On the volatilisation and decomposition of cyanide contaminations

Science of the Total Environment 627 (2018) 1167–1173

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

On the volatilisation and decomposition of cyanide contaminations from gold mining Andreas Brüger a,⁎, Günter Fafilek a, Oscar J. Restrepo B. b, Lucas Rojas-Mendoza b a b

Institute of Chemical Technologies and Analytics, Vienna University of Technology, Getreidemarkt 9, A-1060 Vienna, Austria Departamento de Materiales y Minerales, Facultad de Minas, Universidad Nacional de Colombia, Medellín, Colombia

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• After volatilisation from water, no cyanide destruction via photolysis or oxidation was observed. • High concentrations of cyanide were found in water bodies downstream gold processing centres. • A mass balance of cyanide over the complete cyanidation process is necessary. • A simple and location-independent method for cyanide analysis is applied.

a r t i c l e

i n f o

Article history: Received 13 November 2017 Received in revised form 30 January 2018 Accepted 30 January 2018 Available online xxxx Editor: F.M. Tack Keywords: Cyanide Oxidation Photolysis Artisanal gold mining

a b s t r a c t Cyanide leaching is the predominant process of gold extraction in large scale mining. Current initiatives for reducing the use of mercury in small scale and artisanal mining tend towards the cyanide technology as the only feasible alternative. Thus, the deliberate handling in consideration of the hazardous nature of cyanide compounds is an issue of particular importance. The hydrogen cyanide volatilised during the leaching process and from the tailings solutions after the gold extraction is reported to be destroyed by oxidation and photolysis from the surrounding atmosphere of gold mines and the sunlight. Cyanide solutions, drained into the surrounding waterbodies are stated to volatilise at a high rate, thus detoxifying them and releasing hydrogen cyanide to the air. In this study laboratory experiments and field tests were conducted to deliver basic data for the volatilisation and destruction of cyanide in the environment. In our laboratory tests we observed neither oxidation by the oxygen of air nor photolysis by UV-irradiation of cyanides after volatilisation from water. The whole amount of volatilised cyanide was found in the exhaust gas after absorption in a strong basic solution. Field experiments in Segovia (Colombia) could confirm these findings. Cyanide concentrations in a range 17 to 30 mg/L were measured in a local creek. Hydrogen cyanide amounts of 5 ppm were found in the atmosphere surrounding cyanide leaching facilities. With the findings of this study we want to point out that the concentrations of cyanide in the surrounding of cyanide leaching facilities exceed uncritical limits and a destruction via oxidation and photolysis is not detectable. These conclusions should result in initiatives to protect workers and the surrounding population of gold mines from contaminations of cyanide treatments. © 2018 Elsevier B.V. All rights reserved.

⁎ Corresponding author. E-mail addresses: [email protected] (A. Brüger), guenter.fafi[email protected] (G. Fafilek), [email protected] (O.J. Restrepo B.).

https://doi.org/10.1016/j.scitotenv.2018.01.320 0048-9697/© 2018 Elsevier B.V. All rights reserved.

1168

A. Brüger et al. / Science of the Total Environment 627 (2018) 1167–1173

1. Introduction About 90% of the world's gold mining workforce is represented by artisanal and small-scale gold miners, producing around 15% of the world's gold (UNEP, 2012). In Colombia there are about 200,000 artisanal gold miners, using mainly mercury for the gold processing (Cordy et al., 2011). The amalgamation process suffers a low capability of mercury recovering. Rarely N30% of the gold is captured (UNEP, 2012). The Hg-contaminated tailings are further leached with cyanide, followed by zinc shaving precipitation. After the leaching process, the gold-barren waste, containing metallic mercury and mercuric cyanide complexes as well as other toxic metal complexes with cyanide is discharged into local creeks. This two-stage process of whole-ore amalgamation followed by cyanidation is practiced in artisanal gold mines around the world ((Krisnayanti et al., 2012; Shandro et al., 2009; Telmer and Veiga, 2009). Initiatives were launched in recent years to reducing mercury in artisanal and small-scale gold mining (e.g. the Global Mercury Project (UNIDO, 2004) and the Minamata Convention on Mercury (UNEP, 2013)). Cyanide is recommended as a viable alternative in small scale mining (UNEP, 2012). The utilisation of cyanide leaching processes in artisanal mining requires the formation of mining cooperatives or associations to the extent that the gold ore processing, using cyanide, can hardly be done by individuals. In combination with technically advised processing centres and the appropriate working with cyanide, this would furthermore support projects to reducing mercury in artisanal mining. Approach has been taken to support the collaboration between large and small-scale miners (IFC-CommDev/CASM/ICMM, 2008). In large-scale mining, cyanide leaching is the predominant method for the gold extraction. It is reported to be utilized in 90% of gold producing operations worldwide (Laitos, 2013; Mudder and Botz, 2004). Concerning the use of NaCN for the leaching process, an estimate for the release to the atmosphere of 30%–50% as HCN and a global use of 330,000 t of NaCN for gold mining per annum is quoted by NICNAS (2010), this leads to 55,000 to 90,000 t HCN. Compared to the global atmospheric budget for HCN of 2 mio t this would be a minor contribution to the global HCN release (NICNAS, 2010). Considering more recent data for the usage of NaCN in the gold extraction of 710,000 t, based on 880,000 t world production in 2013 (Mcgroup, 2014) with 81% of this demand used for gold extraction (Kobayashi, 2012) and the global trend of increasing cyanide consumption with decreasing ore grade (Mudd, 2007), the contribution is doubled. Cyanide-leaching accidents in the past have received widespread public attention. The costs of remedial actions after acid mine drainage and leakages of the Summitville mine (USA) exceeded the total profit of gold and silver extraction (CDPH, 2010; Ketellapper et al., 1996). The discussion following the cyanide spill at Baia Mare in Romania when a tailings dam broke in 2000 resulted in the development of the International Cyanide Management Code (ICMC) of best practice for the use of cyanide in the production of gold. This program was designed by a multi-stakeholder steering committee under the guidance of the United Nations Environmental Program (UNEP) and the International Council on Metals and the Environment (ICMI, 2009). The ICMC is a voluntary program for gold mining companies as well as for cyanide producers and transporters. It is considered as an initiative to lift international standards as well (NICNAS, 2010). The ICMC as well as the new international guidelines and limits seem to represent a major step forward to improving social, health, safety and environmental conditions in the gold mining sector. However, despite these improvements, the risk of possible accidents and fatalities remain (Newmont, 2010; Husseini, 2013; T-SG, 2014). In literature there is some claim that small non-lethal doses of cyanide do not give rise to chronic health problems (Logsdon et al., 1999), or quickly metabolize in the body arguing the uncommonness of chronic exposure to cyanides (SWA, 2013). Two detoxification rates

