The oil sands operations in northeast Alberta, Canada, employ unconventional processes to produce synthetic crude oil (SCO). Because the extracted bitumen, ¡®the form of oil in oil sands¡¯, is highly viscous, it requires thermal upgrading to produce SCO. Coking technology is used to convert heavy bitumen fractions to lighter volatile fractions. During this process, an enormous volume of solid coke is produced and the metal impurities (e.g. Al, Fe, Mn, Ni, Ti and V) present in bitumen fractions end-up in the coke particles. As coke demands significant space for storage, oil sands companies are exploring options for placing coke into reclamation landscapes for long term storage and recovery. However, coke holds appreciable amounts of potentially leachable metals that may impede the performance of reclamation landscapes. Although two previous coke leaching studies had showed that coke released metals into water at concentrations exceeding the Canadian guidelines for the protection of aquatic life, the ecotoxicological hazard and risk of these metals were not well characterized. Therefore, the overall goal of this research was to characterize the fate and toxicity of metals associated with coke. In this research, the toxicity of coke leachates collected from oil sands field sites and those artificially generated in the laboratory were evaluated using a standard three-brood Ceriodaphnia dubia tests. Coke leachates (CLs) collected over a period of 20 months from two field lysimeters were found to be acutely toxic to C. dubia. Vanadium concentrations were significantly higher (p¡Ü0.05) than concentrations of all other metals in CLs from both lysimeters, and also in leachates from a laboratory batch renewal leaching study. Furthermore, toxic unit (TU) calculations suggested that Ni and V were likely the cause of CL toxicity, but this was not explicitly proven. Therefore, a chronic toxicity identification and evaluation (TIE) approach was adopted to identify and confirm the cause(s) of CL toxicity. Coke was subjected to a 15 day batch leaching process in the laboratory at pH 5.5 and 9.5 in order to characterize the effect of pH on metals release from coke, and to generate CLs for use in TIE tests. The 7-day LC50 estimates for C. dubia survival were 6.3% and 28.7% (v/v) for CLs generated at pH 5.5 and 9.5, respectively. The dissolved concentrations of Mn, Ni and Zn were high (p¡Ü0.05) in pH 5.5 CL, whereas Al, Mo and V were high (p¡Ü0.05) in pH 9.5 CL. Evidence gathered from a series of chronic TIE tests revealed that Ni and V were the cause of toxicity in pH 5.5 CL, whereas V was the primary cause of toxicity in pH 9.5 CL. Further, the influence of bicarbonate, chloride and sulfate ions on metals release, speciation and Ni and V toxicity was investigated. The type and amount of metals released from coke was significantly influenced by the ion type elevated in the leaching solution. Specifically, sulfate influenced mobilization of Ni, Fe, Mn and Zn from coke, whereas bicarbonate enhanced Al, Mo and V releases from coke. With respect to toxicity, increasing bicarbonate decreased the 7-day Ni2+ IC50 from 6.3 to 2.3 ¦Ìg Ni2+/L suggesting enhanced Ni toxicity at high pH or alkalinity. Conversely, sulfate showed a protective effect against V toxicity to C. dubia. The research presented in this thesis suggests that coke will not be inert when stored in reclamation landscapes and that metals, particularly Ni and V, could reach ecotoxicologically relevant levels in surface waters or substrate porewaters, under favourable leaching conditions. Operationally, efforts should focus on remediation and monitoring of metals released from coke, particularly Ni and V, in impacted wetlands, especially before discharging water into natural wetlands and/or local streams and rivers.

