The chemistry and toxicity of discharge waters from copper mine tailing impoundment in the valley of the Apuseni Mountains in Romania.

Copper mining generates large quantities of waste, tailings, and acid outflows causing long-term environmental impacts and potential threats to human health. Valea Şesei is the largest tailing impoundment in Romania, created by flooding the valley (known as Valea Şesei) of the Metalliferous Mountains (a division of the Apuseni Mountains) with copper mining waste. The present study (i) estimated the total volume of tailings in this area; (ii) screened the concentration of 65 elements (rare earth and platinum group elements, alkali metals and alkali earth metals, transition and post-transition metals and metalloids) and cyanide concentrations in wastewater samples collected from tailing impoundment; (iii) evaluated the toxicity of these water samples using five in vitro bioassays employing human cells isolated from healthy donors and a short-term (1 h) exposure model. The sampled waters were highly acidic (pH 2.1-4.9) and had high electrical conductivity (2.80-15.61 mS cm-1). No cyanides were detected in any sample. Water samples collected from the stream (AMD) inflowing to the tailing impoundment were characterized by the greatest concentrations of alkali metals, alkaline earth metals, transition and post-transition metals, metalloids, rare earth elements, and noble metal group. At other sites, the elemental concentrations were lower but remained high enough to pose a relevant risk. The greatest magnitude of in vitro toxic effects was induced by AMD. Observed alterations included redox imbalance in human neutrophils followed by lipid peroxidation and decreased cell survival, significant aggregation of red blood cells, and increased prothrombin time. The study highlights that Valea Şesei is a large sink for toxic elements, posing environmental and health risks, and requiring action to prevent further release of chemicals and to initiate restoration of the area.

Introduction

Copper (Cu) is one of the most economically important metals with demand expected to significantly increase by 2050 (Elshkaki et al. 2016). Its mining has a significant environmental impact as it is mostly conducted in open pits and generates large quantities of waste. Production of 1 t of Cu typically requires over 150 t of ore to be excavated, crushed, concentrated by froth flotation, and then extracted using different methods depending on the sulfide or non-sulfide nature of the ore (Wills and Napier-Munn 2006). The whole process produces large amounts of tailings, which are the residues of the ore remaining after Cu extraction and contain high levels of toxic metals (e.g., Cd, Fe, Pb, Zn) and metalloids (e.g., As) (Courtney 2013; Wang et al. 2016). Additionally, outflows from mines, known as acid mine drainage, contribute to the allocation of toxic elements dissolved from rocks overrun by water characterized by extremely low pH, primarily due to oxidation of iron sulfide (Akcil and Koldas 2006).

The key issues in decreasing the impact of mining on environmental and human health include (i) mitigation of acid mine drainage effects by preventing oxidation, neutralizing the acid, or collecting the runoff (Akcil and Koldas 2006) and (ii) appropriate management and disposal of tailings (Mleczek et al. 2016). Most conventional impoundment methods involve the construction of surface ponds with raised embankments in which tailings and mine water are systematically deposited downstream, upstream, or centerline (Vick 1990), although methods such as sub-marine disposal or disposal of dewatered tailings as backfill are also in use (Berkun 2005; Fall et al. 2010). Tightened legislation and increased public awareness have forced mine operators to reinforce the stability of embankments and dams, and monitor the process of disposal (Dold 2008). Otherwise, a dam failure and subsequent release of great quantities of waste may cause an environmental disaster and large-scale health threat—a number of such events have already occurred in mining history (Rico et al. 2008).

In some of locations, tailings are still deposited on natural surfaces without raised embankments. One of the most striking examples of this is Valea Şesei, the largest tailing impoundment located in the Apuseni Mountains in Romania (Milu et al. 2002). It was created as a consequence of the open pit Rosia Poieni Cu ore exploitation initiated in 1986 and encompasses an area of the former Geamăna village (Milu et al. 2002), now almost entirely engulfed in tailings. Only a few inhabitants still live in the direct vicinity of the impoundment, the rest having been relocated. As the mine operates, the valley is systematically flooded by tailings and mine water while buildings of the former village are literally drowned in waste, from which the tower of its historical church can still be seen partially protruding (Fig. 1).