have been estimated for men after intravenous injection (e.g. SCOEL, 2010; USEPA, 2010). The rate of 17 μg/kg/min is based on a study using hydrocyanic acid in saline solutions (Loevenhart et al., 1918), the metabolisation data of 1 μg/kg/min was calculated based on effects of nitroprusside on humans (Schulz et al., 1982). Both publications can hardly match the requirements for a contemporary toxicological estimation. A main problem in the collection of information concerning the toxicity of cyanide to humans is the insufficient comparability of literature data. The increasing knowledge about cyanide poisoning restrains the necessity to revise the findings from the mainly prior research with its different state of knowledge and methods. Approaches in other locations of possible cyanide contaminations suggest further investigations of health effects as a result of cyanide leaching (Akinrele, 1986; Dhas et al., 2011; Kumbnani et al., 1990). A second claim is that cyanide is naturally degraded and does not persist in the environment due to oxidation and photolysis in the atmosphere (Logsdon et al., 1999; TRI, 1998). The USEPA describes ‘natural degradation’ as a general term for all of the processes that may reduce the total cyanide concentration in waste (in solution and soil) including volatilisation and evolution of HCN from CN-, (USEPA, 1994). This can be a slightly misleading description as the chemical understanding of ‘degradation’ seems to tend towards decomposition in this context (IUPAC, 2000; Rubo et al., 2000). Because of these varying definitions in literature, the persistence capability of cyanide in the environment should be followed in atmosphere, water and soil: The lifetime of hydrogen cyanide in the troposphere is calculated as 5 years primarily caused by oxidation and 6 months if influenced by ocean uptake; photolysis is described as being negligible due to the fact that the strong H-CN bond cannot be broken by photons available in the troposphere (Cicerone and Zellner, 1983; Kleinböhl et al., 2006). Experiments including the persistence of cyanide in water are published with varying results. Significant oxidation of cyanide with UVA irradiation in synthetic wastewater is found by Ozcan et al. (2012), whereas Simovic (1984) showed that neither the effects of UV light nor oxidation significantly contribute to the removal of cyanide in NaCN solutions. Free cyanide is reported to have potentially high mobility in soil and groundwater environments, depending on factors like pH, humidity and the type of soil, resulting in diffusion up through the soil and release to the atmosphere (Kjeldsen, 1999; NICNAS, 2010). Cyanide-metal complexes can be enriched in soils and represent a potential hazard to the groundwater in releasing toxic metal compounds and free cyanide as a result of decomposition under different circumstances (Kjeldsen, 1999; Shehong et al., 2005). The requirement of further investigations concerning the influence of sunlight, oxidation and volatilisation on cyanide removal in water and air is part of the motivation for this work. The results of these investigations could help to estimate the contamination of the atmosphere surrounding leaching facilities. Available data range from b1 ppm HCN at distances N1 m from the heaps (DeVries, 1996) to values of 0.3– 6 ppm CN in air depending on distance (in a range 2–52 m) to the leach pad and the wind conditions (Feigley et al., 1984). In air screenings as a consequence of the ICMC, the 2011 report of the Yanacocha mine (Peru) indicates areas with hydrogen cyanide levels in excess of 10 ppm (Golder, 2011). The TWA (maximum average airborne concentration to which workers may be exposed without adverse effect over an eight - hour working day, for a five - day working week) is 0.9 ppm, following the recommendations for Indicative occupational exposure limit values (SCOEL, 2010), published by the European Commission. Based on comprehensive data of cyanide exposure in any processing step and every place of possible contamination from gold mines and processing plants, an assessment of the hazard potential for different locations can be done to minimize the pollutions, caused by cyanide compounds. A mass balance as a result of this comprehensive data could help the comparison of cyanide exposures in different gold mines. In this work, one step forward is taken in receiving more information

