31]. The NPs were assessed by 200 keV HR-TEM (Model: JEOL JEM-2010F, Japan) coupled with EDS (Model: Quantax 200, JEOL, Japan). The samples were disconnected and melted in some organic solvents using ultrasonic-sound suspension before being pipetted into Lacy Carbon films supported by Cu grids [25,32]. To avoid contamination, the HR-TEM sample holder was prepared using an advanced plasma system (Model: Gatan Model 950). The HR-TEM point and lattice resolutions were 0.194 nm and 0.1 nm, respectively [28,29].

The elements with higher atomic numbers were detected from the brightness zones of the images, whereas the elements with lower atomic numbers were observed in the dark-field zones of the identified images as described in the literature [33]. Based on the elemental composition, the potential nanominerals were selected from the Inorganic Compound Powder Diffraction File (PDF) database from the International Centre for Diffraction Data. These selected PDFs of the minerals were subsequently compared to the FFT patterns data (interplanar spacing and angles) to calibrate nanominerals [20,34].

As described in Cerqueira et al. [20] and [35], nanominerals typically present some crystal defects, such as added disordering, shearing, or impurities. Hence, there are minor differences between the tested nanominerals and standard minerals in the PDF database [20,21]. The diffraction patterns of unknown nanominerals usually embrace polycrystalline and monocrystalline constituents [34]. The diffraction pattern indexes, including interplanar spacing (d) and angle of crystal plane (<D₁, D₂>) were measured by Digital Micrograph software. By comparing the standard data (from the PDF database) with the measured values, the final identified mineral must meet these restrictions: |d_m–d_s| < 0.01 nm (d_s: standard d value, d_m: measured d value); |<D_m1, D_m2>–<D_s1, D_s2> | < 3º (<D_s1, D_s2>: standard angle value; <D_m1, D_m2>: measured angle value); the indices of crystal face (hkl) of d_s1, d_s2, d_s3 must conform with the equation: h₁k₁l₁ + h₂k₂l₂ = h₃k₃l₃ (h₁k₁l₁, h₂k₂l₂, and h₃k₃l₃ are the crystal faces of d_s1, d_s2, and d_s3, respectively). In addition, the <D_s1, D_s2> is calculated by cell parameters (a, b, c, α, β, γ) and the two identified (h₁k₁l₁) and (h₂k₂l₂). The terms a, b, c describe the three groups of edge lengths, whereas α, β, γ are the intersecting angles.

3. Results

The samples examined were mainly composed of diabase and basalt repositories of a greenish-grey colour and contained plagioclase, pyroxene, quartz, sulfates, chlorites, and Fe-oxides [1]. Furthermore, small crystals of Fe-oxyhydroxide (20~100 nm) associated with larger smectite surfaces (100~500 nm) and smectite lamellae of illites composed of nanochalcopyrite were detected, as shown in Figure 2. Moreover, clay particles of 50 nm size and Fe-oxyhydroxides of 500 nm size were also detected (Figure 3). The mineralogy of Fe-oxyhydroxides was mainly characterised by ferrihydrite with variable crystallinities and goethite. The results of this study indicate that multiple UFPs and NPs containing PHEs were detected in the studied area. As indicated above, the sampled phases were examined directly without prior treatment and by a non-aggressive approach using advanced microscopies. The goal was to simulate the real environment, and consequently determine the probable causes of the diverse UFPs and NPs, including PHEs. Furthermore, a combination of data from a range of utilised techniques, such as the analyses of contamination history, depositional conditions, and geochemistry, are required to thoroughly understand the origin of the nanominerals [34]. The results showed that the majority of the particles occurred as aggregates; although, distinct particles were undetected in these size ranges. Furthermore, the aggregate shapes were irregular, nearly equidimensional, and occasionally rounded outlines. Typically, NPs can preserve or exhibit multifaceted chemical narrations through growth or development in various ecological situations (e.g., Figure 2). For illustration, agronomic activities and mining are known to generate NPs-rich soils [36].