The oil sands operations in northeast Alberta, Canada, employ unconventional processes to produce synthetic crude oil (SCO). Because the extracted bitumen, ¡®the form of oil in oil sands¡¯, is highly viscous, it requires thermal upgrading to produce SCO. Coking technology is used to convert heavy bitumen fractions to lighter volatile fractions. During this process, an enormous volume of solid coke is produced and the metal impurities (e.g. Al, Fe, Mn, Ni, Ti and V) present in bitumen fractions end-up in the coke particles. As coke demands significant space for storage, oil sands companies are exploring options for placing coke into reclamation landscapes for long term storage and recovery. However, coke holds appreciable amounts of potentially leachable metals that may impede the performance of reclamation landscapes. Although two previous coke leaching studies had showed that coke released metals into water at concentrations exceeding the Canadian guidelines for the protection of aquatic life, the ecotoxicological hazard and risk of these metals were not well characterized. Therefore, the overall goal of this research was to characterize the fate and toxicity of metals associated with coke. In this research, the toxicity of coke leachates collected from oil sands field sites and those artificially generated in the laboratory were evaluated using a standard three-brood Ceriodaphnia dubia tests. Coke leachates (CLs) collected over a period of 20 months from two field lysimeters were found to be acutely toxic to C. dubia. Vanadium concentrations were significantly higher (p¡Ü0.05) than concentrations of all other metals in CLs from both lysimeters, and also in leachates from a laboratory batch renewal leaching study. Furthermore, toxic unit (TU) calculations suggested that Ni and V were likely the cause of CL toxicity, but this was not explicitly proven. Therefore, a chronic toxicity identification and evaluation (TIE) approach was adopted to identify and confirm the cause(s) of CL toxicity. Coke was subjected to a 15 day batch leaching process in the laboratory at pH 5.5 and 9.5 in order to characterize the effect of pH on metals release from coke, and to generate CLs for use in TIE tests. The 7-day LC50 estimates for C. dubia survival were 6.3% and 28.7% (v/v) for CLs generated at pH 5.5 and 9.5, respectively. The dissolved concentrations of Mn, Ni and Zn were high (p¡Ü0.05) in pH 5.5 CL, whereas Al, Mo and V were high (p¡Ü0.05) in pH 9.5 CL. Evidence gathered from a series of chronic TIE tests revealed that Ni and V were the cause of toxicity in pH 5.5 CL, whereas V was the primary cause of toxicity in pH 9.5 CL. Further, the influence of bicarbonate, chloride and sulfate ions on metals release, speciation and Ni and V toxicity was investigated. The type and amount of metals released from coke was significantly influenced by the ion type elevated in the leaching solution. Specifically, sulfate influenced mobilization of Ni, Fe, Mn and Zn from coke, whereas bicarbonate enhanced Al, Mo and V releases from coke. With respect to toxicity, increasing bicarbonate decreased the 7-day Ni2+ IC50 from 6.3 to 2.3 ¦Ìg Ni2+/L suggesting enhanced Ni toxicity at high pH or alkalinity. Conversely, sulfate showed a protective effect against V toxicity to C. dubia. The research presented in this thesis suggests that coke will not be inert when stored in reclamation landscapes and that metals, particularly Ni and V, could reach ecotoxicologically relevant levels in surface waters or substrate porewaters, under favourable leaching conditions. Operationally, efforts should focus on remediation and monitoring of metals released from coke, particularly Ni and V, in impacted wetlands, especially before discharging water into natural wetlands and/or local streams and rivers.

Contaminated sediments are periodically subjected to resuspension processes during either storm events or due to dredging. In sediments, metals are often contained in insoluble low bioavailability forms. Upon resuspension, however, biogeochemical processes associated with the exposure to more oxic conditions may lead to transformation and release of the metals, giving rise to exposure and risk in the water column. Batch experiments suggested that oxidation of reduced species and corresponding pH decrease were the most importance processes controlling metals release upon sediment resuspension. A mathematical model was implemented to better understand the complex underlying biogeochemical reactions that affect metals release. The model described the metals dynamics and other inter-related important biogeochemical factors well and was successful at predicting the metals release from different sediment reported in the literature. Tidal and other cyclic variations in oxygen, pH and other relevant parameters in the overlying water may also lead to cyclic transformations and release of metals from surficial sediments. In simulated estuarine microcosm experiments, cyclic variations in pH and salinity due to freshwater/saltwater exchange were shown to lead to cyclic variations in metal release. Both pH and salinity were important factors controlling interstitial dissolved metals concentrations, however, in terms of freely dissolved metals concentrations, which have been considered to be more related with toxicity and bioavailability, pH was the single most important parameter. The mathematical model was extended to the conditions of the cyclic behavior in an estuary and successfully described metals release under such conditions. It is believed that the model can be used to predict the metals behavior in other sediments and conditions by model calibration with a similar experimental approach to that used in this study. / text