A. Brüger et al. / Science of the Total Environment 627 (2018) 1167–1173

concerning the volatilisation and degradation of cyanide in water and air. Experiments were done in the laboratory and as a consequence in field studies in Segovia in the Department of Antioquia, Columbia. In Segovia, the largest gold producing town in Antioquia, artisanal as well as large scale mining activities are found (García et al., 2015). The mining area of Segovia in Colombia is difficult to access and although mining has been carried out for many years, the working conditions are less than ideal. There are difficulties of access, supplies and public order, which makes the measurement and control of the processes difficult, therefore, it is common for discharges and sometimes environmental damage to the soils and water sources ((Bustamante et al., 2016). This situation also influenced the possibilities of collecting samples and thus limited the number of possible measurements in the field experiments. 2. Materials and methods 2.1. Laboratory setup The determination of free cyanide was based on a standard method for the collection of HCN gas and the measurement of CN-concentrations in solution (SMWW, 2012). According to the standard method the volatilised HCN in the reaction flask was collected via suction of laboratory air through the volatilisation flask using a vacuum pump as an aspirator. Volatilisation tests at high pH (between 13 and 14) resulted in an increase of the cyanide concentration in the flask due to water volatilisation. Experiments in an initial pH range of 12.6 to 13 (with NaOH) were accompanied by a significant decrease in pH due to CO2 absorption from the laboratory air. An initial pH 13 solution, decreased to pH 11 after 96 h at an airflow rate of 12 L/h; volatilisation of water in this solution was up to 4% in 24 h depending predominantly on the relative humidity of the atmosphere surrounding the experiment. According to the standard method, the pH of the absorption solution for the collection of HCN is 12.6. Preliminary tests in our laboratory showed that KCN in a CN-concentration of 400 mg/L was stable against volatilisation only in NaOH solutions of pH ≥ 14 (see results). To overcome the above described problems, the standard method was modified as follows: Carbonation and evaporation of water were minimised by adding a 250 mL absorption flask, containing 200 mL of a 1 M NaOH solution preceding the flask containing the cyanide solution. The laboratory air was bubbled through a medium-porosity frit in this absorption flask for the removal of CO2 and the humidification of the laboratory air. After the implementation of the additional absorption flask, the pH didn't change over the period of the experiments and the evaporation of water in the reaction flask was b0.2–0,3% per day. The amount of evaporated water was determined gravimetrically and incorporated into the calculation of the degree of cyanide volatilisation. The volatilisation of HCN was assessed by passing the humidified and CO2-liberated laboratory air through a 500 mL three-neck roundbottom flask, containing 200 mL of buffered cyanide solution (surface area: 78 cm−2, initial cyanide concentration: 400 mg/L (15.37 mM)) without bubbling through the solution. The buffered solutions were produced, following standard preparation methods (Bates and Bower, 1956). The chosen basic range of the reaction solutions (pH 9–pH 14) should cover the whole range of possible pH values found during the cyanide leaching process in gold processing facilities. The HCN loaded air was bubbled through a medium-porosity frit in a 250 mL absorption flask, containing 200 mL absorption solution for the cyanide absorption. For the absorption as well as for further processing of the CN-containing samples, 1 M NaOH solutions were used. A second absorption flask was appended to control the efficiency of HCN absorption and removed, after the comprehensive absorption of cyanide in the first flask was confirmed. The air suction was done using flow rates between 12 L/h and 36 L/h (36 L/h is the maximum to keep the liquid phase in the absorption flask). The flow rate was adjusted with a rotameter, connected to a needle valve (Fischer & Porter). Compared to

