Évaluation de la performance de réacteurs passifs biochimiques à échelle pilote de terrain pour le traitement du drainage minier à forte salinité et faible température

Environmental issues associated with the mining industry are mainly related to contamination from mining effluents. At a mining site, water that accumulates in open pits or underground tunnels may be contaminated, as well as the water flowing from the tailings pond or from the waste rock dumps. Mining effluent contamination results mainly from biogeochemical processes that occur when solid mining waste (waste rocks and mine tailings) are exposed to both water and air. The chemical instability of sulfide minerals when they come in contact with water and atmospheric oxygen can generate acidity and mobilize dissolved metals/metalloids and sulfates in mine water. These contaminated waters must be collected and properly treated in order to reduce their contaminant concentrations to a level that meets environmental standards. Passive treatment systems such as Passive Biochemical Reactors (PBRs) represent an interesting approach for the treatment of mine drainage after mine closure, particularly in economic and environmental terms. Inside a PBR, the aimed treatment mechanism is the precipitation of metals as stable metal sulfides following the reduction of sulfate to hydrogen sulfide under the action of Sulfate-Reducing Bacteria (SRB). PBR systems, which have been the subject of numerous studies in recent years, have shown great promise as a biotechnology for the treatment of mine drainage in temperate or semi-arid climates. However, knowledge concerning PBR performance in northern climates is limited. The combined influence of low temperatures and the potential high salinity of the mine drainage found in northern mines on the effectiveness of PBRs remain poorly known. The main objective of this study was to evaluate the effectiveness of PBR for the treatment of mine drainage in northern climates. In order to achieve this aim, we collaborated with the Raglan mine (Glencore), which is located in the far north of Quebec and where the underground water and the water flowing from the tailings pond (before treatment) have pH values and Ni, Fe and Cu concentrations that do not meet the environmental governmental standards. In total, three pilot-scale PBRs were constructed, operated and monitored at the Raglan mine site. The PBR-AMD reactor was designed for the treatment of Acid Mine Drainage (AMD) and the PBR-CND- 1 and PBR-CND-2 reactors for the treatment of Contaminated Neutral Drainage (CND). The PBR-AMD reactor was monitored for 48 days in the summer of 2018. It was fed with a source of AMD flowing from the Raglan mine tailings pond and which had the following average physicochemical parameters: 27.6 mg/L Ni, 23.7 mg/L Fe, 2.4 mg/L Cu, 383mV ORP, 3 186 mg/L SO42- and pH = 3.7. The PBR-CND-1 and PBR-CND-2 reactors were monitored for 94 days and 44 days, respectively. The PBR-CND-1 reactor was monitored for 57 days in the summer of 2017 until the water froze. The monitoring of the PBR-CND-1 reactor resumed in the summer of 2018 for 37 days. The PBR-CND-2 reactor was monitored in the summer of 2018. Both reactors were fed with a saline Ni-CND pumped from the mine underground galleries. The CND had the following average physicochemical parameters: 25.7 mg/L Ni, 456.0 mg/L SO42-, 234 mV ORP and pH = 7.3 in the summer of 2017 and 22.5 mg/L Ni, 647.1 mg/L SO42-, 218 mV ORP and pH = 7.1 in the summer of 2018. The Hydraulic Residence Time (HRT) in the PBR-AMD reactor was HRT = 60 hours and the HRT in both PBR-CND-1 and PBR-CND-2 reactors was HRT = 24 hours. During the tests, the physicochemical parameters of the effluents at the entry and exit of the PBRs were regularly checked. The results of the tests showed that neither the PBR-AMD reactor nor the PBR-CND-1 and PBR-CND-2 reactors developed favorable conditions for SRB activity. The absence of SO42- removal, the persistent oxidizing conditions and the absence of sulfide detection at the outlet of the PBRs suggest that the sulfate-reduction process did not occur in the PBR-AMD, PBR-CND-1 and PBR-CND-2 reactors or at least was negligible. The removal of metals as metal sulfides in the three PBRs is therefore unlikely. The low temperatures recorded in the PBRs during monitoring (5°C on average) may explain why the SRB did not develop in the reactors. For the PBR-CND-1 and PBR-CND-2 reactors, the high salinity of the CND could also have had a negative impact on SRB populations. In the PBR-AMD reactor, removal of Ni (between 93% and 95% at the beginning and then between 53% and 56% at the end), Fe (between 96% and 99%) and Cu (99%) has been observed. In the absence of SRB activity, the major mechanisms governing the removal of metals in the PBR-AMD reactor are: (1) precipitation, in the form of hydrated oxy-hydroxides and carbonates, related to pH changes, (2) co-precipitation with oxy-hydroxides and (3) sorption on organic matter. The sorption and co-precipitation mechanisms are particularly important for Ni, whose stable solid phases are not expected to form at pH 7-8. The lower removal efficiency of Ni in the PBR-AMD reactor compared to Fe and Cu removal can be explained by the competition between Ca2+ ions, Fe, Cu and Ni for the sorption sites of organic matter. The bonds that Cu and Fe form with organic matter are stronger than Ni which is more mobile. Ni removal in the PBR-CND-1 reactors (99% in the summer of 2017 and 83% in the late summer of 2018) and PBR-CND-2 (between 95% and 99% at the beginning then between 61% and 83% at the end) is mainly governed by sorption mechanisms on the organic matter and on the wood ash of the reactive mixture. The decrease in Ni-removal rate over time in the PBR-AMD, PBR-CND-1 and PBR-CND-2 reactors can be attributed to the gradual saturation of active sorption sites of the reactive mixture. The PBR-AMD, PBR-CND-1 and PBR-CND-2 reactors failed to meet all the AMD and CND treatment objectives at the Raglan mine site. The limited performance of the reactors might be due to the lack of SRB activity. The use of native SRB, adapted to low temperature and high salinity, in reactive mixtures could be a solution to enhance the development and activity of SRB in PBRs and therefore increase their performance. Finally, the results of this study, combined with additional tests, will allow optimizing the design criteria for large-scale passive treatment systems at the Raglan mine and on other mining sites in the northern climate areas.