Fig. 1

Acid mine drainage inflowing from the Rosia Poieni copper mine to Valea Şesei (a) and views of different parts of tailing impoundment (b–e)

The assessment of environmental and health risks arising from the presence of such deposits of tailings as found in Valea Şesei is highly important if one considers its size and location, and is likely to require an interdisciplinary approach. Over the years, a number of unverified claims as well as anecdotal evidence regarding threats associated with this area have been raised by the media and various websites. These have included the presence of a high concentration of cyanide compounds and mercury in the wastewaters. As shown by only a few studies, waters draining freely from the ore deposit are contaminated with toxic metals (Milu et al. 2002; Melenti et al. 2011) while the entire area contributes significantly to the pollution of the river Aries and its tributaries (Levei et al. 2013). The wastes are continuously deposited as the mine operates, highlighting the necessity for monitoring of this area as regards toxicological risks.

The present study was undertaken in an effort to evaluate the chemical composition of wastewaters generated by the Rosia Poieni Cu mine and the associated toxicological risks. This was assessed by (i) characterizing the morphometry of the Valea Şesei tailing impoundment; (ii) a broad, multi-element ICP-OES screening (total of 67 elements) and analyses of cyanide concentrations of wastewaters from this area, and (iii) toxicological studies of wastewaters by a battery of in vitro bioassays employing cells isolated from healthy human donors. The present study represents the most up to date information on the chemistry and toxicity of wastewaters from Valea Şesei and introduces biomedical tools as part of an integrated ecotoxicological assessment.

Materials and methods

Study area

The Valea Şesei is the largest Romanian tailing deposit located in the Metaliferi Mountains, a division of the Apuseni Mountains (Western Carpathians) created by flooding the dammed valley with tailings and wastewater originating from Rosia Poieni Cu ore, its processing, and smelting (Fig. 2). The exploitation is carried out in the open pit manner and is currently the second largest in Europe, with more than a billion tonnes of porphyry-type ore containing on average 0.36% Cu (Milu et al. 2004). The dam, built from limestone blocks and gravel, is located in the northern part of the impoundment area, has a height of 118 m, and has an inclination of 33° (Duma 1998; Melenti et al. 2011). The nearest water course is the river Aries, located downhill approximately 3.5–4.0 km from the dam, progressing to the rivers Mures, Tisa, and finally through the river Danube into the Black Sea. Geological structures underneath the tailing deposit consist of sedimentary rocks (type not specified) on the west side and crystalline limestone on the east side. It was reported that groundwater up to 8 m below tailing pond are being affected by acid water seepage (Melenti et al. 2011). The surface waters can be found mainly in the southern part where the largest stream inflow is located and in the south-eastern part (Fig. 2a). The area surrounding the tailing storage is hilly and covered by deciduous forests with a prevalence of beech or coniferous trees (Melenti et al. 2011). There is no fence or whatsoever to block an access to impoundment and effluent. The impoundment surroundings are still inhabited by a few locals.

Fig. 2

Location of studied area and sampling points (a) and morphometric characteristics of Valea Şesei tailing impoundment (b)

Quantification of morphometry of tailing impoundment area

An ortophotomap of Valea Şesei originating from 2012 was laid on a military topographic map scale 1:25k (HTM sheets/Hărți topografice militare/: L-34-59-C-d; L-34-71-A-b) using ArcView software (Environmental Systems Research Institute, USA). Due to terrain generalization, the shoreline of the tailing impoundment area frequently crossed contour lines at 690 and 700 m a.s.l. Its course was most consistent with the outline of the 690 m a.s.l. contour line; thus, it was used for cartometric measurements. The topographic map was realized using a 10-m contour interval. The bathymetric map was realized using those contours lower than 690 m that are precise enough to reflect isobaths of the Valea Şesei impoundment area. The total volume of impoundment was evaluated with a bathymetric map applying the formula for the truncated cone (Håkanson 1981).

Sampling

Sampling took place during summer 2016. The water samples were collected with a telescopic sampler from near the surface of the stream flowing from the area of the Rosia Poieni Cu mine to Valea Şesei (AMD, 46° 19′ 04.3″ N 23° 11′ 55.7″ E) and from three sites WW-1 (46° 19′ 19.6″ N 23° 12′ 33.1″ E), WW-2 (46° 19′ 31.3″ N 23° 12′ 43.8″ E), and WW-3 (46° 19′ 09.8″ N 23° 13′ 22.2″ E) located along the shoreline of the tailing impoundment (Fig. 2). Samples were collected in 0.5-L polypropylene bottles. Three replicates were collected for each site.