1169

natural wind intensities, these flow rates are rather low but the target is reached with the complete absorption of volatilised HCN. The decrease of volume in the absorption flask due to volatilisation of water (up to 0.1% per day) was neglected at low cyanide absorption concentrations and incorporated in the calculations with cyanide absorption concentrations, higher than 100 mg/L in the absorption flask. A schematic of this set-up is shown in Fig. 1. The light source for the UV irradiation consisted of UVA fluorescent tubes (Sylvania Lynx BL350). This UV source has a wavelength maximum at 350 nm and a light irradiance of 66 W/m2 measured between 320 and 390 nm at a distance of 4 cm. Relative intensities of light emitted by the tubes as a function of wavelength were measured with an ocean optics fibre optic spectrometer, calibrated against a white light source of known colour temperature. The comparison to the solar irradiance (direct + circumsolar light AM1,5, ASTM G-173 with an irradiance of 24 W/m2 between 320 and 390 nm) is shown in Fig. S 1 in the supplementary information. The UV irradiation in our experiments was done, using three fluorescent bulbs placed in a box with a reflecting background. The light source was placed in a distance of 1.5 cm to the reaction flask, measured between the edge of the bulbs and the outer surface of the flask. The radiation loss through the glass of the flask is approximately 15% in the used wavelength range and can be followed in the supplementary information (Fig. S 1). The setup for the photolysis-experiments was similar to the volatilisation experiments described above. Buffered cyanide solutions (pH = 10.3, initial cyanide concentration: 400 mg/L) were filled into the reaction flask and irradiated between 24 h and 72 h. Air suction was done with a flow rate of 13 L/h. The CN concentration in the reaction flask and in the absorption flask was analysed after the experiments to determine the cyanide loss due to photolysis. 2.2. Setup for the field measurements Investigations of the cyanide content in the air surrounding processing plants were done at three locations. At the Entable JM, cyanide in air was measured in a distance of 1 m to a vat during a cyanide leaching process, activating the slurry with gas introduction. In the second location, the Mina La Campana, measurements were done inside a tank for the storage of cyanide solution before usage, in a distance of 1 m to the cyanide solution. The same distance was followed in the third location, the Mina Cogote, where the leaching was done in a tank equipped with an agitator and with gas introduction. The method for the field investigations was similar to the method described in the laboratory volatilisation experiments, but was adapted for a simple location-independent application. The measurements were done in a distance of 1 m to the cyanide containing slurry. The influence of wind was excluded by choosing an absorption period with calm conditions. These were done using a stand-alone setup for sample drawing (supplementary information Fig. S 2). A membrane air pump, (Zhongshan Jingdian Electric Appliance, 0.1–0.8 L/min), was connected to a 1.2 Ah lead-acid rechargeable battery (Exide Powerfit) to allow operations of about 6 h. To ensure longer experiment durations, a 3 W/630 cm2 amorphous solar panel could be connected to the unit. The air-flow rate was controlled using a flow meter (Fischer & Porter). pH measurements were done with universal indicator strips. The whole equipment was designed for a simple transportation with durable materials and varying applications, considering the different architecture of the goldmines in this region sometimes difficult to access. The sampling for the cyanide measurements in water was done at three positions in a river flowing through the area of gold mining (Quebrada la Cianurada). The locations of the sampling were chosen in order to compare different areas of potentially high cyanide levels and to make an estimation about the influence of contaminated side streams to the main stream of the river that collects most of the drains of the gold mines. One location was in a tributary stream, flowing through a mining area, the second point for the sampling was in the

1170

A. Brüger et al. / Science of the Total Environment 627 (2018) 1167–1173

G F air

E

B D

C

A

A. Absorption flask with fritted glass diffuser for CO2 removal and H2O saturation, pH 14 B. Reaction flask with sample solution, with optional UV-radiation C. UV-lamp D. Absorption flask with fritted glass diffuser for absorption of volatilised HCN, pH 14 E. Suction flask trap F. Rotameter G. Vacuum pump Fig. 1. Apparatus for the determination of volatilisation and decomposition of cyanide with optional UV irradiation.

main stream before the tributary stream mouth and the third one after this conflux, both in a distance of about 100 m to the tributary stream mouth (Fig. 2.). The sampling area was located downstream outside of the town. 2.3. Cyanide analysis Analysis of the CN-content was done by titration against AgNO3 with p-dimethylaminobenzalrhodanine, (all reagents are Merck p.a.). Using a 10 mL burette, half droplets with 0.02 mL volume could be produced during titration as a minimum. The temperature in the laboratory was constant at 23.5 °C (±0.4 °C). The influence of the temperature on the described analytical method was determined by measuring the CN content at a temperature range 13 °C to 33 °C. No difference was noticed in

this range. All samples were measured within minutes after collection. The amount of photolysis and oxidation of HCN was determined in comparing the HCN loss in solution and the absorbed cyanide as a result of volatilisation. A difference between the sum of these values and the initial concentration of cyanide in the solution should give an argument for the degradation of HCN as a result of oxidation, photolysis or hydrolysis to ammonia. In order to ensure that measuring equipment is not destroyed over the period of field investigations during transportation through rough terrain, the titration was done with 10 mL and 1 mL pipettes depending on the expected cyanide content in the samples obtained from the air experiments. As preliminary experiments showed low cyanide absorption rates in leaching solutions of a pilot plant at the Universidad Nacional de Colombia the detection limit, using a 1 mL graduated

Fig. 2. Sampling in the Quebrada la Cianurada (Location before contributing mine stream).