In this study, Hg was freely detected in such samples, as observed in Figure 3. Furthermore, it was observed that when in contact with oxygen and water, the sulfides degraded the amorphous phases of clays and quartz, which are crucial elements that prevent the mobilisation of PHEs, such as Hg and As. The detected PHEs manifested a high capacity for adsorption to Fe in the presence of Fe-carbonate-containing feldspar potassium (Figure 4). This can be attributed to the high surface area and high surface charge of feldspar potassium [37] Figure 5 illustrates the transformations of cubic and spherical pyrites (i.e., 30 μm–100 μm) into jarosite (i.e., 10 nm–50 nm), which are subsequently transformed into amorphous phases. It was observed that during such mineralogical modifications, PHEs, such as Hg, As, Cd, and Pb, are present in both crystalline and amorphous phases.

Figure 6 illustrates the transformations of calcite. The results of the SAED analysis proved that only gypsum was found in the crystalline phase during the transformation (Figure 6). Moreover, the EDS analysis revealed that different types of PHEs, including Cd, Hg, Ti, and Sr, were also found in lower proportions during the transformation. It is evident that hematite in the association of As, Cr, Hg, V, and Zn was present in all of the samples (Figure 7). Figure 7 exemplifies the EDS results of the HRTEM image of typical hematite with As and Hg, and there is no proof of presence of small traces of aluminosilicates, which could be detected due to the interactions between the minerals [16,38,39].

4. Discussion

The metallic mineralisation could be linked to sulphide, principally pyrite, with ores in the forms of veinlet and stock work. The geo-characteristics of the PHEs were directly affected by particle size distribution, mineralogical abundance, form factors, mineral association types, and other environmental factors, including temperature and humidity [34]. Most aggregates are categorized by the size and geochemical heterogeneity of the constituent particles, which range in sizes from hundreds to tens of nanometres with distinctive curvy outlines [40,41]. The identified NPs by advanced microscopy (AM) show variation from the multifaceted combinations of inorganic and organic phases ranging from several amorphous assemblies to carbonaceous matter [23,34]. Gasoline and diesel combustion results in several C-NPs linked with PHEs, polycyclic aromatic hydrocarbons (PAHs), and heterocyclic aromatic mixtures generated from evaporation and the incomplete burning of lubricants [42]. The air incidental NPs can also grow into larger UFPs that perform as cloud compression nuclei and help light absorption, which disturb global heating progressions [43]. Clays frequently contain secondary minerals formed by the alterations of primary minerals, such as calcite, pyrite, marcasite, and Fe-sulfides [44,45].

In acidic soils, the distribution of Zn on phyllosilicate surfaces occurs mainly by electrostatic interactions [46]. However, Zn was mostly detected in the sphalerites, Al/Fe-oxyhydroxides, and organic matter of the soils examined in this study [29,47]. Such surface coatings of montmorillonites by Fe-oxyhydroxides can compensate for the adverse effects of clays under acidic conditions. Several PHEs, including As, Hg, Cu, Cr, and Pb, have also been detected in the studied soils using TOF-SIMS analysis [20]. The observation described above is critical because gold mines normally contain As, Ti, and Cr [1] In addition, Hg is generally used to prevent the mobilisation of toxic elements during gold extraction. The occurrences of PHEs and mineralogical phase transformations were also detected in the artisanal mining soils [20,29]. The geo-accumulation index manifested the extreme contamination of As and Hg in the illegal mining areas of Colombia. Hence, due to the high volatility of Hg, artisanal gold mining causes serious threats globally [2,4].

It has been demonstrated that the amorphous phases of clay and quartz can immobilise the PHEs [12], thereby causing severe health risks (probably due to the inhalation of ultra-fine particulate materials) in the gold mining areas. Therefore, the results obtained in this study could assist government authorities to thwart the clandestine mining practices in Colombia and other countries around the globe.

5. Conclusions

This study analysed and discussed the relationship between PHEs and micro- or nanometric mineralogical phases in the Cauca region of Colombia, where an illegal mining practice with Hg and cyanide occurs. The study showed that PHEs (Hg, As, Cd, and Pb) are strongly associated with both crystalline (quartz) and amorphous phases of clays. PHEs demonstrated a high ability for adsorption to Fe in the existence of Fe-carbonate-containing feldspar potassium.

Therefore, the present study could serve as a reference for future research on the recovery of artisanal gold mining areas. The appropriate use of community dynamics (such as available resources) and the necessary supports from the government could effectively prevent clandestine gold mining practices. Future research should focus on the implementation of economically practical methods for the sustainable recovery of gold.