Chemical analyses

General physicochemical properties

Electrical conductivity (EC) and pH of collected samples were measured in situ using a portable multi-parameter probe HI9829 (Hanna Instruments, USA).

Elemental composition

The water samples were filtered through 0.45-μm filters (Whatman, UK) and acidified with suprapure nitric acid (Sigma-Aldrich, Germany). Elemental analysis was performed using the inductively coupled plasma optical emission spectrometer Agilent 5100 ICP-OES (Agilent, USA). A simultaneous axial and radial view of plasma was obtained by a synchronous vertical dual view (SVDV) using dichroic spectral combiner (DSC) technology. The following common conditions were applied: radio frequency (RF) power 1.2 kW, nebulizer gas flow 0.7 L min−1, auxiliary gas flow 1.0 L min−1, plasma gas flow 12.0 L min−1, charge-coupled device (CCD) temperature −40 °C, viewing height for radial plasma observation 8 mm, accusation time 5 s, three replicates. The calibration was performed using standard analytical solutions (Merck, Germany). A total of 66 elements were analyzed and grouped as follows: alkali metals (Cs, K, Li, Na, Rb), alkaline earth metals (Ba, Be, Ca, Mg, Sr), transition metals (Cd, Co, Cr, Cu, Fe, Hf, Hg, Mn, Mo, Nb, Ni, Ta, Ti, V, W, Zn, Zr), post-transition metals (Al, Bi, Ga, In Pb, Sn, Tl), metalloids (As, B, Ge, Sb, Si, Te), rare earth elements (Ce, Eu, Er, Gd, La, Nd, Pr, Sc, Sm, Dy, Ho, Lu, Tb, Tm, Y, Yb), and noble metals (Ag, Au, Ir, Os, Re, Rh, Ru, Pd, Pt). Specific information on applied wavelengths and detection limits for each element is given in the Supplementary Data (Table S1).

Cyanide concentration

After collection, water samples were preserved by adjusting their pH to 10 with sodium hydroxide after which they were centrifuged at 4500g. Total cyanide concentration was determined by a continuous flow analysis system with a photometric flow detector SAN++ (Skalar, Netherlands) according to ISO 14403-2: 2012(E). The detection limit was 5.0 μg L−1.

Toxicological studies

Experimental design

The toxicological activity of discharge waters collected from Valea Şesei was assessed and compared using a battery of in vitro assays employing human cells. Water samples were prepared by centrifugation at 4500g and filtered through the injection filter with a nominal pore size of 0.2 μm (Sartorius, Germany). The potential inducement of oxidative stress and its detrimental effects was tested in isolated neutrophils and included investigations of intracellular reactive oxygen species (ROS) generation, peroxidation of lipids, and cell survival. The potential effect on coagulation parameters, prothrombin time, and international normalized ratio was tested in separated human plasma, whereas red blood cell aggregation rate was assessed in whole blood via erythrocyte sedimentation rate. The final concentration of waters in exposed samples was always 1%. For each assay, peripheral blood samples were collected from three healthy (screened by physical examination, medical history, and initial blood tests) non-smoking and normal weighted (BMI 18.5–24.9) human donors at the Regional Centre of Blood and Blood Treatment in Poznan, Poland, according to accepted safeguard standards and legal requirements.

Human neutrophils isolation

Human neutrophils were isolated from blood samples collected in lithium heparin tubes (Becton-Dickinson, USA) using a one-step density-gradient centrifugation on Gradisol G of specific gravity of 1.115 g mL−1 (Polfa, Poland) at 400g at room temperature for 30 min. The residual erythrocytes were removed from the cell population by hypotonic lysis. The purity of the neutrophils (>90%) was verified by counting under a light microscope after May-Grunwald-Giemsa staining.