A. Brüger et al. / Science of the Total Environment 627 (2018) 1167–1173

1171

Fig. 3. Volatilisation of HCN in water at pH 9 in % of the initial concentration. Squares: measured HCN concentration in the gas phase identified via absorption in NaOH (HCNg) as a function of volatilisation time t. Solid line: regression for 1st order kinetic (1-exp(−t/t0)) with the time constant (t0) of 7,24 s.

pipette was tested. Droplets of 0.01 mL volume could be produced on the wall of the titration flask and shaken into the cyanide solution. According to calculations, 1 ppm CN is equal to 1.064 mg/m3 at 30 °C. With a gas flow rate of 12 L/h in an absorption flask with a liquid volume of 200 mL, this correlates to 0.013 mL AgNO3 solution for an absorption period of 2 h and 0.006 mL for a duration of 1 h. For a more accurate investigation of the detection limit, the titrant was diluted 1:10 and mixed with cyanide-solutions with concentrations ranging 0.1 to 0.5 mg/L in further preliminary laboratory experiments. These tests confirmed a detection limit of 0.1 mg/L (equates to 2 ppm with 1 h absorption time and 1 ppm with 2 h absorption time in our experiments). For more precise data, the analysis method would have to be changed towards a more complex or location-dependent procedure. The analysis of samples from the river was done 4 h after sampling and storing in glass bottles. The dark coloured bottles were completely filled with the samples and stored cold (10 °C) in the dark until the analysis was done. For the examination of the stability of cyanide solutions, handled in this way, laboratory tests were done using cyanide solutions (40 mg/L) in tap water, kept in the same glass bottles under similar conditions. The cyanide content decreased b0.1%/h in durations up to 10 h and was consequently neglected.

3. Results and discussion 3.1. CN-volatilisation/oxidation The oxidation versus volatilisation of cyanide by atmospheric oxygen was followed in two experimental strategies.

With cyanide solutions at a constant pH of 9 over a volatilisation period of 1 h to 96 h, it can be shown that the curve in Fig. 3 exhibits first order kinetics. In these experiments, besides the volatilisation of HCN, no further reaction was found. In each volatilisation experiment without further treatment (e.g. UV irradiation), the amount of HCN loss in the buffered solution was equal to the HCN content in the absorption flask. The second route was followed in investigating the HCN content in the gas phase surrounding cyanide solutions in a range of pH 9 to 14. The duration of the experiments was adapted to the extent of cyanide volatilisation from solutions over this pH range – the longest volatilisation experiment was completed after 13 days. Again, the amount of HCN loss in the solution was equal to the volatilised HCN content in the absorption flask. For a plot of comparable results, the half-life of HCN in the buffered solutions was calculated using first order kinetics. The results are summarized in Table 1 – measurements at pH 14 didn't result in detectable HCN absorption from the gas phase. The plot of the HCN half-life in solution against the pH with its logarithmic behavior is shown in the supplementary information (Fig. S 3).

3.2. UV irradiation Photolysis represents another possible degradation reaction besides oxidation. Table 2 depicts the results of experiments with three different irradiation periods. Similar to the results of the oxidation experiments, the HCN loss in solution was equal to the amount of absorbed cyanide as a consequence of volatilisation. The amount of volatilised HCN was in good agreement with data found in the volatilisation experiments. The half-life calculated from the data of Table 2 is 95,5 h (SD = 2,2 h). The total amount

Table 1 Half-life (thl) and time constant of HCN in buffered solutions at different pH (average value ± standard deviation). pH

14

13

12.6

12

11

10.5

10

9.5

9

thl t0

n. d. n. d.

440 ± 50 d 640 ± 70 d

176 ± 16 d 254 ± 23 d

53 ± 6 d 76 ± 8 d

238 ± 17 h 343 ± 24 h

107 ± 5 h 150 ± 6 h

38.4 ± 1,4 h 55.4 ± 2,3 h

11.6 ± 1,2 h 16.7 ± 1,7 h

5.0 ± 0,5 h 7.2 ± 0,7 h

1172

A. Brüger et al. / Science of the Total Environment 627 (2018) 1167–1173

Table 2 HCN decrease in a buffered solution (pH: 10.3) and HCN gas absorption after UV irradiation. HCNl: concentration in the buffered solution, HCNg: concentration in the gas phase identified via absorption in NaOH. HCNl + HCNg: total amount of HCN after irradiation, c0: initial concentration of cyanide in the buffered solution. Time

