Pierre Gélinas

Multiphase transfer processes in waste rock piles producing acid mine drainage: 1: Conceptual model and system characterization

Acid mine drainage (AMD) results from the oxidation of sulfides, mainly pyrite, present in mine wastes, either mill tailings or waste rock. This is the first of two papers describing the coupled physical processes taking place in waste rock piles undergoing AMD production. Since the oxidation of pyrite involves the consumption of oxygen and the production of heat, the oxidation process initiates coupled processes of gas transfer by diffusion and convection as well as heat transfer. These processes influence the supply of oxygen that is required to sustain the oxidation process. This first paper describes a general conceptual model of the interaction of these coupled transfer processes. This general conceptual model is illustrated by the physicochemical conditions observed at two large sites where extensive characterization programs revealed widely different properties. The South Dump of the Doyon mine in Canada is permeable and has a high pyrite oxidation rate leading to high temperatures (over 65 degrees C), thus making temperature-driven air convection the main oxygen supply mechanism. The Nordhalde of the Ronnenberg mining district in Germany contains lower permeability material which is less reactive, thus leading to a more balanced contribution of gaseous diffusion and convection as oxygen supply mechanisms. The field characterization and monitoring data at these sites were thoroughly analyzed to yield two coherent sets of representative physical properties. These properties are used in the second paper as a basis for applications of numerical simulation in AMD-producing waste rock piles.

Geochemical characterization of acid mine drainage from a waste rock pile, Mine Doyon, Québec, Canada

Water quality in the unsaturated and saturated zones of a waste rock pile containing sulphides was investigated. The main objectives of the project were (1) the evaluation of geochemical trends including the acid mine drainage (AMD)-buffering mechanism and the role of secondary minerals, and (2) the investigation of the use of stable isotopes for the interpretation of physical and geochemical processes in waste rock. Pore water in unsaturated zone was sampled from suction lysimeters and with piezometers in underlying saturated rocks. The investigation revealed strong temporal (dry period vs. recharge period), and spatial (slope vs. central region of pile) variability in the formation of acid mine drainage. The main secondary minerals observed were gypsum and jarosite. There was a higher concentration of gypsum in solid phase at Site TBT than at Site 6, suggesting that part of the gypsum formed at Site 6 in the early stage of AMD has been already dissolved. Formation of secondary minerals contributed to the formation of AMD by opening of foliation planes in waste rock, thus increasing the access of oxidants like O2 and Fe3+ to previously encapsulated pyrite. The behavior of several dissolved species such as Mg, Al, and Fe2+ can be considered as conservative in the leachate. Stable isotopes, deuterium and 18O, indicated internal evaporation within the pile, and were used to trace recharge pulses from snowmelt. Isotope trends for 34S and 18O(SO4) indicated a lack of sulfate reduction and zones of active oxidation of pyrite, respectively. Results of numerical modeling of pyrite oxidation and gas and water transport were consistent with geochemical and isotopic trends and confirmed zones of high evaporation rate within the rock pile close to the slope. The results indicate that physical and chemical processes within the pile are strongly coupled and cannot be considered separately when oxidation rates are high and influence gas transport as a result of heat generation.

Gas production and migration in landfills and geological materials

Landfill gas, originating from the anaerobic biodegradation of the organic content of waste, consists mainly of methane and carbon dioxide, with traces of volatile organic compounds. Pressure, concentration and temperature gradients that develop within the landfill result in gas emissions to the atmosphere and in lateral migration through the surrounding soils. Environmental and safety issues associated with the landfill gas require control of off-site gas migration. The numerical model TOUGH2-LGM (Transport of Unsaturated Groundwater and Heat-Landfill Gas Migration) has been developed to simulate landfill gas production and migration processes within and beyond landfill boundaries. The model is derived from the general non-isothermal multiphase flow simulator TOUGH2, to which a new equation of state module is added. It simulates the migration of five components in partially saturated media: four fluid components (water, atmospheric air, methane and carbon dioxide) and one energy component (heat). The four fluid components are present in both the gas and liquid phases. The model incorporates gas-liquid partitioning of all fluid components by means of dissolution and volatilization. In addition to advection in the gas and liquid phase, multi-component diffusion is simulated in the gas phase. The landfill gas production rate is proportional to the organic substrate and is modeled as an exponentially decreasing function of time. The model is applied to the Montreals CESM landfill site, which is located in a former limestone rock quarry. Existing data were used to characterize hydraulic properties of the waste and the limestone. Gas recovery data at the site were used to define the gas production model. Simulations in one and two dimensions are presented to investigate gas production and migration in the landfill, and in the surrounding limestone. The effects of a gas recovery well and landfill cover on gas migration are also discussed.