Intracellular reactive oxygen species assay

Neutrophils were loaded for 30 min at 37 °C in darkness with 20 μM of 2′,7′-dichlorofluorescin diacetate (DCFDA; Abcam, UK), a fluorogenic dye that measures hydroxyl, peroxyl, and other ROS activities within the cell. Neutrophils were then washed, dissolved in PBS, and seeded in a black clear bottom 96-well plate at a density of 20 × 104 cells per well (99 μL aliquots/well). Water samples were added in the amount of 1 μL and the plate was incubated at 37 °C for 1 h. Fluorescence of DCFDA was measured kinetically after 5, 15, 30, and 60 min of incubation using a Synergy HTX multi-mode plate reader (BioTek, USA) at an excitation of 495 nm and emission of 528 nm. The background signal, measured in exposed neutrophils not loaded with DCFDA, was withdrawn from the corresponding samples. The final results were presented as a percentage of a parallel control constituted of neutrophils incubated with 11 μL of PBS.

Lipid peroxidation assay

Lipid peroxidation was analyzed using a Lipid Peroxidation Colorimetric/Fluorometric Assay Kit (BioVision, UK) by means of malondialdehyde (MDA) content. Human neutrophils were seeded in a 96-well plate at a density of 20 × 104 neutrophils per well (99 μL aliquots/well) and exposed to 1 μL of each discharge water sample for 1 h at 37 °C. The control was constituted of cells incubated with 11 μL of PBS. After the experiments, cells were harvested from each well and homogenized on ice in 300 μL of provided lysis buffer (with the addition of butylated hydroxytoluene to prevent artificial lipid peroxidation) and centrifuged to remove insoluble material. The resulting 200 μL of supernatants was transferred to a microcentrifuge tube and supplemented with 600 μL of thiobarbituric acid (TBA) to generate an MDA–TBA adduct. To accelerate the reaction, samples were incubated at 95 °C for 60 min and the final product was measured colorimetrically at 532 nm. Three technical replicates were conducted for each donor. The calculated values were compared to a calibration curve prepared using MDA standard (BioVision, UK). The coefficient of variation (r 2) for the calibration curve was 0.99. The final results were presented as a percentage of a parallel control.

Cell viability assay

Cell viability was determined using the colorimetric 3-(4,5-dimethylthiazol-2-yl)-2,5 diphenyltetrazolium bromide (MTT) metabolic activity assay (BioVision, UK) according to the manufacturer’s instructions. Human neutrophils were seeded in a 96-well plate at a density of 20 × 104 neutrophils per well (99 μL aliquots/well) and exposed to 1 μL of each water sample for 1 h at 37 °C. Afterwards, cells were washed, seeded again, and 10 μL of MTT was added to each well for 2 h. Neutrophils were then treated with 10% sodium dodecyl sulfate in 0.01 HCl and incubated for another 6 h in darkness to dissolve formazan crystals. The optical density (OD) of the final product (the formazan crystals) was measured at 570 nm using a Synergy HTX microplate reader (BioTek, USA). The final results were presented as a percentage of a parallel control constituted of neutrophils incubated with 1 μL of PBS.

Coagulation assay

The effects of discharge waters on blood coagulation level were assessed with the prothrombin time measurement and calculation of the international normalized ratio (INR). Venous blood was collected in tubes containing 3.2% sodium citrate as an anticoagulant (Sarstedt S-Monovette® 9NC). The ratio of blood to anticoagulant was no less than 10:1. Plasma was obtained by centrifugation at 1500×g at 21 °C for 15 min, pipetted and divided into five 990 μL aliquots. The first aliquot was used as a control (with 10 μL of PBS), the subsequent aliquots were exposed for 1 h to 10 μL of each water sample at 37 °C. The prothrombin time was measured immediately with the optical nephelometric method using a K-3002 Optic coagulometer (Kselmed, Poland) in a certified hematological laboratory. INR was calculated from the following formula:

The normative laboratory range for prothrombin time and INR was 12–16 s and 0.9–1.2, respectively. Additionally, the plasma samples (1 mL) were exposed to a 10-fold higher volume of samples (100 μL) to magnify the potential effects; all samples were then photographed after 5 min.

