Heated column experiments: A proxy for investigating the effects of in situ thermal recovery operations on groundwater geochemistry

Highlights

• Heated column design mimics aquifer heating from an in situ thermal recovery well.

• Dispersivity increases up to >10× when temperatures increase from 25 °C to 80 °C.

• Temperature increase affects metal(loid) concentrations within the column.

• Dominant As release mechanism is attributed to thermal desorption from Fe oxides.

Abstract

In situ thermal recovery is utilized extensively for unconventional bitumen extraction in the Cold Lake-Beaver River (CLBR) basin in Alberta, Canada. Public health concerns have been raised over potable groundwater contamination and arsenic release adjacent to these operations within the CLBR basin, which have been linked to subsurface heating of aquifer sediments. Under localized heated conditions, As-bearing aquifer sediments have been shown to undergo water-rock interactions and release constituents at near neutral pH conditions; however, release mechanisms have yet to be conclusively reported. To investigate the hydrogeochemical processes of aquifer heating and solute transport in detail, this study applies a novel heated column design to mimic saturated aquifer materials in contact with a thermal recovery well while constraining flow and geochemical conditions. Two column experiment scenarios were considered using: 1) quartz [SiO₂] sand with 0.6 wt% pyrite [FeS₂]; and 2) aquifer sediments collected from the CLBR region. Heated temperature gradients between 50 °C and 90 °C were maintained within a 0.6 m section of the 3 m column with a flow rate of one pore volume per week. During heated low oxygen (<3 mg L⁻¹) conditions, results generally show increases in pH, Al, As, B, Mn, Mo, Si and Zn concentrations within and downgradient of the column heating section. Constituent release is primarily attributed to thermal desorption from Fe oxides, clay and silicate mineral dissolution, competitive anion exchange, and increased mixing. Overall results suggest that these mechanisms are responsible for increasing constituent concentrations in groundwater adjacent to in situ thermal recovery operations.

Introduction

Proven unconventional oil reserves in Canada are primarily bituminous oil sands with an estimated volume of approximately 171 billion bbl (Dusseault, 2001; Lines, 2008; McGlade, 2012). Approximately 20% of these oil sand reserves are sufficiently shallow for surface mining extraction, with the remaining 80% located at depth. Extraction of heavy and extra-heavy oils from low-permeability units below 80 m from surface can be performed in situ with steam aided extraction processes commonly referred to as thermal recovery (Chopra et al., 2010). Two of the most extensively used thermal recovery methods include steam assisted gravity drainage (SAGD) and cyclic steam stimulation (CSS). Both methods inject pressurized steam at 200 °C – 300 °C (Fennell, 2008; Giraldo and MacMillan, 2016) into the permeable bitumen substrate, thereby lowering the viscosity of the oil and allowing the steam-oil mixture to be extracted to surface (Jiang et al., 2010).

In comparison to surface oil sands mining, in situ thermal bitumen extraction has the capacity to generally satisfy higher environmental standards by producing significantly lower quantities of solid waste such as tailings, in addition to creating much less surface disturbance. However, to access the bituminous strata, in situ operations require installation of injection and recovery wells through regional non-saline aquifers, which provide an abundant supply of potable groundwater to municipal and rural residents and the agricultural industry (Fennell, 2011). Consequently, public health concerns have been raised over the potential impacts to potable groundwater aquifers from thermal injection wells (Fennell, 2008; Javed and Siddique, 2016; Alberta Energy Regulator, 2016) as well as to surface environments from uncontrolled release of bitumen from the reservoir (Korosi et al., 2016).

Previous studies have demonstrated that heat transfer to aquifer sediments cause water-rock interactions and compositional changes in the groundwater geochemistry (Fennell, 2008; Bonte et al., 2013; Moncur et al., 2015b; Javed and Siddique, 2016). Elevated temperatures alter the thermodynamic and kinetic conditions of local aquifer systems, which can lead to increases in mineral solubility, reaction rates, and dissolved constituent concentrations (Nordstrom and Munoz, 1994; Stumm and Morgan, 1995). Subsequently, the mobility of these constituents can then become influenced by the physical transport of thermal and groundwater flow regimes (Domenico and Schwartz, 1990). Their ultimate fate is dictated by the chemical nature of the individual species as well as the downgradient geochemical conditions (Appelo and Postma, 2005). Arsenic (As) has been identified as a primary constituent of concern in these systems which has surpassed background concentrations by an order of magnitude in groundwater monitoring wells downgradient from thermal recovery operations located in the Cold Lake-Beaver River (CLBR) basin of Alberta (Fennell, 2008; Canadian Natural Resources Limited, 2006).

Arsenic concentrations greater than the 0.01 mg L⁻¹ limit for drinking water set by the World Health Organization (WHO) have been identified in natural groundwaters within Canada (Moncur et al., 2015a), Europe, Asia and the Americas (Sharma and Sohn, 2009). These elevated concentrations are generally attributed to the presence of As-bearing minerals in the aquifer and aquitard units coupled with slow groundwater flow rates, and either elevated pH conditions (>8.5) or strong subsurface reducing conditions at neutral pH (Smedley and Kinniburgh, 2002). Anthropogenic activities can also introduce As into groundwater systems and have been recorded in areas surrounding industrial sites including metal mines and tailings ponds; surface releases of geothermal waste fluids; aquifer thermal energy storage systems; pressure-treated wood facilities; and in situ thermal recovery operations (Fennell, 2008; Ilgen et al., 2011; Bonte et al., 2013; Moncur et al., 2015a).

Laboratory batch experiments and field-scale tests have investigated high temperature (50 °C – 150 °C) reactions for the Athabasca aquifer sediments and mechanisms of As release. Although all conclusions have indicated that elevated temperatures are at the root cause of As liberation, proposed mechanisms thus far have included reductive dissolution of Fe-rich smectite clay (Fennell, 2008), sulfide mineral dissolution (Moncur et al., 2015a), and ion exchange (Javed and Siddique, 2016). Based on these findings, there remains a gap in understanding the hydrogeochemical and reactive transport processes that are responsible for producing elevated As concentrations from SAGD and CSS operations. To examine these processes in greater detail, this study applies two heated column experiment scenarios using: 1) quartz [SiO₂] sand with 0.6 wt% pyrite [FeS₂] and 2) aquifer sediments collected from the Empress Formation within the CLBR region. Together these experiments mimic the conditions of a saturated aquifer unit in contact with an in situ thermal recovery well and help to elucidate the controlling geochemical and transport behavior of dissolved constituents under localized elevated temperature conditions. Particular focus is directed towards investigating thermodynamic and kinetic water-rock interactions, as well as the mechanisms of temperature dependent solute release, migration and fate in saturated media.