Aquifer washing by micellar solutions: 1: Optimization of alcohol–surfactant–solvent solutions

Abstract Phase diagrams were used for the formulation of alcohol–surfactant–solvent and to identify the DNAPL (Dense Non Aqueous Phase Liquid) extraction zones. Four potential extraction zones of Mercier DNAPL, a mixture of heavy aliphatics, aromatics and chlorinated hydrocarbons, were identified but only one microemulsion zone showed satisfactory DNAPL recovery in sand columns. More than 90 sand column experiments were performed and demonstrate that: (1) neither surfactant in water, alcohol–surfactant solutions, nor pure solvent can effectively recover Mercier DNAPL and that only alcohol–surfactant–solvent solutions are efficient; (2) adding salts to alcohol–surfactant or to alcohol–surfactant–solvent solutions does not have a beneficial effect on DNAPL recovery; (3) washing solution formulations are site specific and must be modified if the surface properties of the solids (mineralogy) change locally, or if the interfacial behavior of liquids (type of oil) changes; (4) high solvent concentrations in washing solutions increase DNAPL extraction but also increase their cost and decrease their density dramatically; (5) maximum DNAPL recovery is observed with alcohol–surfactant–solvent formulations which correspond to the maximum solubilization in Zone C of the phase diagram; (6) replacing part of surfactant SAS by the alcohol n -butanol increases washing solution efficiency and decreases the density and the cost of solutions; (7) replacing part of n -butanol by the nonionic surfactant HOES decreases DNAPL recovery and increases the cost of solutions; (8) toluene is a better solvent than D -limonene because it increases DNAPL recovery and decreases the cost of solutions; (9) optimal alcohol–surfactant–solvent solutions contain a mixture of solvents in a mass ratio of toluene to D -limonene of one or two. Injection of 1.5 pore volumes of the optimal washing solution of n -butanol–SAS–toluene– D -limonene in water can recover up to 95% of Mercier DNAPL in sand columns. In the first pore volume of the washing solution recovered in the sand column effluent, the DNAPL is in a water-in-oil microemulsion lighter than the excess aqueous phase (Winsor Type II system), which indicates that part of the DNAPL was mobilized. In the next pore volumes, DNAPL is dissolved in a oil-in-water microemulsion phase and is mobilized in an excess oil phase lighter than the microemulsion (Winsor Type I system). The main drawback of this oil extraction process is the high concentration of ingredients necessary for DNAPL dissolution, which makes the process expensive. Because mobilization of oil seems to occur at the washing solution front, an injection strategy must be developed if there is no impermeable limit at the aquifer base. DNAPL recovery in the field could be less than observed in sand columns because of a smaller sweep efficiency related to field sand heterogeneities. The role of each component in the extraction processes in sand column as well as the Winsor system type have to be better defined for modeling purposes. Injection strategies must be developed to recover ingredients of the washing solution that can remain in the soil at the end of the washing process. ©1997 Elsevier Science B.V.

Aquifer washing by micellar solutions : 2. DNAPL recovery mechanisms for an optimized alcohol-surfactant-solvent solution

Abstract A large sand column experiment is used to illustrate the principles of complex organic contaminants (DNAPL) recovery by a chemical solution containing an alcohol ( n -butanol), a surfactant (Hostapur SAS), and two solvents ( d -limonene and toluene). The washing solution is pushed by viscous polymer solutions to keep the displacement stable. The main NAPL recovery mechanisms identified are: (1) immiscible displacement by oil saturation increase (oil swelling), oil viscosity reduction, interfacial tension lowering, and relative permeability increase; (2) miscible NAPL displacement by solubilization. Most of the NAPL was recovered in a Winsor, type II system ahead of the washing solution. The 0.8 pore volume (PV) of alcohol–surfactant–solvent solution injected recovered more than 89% of the initial residual DNAPL saturation (0.195). Winsor system types were determined by visual observation of phases and confirmed by electrical resistivity measurements of phases and water content measurements in the oleic phase. Viscosity and density lowering of the oleic phase was made using solvents and alcohol transfer from the washing solution. Small sand column tests are performed to check different rinsing strategies used to minimize washing solution residual ingredients which can be trapped in sediments. An alcohol/surfactant rinsing solution without solvent, injected behind the washing solution, minimizes solvent trapping in sediments. More than five pore volumes of polymer solution and water must be injected after the rinsing solution to decrease alcohol and SAS concentrations in sediments to an acceptable level. To obtain reasonable trapped surfactant concentrations in sediments, the displacement front between the rinsing solution and the subsequent the following polymer solution has to be stable.

The mineralogy of sensitive clays in relation to some engineering geology problems — An overview

Abstract This paper deals with the nature of the sensitive clays of Eastern Canada. Reference is made to some engineering geology problems which illustrate the dominant role of the mineralogy in their behaviour. Properties discussed include: thixotropy, permeability of natural clay membranes, diffusion of trace elements and other chemical products, sensitivity, erodibility and landslide susceptibility, artificial cementation and stabilization. Future research avenues are indicated.

Localisation géoélectrique des biseaux salés dans la région côtière au Bénin: Exemple de terrain

RésuméSoixante treize sondages électriques de type Schlumberger ont été réalisés sur un site représentatif de la région côtière au sud du Bénin. L’interprétation multicouche des sondages, soutenue par des corrélations avec les forages existants sur le site, a permis une description plus détaillée des différentes couches géoélectriques. Dans les cordons littoraux, les horizons aquifères reconnus et individualisés confirment l’invasion saline de l’aquifère superficiel mais, ìls indiquent aussi la présence d’autres biseaux salés dans les multiples nappes de l’aquifère profond. À l’intérieur des terres, sur le « plateau de Godomey », les deux aquifères ne montrent aucun signe apparent de contamination saline.AbstractSeventy-three vertical electrical soundings with Schlumberger arrays have been carried out on a representative zone of the coastal area in Southern Bénin. A suitable combination of multilayer interpretation and correlation of some soundings with existing drilling wells have permitted more detailed description of the different geoelectrical layers. In the barrier beach, the reconnaissance of the aquifers confirms the intrusion of saline water in the superficial aquifer and also in the deeper multilayered aquifer. Far inland, on the Godomey Plateau, these two aquifers don’t show any sign of salt water intrusion.