Erythrocyte sedimentation rate assay

To assess whether discharge water samples may affect the erythrocyte sedimentation rate (ESR) also known as Biernacki’s reaction, 5 mL of venous blood of healthy donors was collected directly into BD Vacutainer Seditainer™ glass tubes containing 3.2% sodium citrate as an anticoagulant (4:1) in quintuplicate. Water samples were immediately added in the volume of 50 μL; tubes were inverted eight times to mix and inserted into the BD Vacutainer Seditainer™ stand for 5 h. ESR was measured hourly and given as millimeters of sedimented red blood cells.

Statistical analyses

The results were analyzed using STATISTICA 10.0 software (StatSoft, USA). Differences in elemental composition between the investigated water samples were evaluated using one-way ANOVA with the Tukey post hoc test. In toxicological studies, the paired sample t test was applied to compare the difference in effects induced by each water sample with the control. p < 0.05 was considered as statistically significant.

Results

Morphometric characteristics of tailing impoundment

The detailed morphometry of Valea Şesei tailing impoundment is presented in Fig. 2b. The total surface area is 1.426 km2 whereas the volume of deposited tailing amounts to 36.01 hm3. The depth is systematically rising in the direction of the dam with mean and maximum depth of 25.3 and 70 m, respectively (Fig. 2b).

General physicochemical properties

All investigated water samples were characterized by low pH and high electrical conductivity. The determined ranges of pH were as follows: AMD—2.1–2.2, WW-1—2.5–2.6, WW-2—2.8–2.9, and WW-3—4.8–4.9; whereas mean ± SD electrical conductivity was 15,610 ± 80 μS cm−1, 7163 ± 57 μS cm−1, 3746 ± 62 μS cm−1, and 2800 ± 55 μS cm−1 for AMD, WW-1, WW-2, and WW-3, respectively.

Elemental composition

Of the 65 investigated elements, the concentration of 17 of them were below the level of detection (Au, Cs, Er, Ga, Ge, Hf, Hg, Nb, Pb, Pd, Re, Ru, Sb, Sm, Sn, Ta, and Tm).

Alkali metals and alkaline earth metals

The concentration of alkali metals was diverse within the studied waters. The AMD site was characterized by the lowest Na and K concentrations which, in turn, reached the greatest level at WW-3 and exceeded 14 and 30 mg L−1, respectively. The highest concentration of Li, exceeding 0.6 mg L−1, was found at AMD whereas Rb concentrations were similar at all studied sites and amounted on average to 3.0 mg L−1 (Fig. 3).

Fig. 3

A–D The concentrations of selected alkali metals in water samples collected at Valea Şesei. Identical superscripts (a, b, c) denote no significant (p > 0.05) difference between mean values in rows according to Tukey’s HSD test (ANOVA). DL detection limit. Cs was not detected in any sample

Similarly, the concentrations of alkaline earth metals were also significantly different in the investigated waters. In general, Mg and Ca were the most abundant elements of this group. The AMD site was characterized by the highest concentration of Mg (exceeding 550 mg L−1) and Be (over 0.1 mg L−1) but decidedly the lowest levels of Ba and Ca. The concentration of Ca was at least 2-fold higher at other sites and amounted to 650 mg L−1 at WW-2 (Fig. 4).

Fig. 4

A–E The concentrations of selected alkaline earth metals in water samples collected at Valea Şesei. Identical superscripts (a, b, c) denote no significant (p > 0.05) difference between mean values in rows according to Tukey’s HSD test (ANOVA)

Selected transition and post-transition metals

Generally, the greatest number of transition metals and their highest concentrations were identified at the AMD site (Table 1). Cr, Ti, and Zr were detected only at this site. Fe, Zn, Cu, and Mn were the most abundant elements of this group, with mean concentrations exceeding 4500, 600, 250, and 60 mg L−1, respectively. The AMD site was also characterized by increased levels of Co (over 2.0 mg L−1), Cd, and Ni (over 1.0 mg L−1 each) and a relatively high concentration of V and W (over 0.5 mg L−1 each). The decidedly lowest concentration of transition metals was found at WW-2, with a number of elements below the limit of detection and trace quantities of Fe and Zn (Table 1.)