HCNl/mg·L−1

HCNg/mg·L−1

Volatilisation/%

HCNl + HCNg/mg·L−1

c0/mg·L−1

24 h 48 h 72 h

334 227 237

65 120 160

16.2 30.1 40.3

399 397 397

400 398 397

of HCN did not change during the experiments using UV light for up to 72 h irradiation time, no HCN degradation via photolysis was found in these experiments. 3.3. Ambient air measurements in Segovia The measurements of the air surrounding gold processing facilities were done at three locations, where cyanide solutions are stored in open tanks and freely accessible. The results as well as the determined concentrations of the cyanide solutions in the open tanks with their pH are listed in Table 3. Hence with this method it was shown that the TWA limit of 0,9 ppm cyanide is markedly exceeded at all measured areas. A more precise analysis would be permitted in the extension of the absorption time. The data collected from the ambient air of small scale goldmines in Segovia as well as the data found for large scale goldmines confirm the results from laboratory experiments. Further measurements will follow to determine the extent of the area, where values higher than 1 ppm cyanide can be found and for the application of our method for the determination of CN concentrations in the air surrounding large scale mines. It is expected that measurements near leach heaps will yield higher cyanide concentrations due to the higher contact area of cyanide with the surrounding air and the crushed ores in accordance with measurements already done in large scale mines. CN concentrations of 5 ppm or higher in large scale goldmines should conduct research regarding the cause of the HCN-emission sources and lead to the systematic reduction of contamination sources for the protection of the workers. 3.4. Water analysis in Segovia Samples from the Quebrada la Cianurada that collects most of the drains of the town were taken at three different locations. The cyanide level of the mine stream was considerably higher than the values found in the Quebrada la Cianurada. However, this cyanide level did not change the concentration of the main stream evidently (Table 4). The high level of cyanide concentration in the main stream and a comparatively low discharge of the side stream are considered to be the main reasons for these results. The existing results indicate high cyanide concentrations in the drains of the gold processing facilities. The impact of the cyanide volatilisation along the stream could not be determined due to the small number of measuring points. The high CN-concentration at the location before the contributing mine stream are caused by the processing plants predominantly found inside the town. Only a few gold mines have their own processing plants in their mining area. Since the sampling area was downstream outside the town, the Quebrada la Cianurada is already fed by the discharge from the processing plants of the town when it passes the area where the samples were taken. The results of cyanide analysis in stream water revealed very high concentrations, compared to common regulations. According to the requirements of the International Cyanide Management Code standards, cyanide process solution discharges to surface waters should not exceed 0.5 mg/L WAD cyanide nor result in a concentration of free cyanide in excess of 0.022 mg/L within the receiving surface water body, and downstream of any mixing zone (ICMI, 2009).

The cyanide concentrations found in the Quebrada la Cianurada are comparable to results from samples of draining waters with similar pH found in a gold mine in Sardinia/Italy, that receive seepage from the tailings impoundment (Da Pelo et al., 2009). The persistence of free cyanide in stream water has to be considered in the assessment of potential sources for cyanide contaminations in a wide area. The reduction of the HCN concentration cannot be reduced to levels below regulatory concern as a result of volatilisation under these conditions (Logsdon et al., 1999). However, for an advanced investigation, sampling at different side streams as well as the screening of the main stream on different locations has to be done. These results will enable us to provide the volatilisation rate of cyanide for this river.

4. Conclusions The often stated assumption that natural oxidation and photolysis would reduce the cyanide content in air to harmless values could not be confirmed in areas, where workers have to operate with cyanide solutions. In laboratory experiments it was shown that the volatilisation is the only process in the cyanide decline of pH buffered solutions in experimental periods up to 13 days. The described method is suitable for the application by local employees in processing facilities for the determination of the cyanide pollution in the atmosphere. An air pump together with an absorption flask and a flow meter could be combined with an existing cyanide-analysis equipment (existent in many gold processing facilities) to identify areas with possible health risks for the workers. The results of the cyanide analysis in the creek show that the concentrated drains from the surrounding processing facilities are too high to be brought to acceptable levels by dilution from the creek or volatilisation. For a comprehensive assessment of the contamination potential of cyanide leaching a mass balance of cyanide is inevitable. The analysis of cyanide volatilisation in any process involving this compound in addition to existing data of the leaching process could lead to these results. Estimations, without these considerations seem to represent an important factor for a widespread existence of different standards concerning contamination limits. The localization and collection of data for any relevant contamination source would be an important contribution to minimize controversies regarding the hazardousness of the cyanide leaching process and to optimize the process operation with a view to minimal contamination. In small-scale as well as in large-scale gold mining, conflicts as a result of different attitudes in gold mining especially related to cyanide leaching can result in substantial expenses and time delays in the workflow of gold processing. Table 3 Cyanide concentration in the air surrounding three processing tanks and CN concentration in the solutions in the tanks. Containing cyanide leaching solutions.

Entable JM Mina La Campana Mina Cogote

pH in leaching solution

Absorption time/h

CN conc. in leaching solution/g·L−1

CN conc. in air/ppm

12.5 12.3

1 1

0.83 1.04

5 5

11.6

2

1.32

4

A. Brüger et al. / Science of the Total Environment 627 (2018) 1167–1173

1173

Table 4 Cyanide concentration in the Quebrada la Cianurada.

Cyanide conc./mg·L−1 pH

Location before contributing mine stream

Mine stream (tributary stream)

Location after contributing mine stream

17 6.5

30 8.2

18 7.2

Discussions with mine workers in Segovia have indicated a great interest in the collection of data from air and water contaminations in common with their readiness to reduce the contaminant loads. These findings should provide information for a revaluation of the risks and improvements of the cyanide leaching process in the context of an upcoming interest in an advanced cyanide leaching technology in small-scale and large-scale gold mining as well as initiatives to protect workers, surrounding population and environment from contaminations as a result of it. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.scitotenv.2018.01.320. References Akinrele, I.A., 1986. Hydrocyanic acid hazard during large scale cassava processing. Trop. Sci. 26, 59–65. Bates, R.G., Bower, V.E., 1956. Alkaline solutions for pH control. Anal. Chem. 28, 1322–1324. Bustamante, N., Danoucaras, N., McIntyre, N., Díaz-Martínez, J.C., Restrepo-Baena, O.J., 2016. Review of improving the water management for the informal gold mining in Colombia. Revista Facultad de Ingeniería Universidad de Antioquia, pp. 163–172. CDPH -Colorado Department of Public Health and Environment, 2010. Summitville mine superfund site, 2010. Five-Year Review Report, Rio Grande County, Colorado. Cicerone, R.J., Zellner, R., 1983. The atmospheric chemistry of hydrogen cyanide (HCN). J. Geophys. Res. 88, 10689. Cordy, P., Veiga, M.M., Salih, I., Al-Saadi, S., Console, S., Garcia, O., et al., 2011. Mercury contamination from artisanal gold mining in Antioquia, Colombia: the world's highest per capita mercury pollution. Sci. Total Environ. 410–411, 154–160. Da Pelo, S., Musu, E., Cidu, R., Frau, F., Lattanzi, P., 2009. Release of toxic elements from rocks and mine wastes at the Furtei gold mine (Sardinia, Italy). J. Geochem. Explor. 100, 142–152. DeVries, F.W., 1996. On the Atmospheric Non-Impact of Cyanide Releases. Randol International, pp. 163–166. Dhas, P.K., Chitra, P., Jayakumar, S., Mary, A.R., 2011. Study of the effects of hydrogen cyanide exposure in Cassava workers. Indian J. Occup. Environ. Med. 15, 133–136. Feigley, P.H., Heisinger, J.F., Douglass, R.J., 1984. The use of rodents in identifying and monitering potential environmental impacts from cyanide at a heap leaching facility in South Dakota. Conference on Cyanide and the Environment. Geotechnical Engineering Program. Colorado State University Fort Collins Colorado, Tucson, Arizona, pp. 175–188. García, O., Veiga, M.M., Cordy, P., Suescún, O.E., Molina, J.M., Roeser, M., 2015. Artisanal gold mining in Antioquia, Colombia: a successful case of mercury reduction. J. Clean. Prod. 90, 244–252. Golder Associates Pty Ltd, 2011. Minera Yanacocha S.R.L ICMC Recertification Audit Summary Audit Report, West Perth, Australia. Husseini, G., 2013. Focus on transportation safety within the chemical handling industry. 2nd Biennial National Safety Conference. Factories Inspectorate Department of the Ministry of Employment and the Bureau of Public Safety, Tema, Ghana. ICMI. International Cyanide Management Institute, 2009. Implementation Guidance for the International Cyanide Management Code. p. 33. IFC-CommDev/CASM/ICMM -Communities and Small-Scale Mining/International Finance Corporation‘s Oil, 2008. Gas and Mining Sustainable Community Development Fund/ International Council on Mining and Metals, Working Together: How large-scale mining can engage with artisanal and small-scale miners. It can be found at: https://www.commdev.org/wp-content/uploads/2015/06/Working-together-Howlarge-scale-mining-can-engage-with-artisanal-and-small-scale-miners.pdf. IUPAC, 2000. Compendium of Chemical Terminology. 2nd edn. (The “Gold Book”) Version: 2.3.3. (2014) vol 17. Blackwell Scientific Publications, Oxford. Ketellapper, V.L., Pendleton, J.A., Posey, H.H., Long, M.B., 1996. The Mining History and Environmental Clean-up at the Summitville Mine. Colorado Geological Survey, Denver, CO. Kjeldsen, P., 1999. Behaviour of cyanides in soil and groundwater: a review. Water Air Soil Pollut. 115, 279–307.