Table 1

Mean ± SD (maximum) content (mg L−1) of transition and post-transition metals in water samples (n = 3) collected at Valea Sesei

	AMD	WW-1	WW-2	WW-3
Transition metals
Cd	1.08 ± 0.18a (1.20)	0.22 ± 0.01b (0.24)	<DL	0.01 ± 0.002c (0.0167)
Co	2.20 ± 0.35a (2.47)	0.88 ± 0.07b (0.96)	<DL	0.003 ± 0.006c (0.004)
Cr	0.02 ± 0.02 (0.04)	<DL	<DL	<DL
Cu	259.3 ± 52.4a (291.1)	86.4 ± 6.4b (93.0)	<DL	2.77 ± 0.22c (2.95)
Fe	4669.6 ± 727.9a (5118.2)	410.9 ± 46.7b (459.1)	0.23 ± 0.41c (0.71)	0.006 ± 0.01c (0.01)
Mn	62.3 ± 11.4a (69.8)	25.2 ± 1.8b (26.9)	2.45 ± 0.74c (3.28)	3.82 ± 0.31c (4.09)
Mo	0.01 ± 0.007a (0.015)	0.002 ± 0.003a (0.006)	0.005 ± 0.005a (0.01)	0.02 ± 0.003a (0.025)
Ni	1.48 ± 0.26a (1.65)	0.36 ± 0.03b (0.39)	<DL	0.03 ± 0.007c (0.035)
Ti	0.03 ± 0.03 (0.200)	<DL	<DL	<DL
V	0.508 ± 0.165a (0.606)	<DL	0.001 ± 0.001b (0.003)	<DL
W	0.711 ± 0.228 (0.902)	<DL	<DL	<DL
Zn	610.9 ± 129.9a (692.7)	65.6 ± 7.8b (74.2)	0.04 ± 0.02d (0.06)	4.89 ± 0.35c (5.14)
Zr	0.03 ± 0.03 (0.06)	<DL	<DL	<DL
Post-transition metals
Al	4488.3 ± 831.1 a (5014.8)	826.1 ± 68.1b (894.4)	0.19 ± 0.17d (0.34)	8.94 ± 0.14c (9.10)
Bi	0.61 ± 0.10a (0.71)	0.07 ± 0.08b (0.18)	0.14 ± 0.03b (0.17)	0.07 ± 0.03b (0.10)
In	0.50 ± 0.11a (0.62)	0.26 ± 0.05b (0.32)	0.35 ± 0.04ab (0.40)	0.34 ± 0.05ab (0.39)
Tl	0.09 ± 0.09a (0.20)	<DL	0.01 ± 0.02a (0.04)	0.03 ± 0.06a (0.11)

Au, Ga, Hf, Hg, Nb, Pd, Re, Sn, and Ta were not identified in any sample. Identical superscripts (a, b, c, d) denote no significant (p > 0.05) difference between mean values in rows according to Tukey’s HSD test (ANOVA)

DL detection limit

Al was the most abundant element from the group of post-transition metals. Its high concentrations were identified for each site except WW-2 (Table 1). At AMD, its mean level exceeded 4400 mg L−1. The concentrations of other elements of this group were relatively low. The highest concentrations of Bi and In (>0.5 mg L−1) were found at the AMD site (Table 1).

Metalloids

The highest concentration of detected metalloids was again found at the AMD site. As and Te were identified above the limit of detection only for this site. Compared to WW-1, WW-2, and WW-3, Si concentrations at AMD were 3-fold, 66-fold, and 18-fold higher, respectively (Table 2).

Table 2

Mean ± SD (maximum) content (mg L−1) of metalloids in water samples (n = 3) collected at Valea Sesei

	AMD	WW-1	WW-2	WW-3
As	0.107 ± 0.05 (0.153)	<DL	<DL	<DL
B	6.63 ± 1.38a (7.50)	0.54 ± 0.07b (0.63)	0.05 ± 0.07b (0.06)	0.03 ± 0.02b (0.04)
Si	85.7 ± 14.9a (95.2)	28.1 ± 2.1b (30.2)	1.30 ± 0.07c (1.36)	4.70 ± 0.4c (5.0)
Te	0.51 ± 0.25 (0.75)	<DL	<DL	<DL

Ge and Sb were not identified in any sample. Identical superscripts (a, b, c) denote no significant (p > 0.05) difference between mean values in rows according to Tukey’s HSD test (ANOVA)