Kleinböhl, A., Toon, G.C., Sen, B., Blavier, J.-F.L., Weisenstein, D.K., Strekowski, R.S., et al., 2006. On the stratospheric chemistry of hydrogen cyanide. Geophys. Res. Lett. 33, L11806. Kobayashi, Y., 2012. Petrochemical downstream in Saudi Arabia. Japanese-Saudi Business Opportunities Forum. Asahi Kasei Chemicals Corp, Tokyo, Japan. Krisnayanti, B.D., Anderson, C.W., Utomo, W.H., Feng, X., Handayanto, E., Mudarisna, N., et al., 2012. Assessment of environmental mercury discharge at a four-year-old artisanal gold mining area on Lombok Island, Indonesia. J. Environ. Monit. 14, 2598–2607. Kumbnani, H.K., Singh, A., Singh, G., Parimoo, P., 1990. Limbs' malformation in southern Nigeria—garri (Manihot esculenta) a probable cause. J. Hum. Ecol. 1, 189–191. Laitos, J.G., 2013. Cyanide mining and the environment. Pace Environ. Law Rev. 30, 869–949. Loevenhart, A.S., Lorenz, W.F., Martin, H.G., Malone, J.Y., 1918. Stimulation of the respiration by sodium cyanide and its clinical application. Arch. Intern. Med. 21, 109–129. Logsdon, M., Hagelstein, K., Mudder, T., 1999. The Management of Cyanide in Gold Extraction. International Council on Metals and the Environment (since 2001: ICMM-International Council on Mining and Metals), Ottawa, Ontario CANADA, p. 44. Mcgroup -MRCL, 2014. Sodium Cyanide: 2014 World Market Outlook and Forecast up to 2018. Birmingham, UK: via. http://mcgroup.co.uk/news/20141219/global-sodiumcyanide-demand-grow-stably-upcoming-years.html. Mudd, G.M., 2007. Global trends in gold mining: towards quantifying environmental and resource sustainability. Resour. Policy 32, 42–56. Mudder, T.I., Botz, M.M., 2004. Cyanide and society a critical review. Eur. J. Miner. Process. Environ. Prot. 4, 62–74. Newmont, G., 2010. Newmont Ghana Finalizes Compensation for Accidental Overflow at Ahafo Mine. Newmont Mining Corporation, Accra. NICNAS -National Industrial Chemicals Notification and Assessment Scheme Commonwealth of A, 2010. Sodium Cyanide. Sydney, Australia. Ozcan, E., Gok, Z., Yel, E., 2012. Photo/photochemical oxidation of cyanide and metal–cyanide complexes: ultraviolet A versus ultraviolet C. Environ. Technol. 33, 1913–1925. Rubo, A., Dickmann, A., Gos, S., 2000. Laboratory Simulation of HCN Emissions From Tailings Ponds. A. A. Balkema, pp. 307–313. Schulz, V., Gross, R., Pasch, T., Busse, J., Loeschcke, G., 1982. Cyanide toxicity of sodium nitroprusside in therapeutic use with and without sodium thiosulphate. Klin. Wochenschr. 60, 1393–1400. SCOEL -Scientific Committee on Occupational Exposure Limits, 2010. Recommendation From the Scientific Committee on Occupational Exposure Limits for Cyanide (HCN, KCN, NaCN). European Commission Employment, Social Affairs and Inclusion, p. 16. Shandro, J.A., Veiga, M.M., Chouinard, R., 2009. Reducing mercury pollution from artisanal gold mining in Munhena, Mozambique. J. Clean. Prod. 17, 525–532. Shehong, L., Baoshan, Z., Jianming, Z., Xiaoying, Y., 2005. The distribution and natural degradation of cyanide in goldmine tailings and polluted soil in arid and semiarid areas. Environ. Geol. 47, 1150–1154. Simovic, L., 1984. Kinetics of Natural Degradation of Cyanide from Gold Mill Effluents. Faculty of Graduate Studies. McMaster University, Hamilton, Ontario, USA, p. 369. SMWW APHA, 2012. 4500-CN− cyanide. In: Rice, E.W. (Ed.), Standard Methods for the Examination of Water and Wastewater. American Public Health Association, Washington, DC. SWA -Safe Work Australia, 2013. Guide for Preventing and Responding to Cyanide Poisoning in the Workplace. Commonwealth of Australia. Telmer, K., Veiga, M., 2009. World emissions of mercury from artisanal and small scale gold mining. In: Mason, R., Pirrone, N. (Eds.), Mercury Fate and Transport in the Global Atmosphere. Springer, US, pp. 131–172. TRI -National Safety Council - Toxics Release Inventory, 1998. TRI/Right-To-Know Communications Handbook - Chemical Fact Sheet. Environmental Health Center - A Division of the National Safety Council, Washington DC, p. 40. Trans-Siberian Gold plc, 2014. Annual Report and Accounts 2013, Cambridge, UK. UNEP -United Nations Environment Program, 2013. Minamata Convention on Mercury. Nairobi, Kenya. UNEP -United Nations EP, 2012. Reducing Mercury Use in Artisanal and Small-Scale Gold Mining A Practical Guide. United Nations Environment Programme, Geneva, Switzerland https://wedocs.unep.org/bitstream/handle/20.500.11822/11524/reducing_mercury_artisanal_gold_mining.pdf?sequence=1&isAllowed=. UNIDO -United Nations Industrial Development Organization, 2004. Protocols for Environmental and Health Assessment of Mercury Released by Artisanal and SmallScale Gold Miners. Vienna, Austria. USEPA -U.S. Environmental Protection Agency, 1994. Technical Report - Treatment of Cyanide Heap Leaches and Tailings, EPA 530-R-94-037, Washington, DC. USEPA -U.S. Environmental Protection Agency, 2010. Toxicological Review of HCN and CN Salts, Washington, DC.