Factors influencing non-aqueous extraction process of bitumen to mitigate the environmental impacts

Highlights

• Non-aqueous extraction process and engineering unit were developed, which uses much less water and reduce the release of major hazardous chemicals to oil sands tailings.

• The Factors influencing the process are studied: grades of ores (high, medium and low-grade); different solvents (toluene, pentane and other mixed solvents), and different S/Bs were studied.

• The higher-grade ore contains more heavy bitumen, the higher mitigating of the environmental impacts.

• The recovery rate of cyclopentane is higher than that of toluene.

• Mitigating the environmental impact effects were studied experimentally, efficiency of bitumen recovery was able to reach over 99 % wt %.

Abstract

Environmental pollutants' transport through tailing extraction process is a big challenge. Hazardous materials in tailings ponds increase with high negative environmental impacts. In this paper, a new type of bench no-water-extracted bitumen process was developed. Factors such as three oil sand ores, namely high, medium, and low-grade in the Athabasca deposits have been exposed to five different light hydrocarbon solvents. The bitumen recovery efficiency of the novel system was investigated with respect to solvent-to-bitumen ratios (S/Bs) as well as solvent and ore properties. The results showed that this novel bench-scale non-aqueous extraction process could significantly reduce water consumption and improve solvent recycling and reduce the release of major hazardous chemicals to oil sands tailings water. The new solvent recovery and bitumen quality for all treated ores were as good as the outcomes from hot water leaching and naphtha froth treatment. When low boiling point solvents (especially cyclopentane) are used to treat high grade ores, the efficiency of bitumen recovery was able to reach over 99 % wt % (by weight %). At the same time, the solvent loss rate could be controlled below 4 barrels (bbl) of solvent per 1000 bbl of extracted bitumen (0.4 %). This indicates the evident effectiveness of non-aqueous extraction of bitumen to mitigate the environmental impacts.

Introduction

Alberta's total oil reserves are about 171 billion barrels, almost all of which are contained in the Athabasca oil sands (Al-dahhan and Mahmood, 2019; Masliyah et al., 2008; Long et al., 2007). These reserves could provide a reliable source of energy for North America. Unfortunately, those tailings ponds contained thousands of tonnes of various undesirable materials such as dissolved chemicals including some toxic organic compounds. Furthermore, tailings from oil sands can contain more hazardous for environment materials such as napthenic acids, polycyclic aromatic hydrocarbons, phenolic compounds, ammonia, lead, mercury, and other metals (Rosa et al., 2017). For large oil sands deposits, most studies on the extraction of bitumen from oil sands are carried out based on the hot water extraction process (HWEP) (Hooshiar et al., 2012; Long et al., 2007). For example, Clark hot water extraction (CHWE) process has been used to produce commercial bitumen from Athabasca oil sands for more than 40 years (Zhou et al., 2004). Even so, the water-based technology faces many challenges: i) the process is water-intensive (three to four barrels of water are needed to produce a barrel of asphalt) (Yuan et al., 2019a, b); ii) every year, an amount of hazardous materials in tailings ponds increases, thus, generating serious, long-term environmental impacts (Lyu et al., 2019; Rosa et al., 2017; Yuan and Elektorowicz, 2020; Yuan et al., 2020); iii) high thermal capacity of water leads to the high energy consumption (Jensen and Hylland, 2019) and high greenhouse gas (GHG) emissions (Mahmood and Elektorowicz, 2015; Wu and Dabros, 2012). Therefore, there is an urgent need for a novel non-aqueous bitumen extraction process and other advanced technologies to mitigate the impact of such water conservation for better environmental protection.

The successful implementation of a new, environmental mitigating, technologies would significantly reduce the environmental impacts of the oil sand industry (Foght et al., 2017; Rosa et al., 2017), such as river ecology and groundwater conservation (Chawla, 2015), land reclamation and GHG reduction (Mukherjee et al., 2018; Masliyah et al., 2008). The main environmental benefits are reduction of water consumption in the process of asphalt mining, improvement of energy efficiency, and reduction of greenhouse gas emissions (Brandl et al., 2015). For example, the removal of polluted water from the Athabasca River would help to preserve the ecology (Bian et al., 2015; Huang et al., 2018). Furthermore, this novel non-aqueous bitumen extraction process will increasingly contribute to economic activity. Significant results have repercussions in the form of spin-off benefits to related industries such as Carbon tax of Canada, operation cost, and water consumption (Cañizares et al., 2009).

In this paper, some guiding documents of water source protection planning in the process of developing a new bench-scale non-aqueous bitumen extraction process are referred from the perspective of legislation, policy and planning (Bourgès-Gastaud et al., 2017). For example, the Canadian Council of Ministers of the Environment (CCME) has published a guide to source-to-source: a multi-barrier approach to safe drinking water (CCME, 2004), where CCME guidance has intended to help municipalities in development of water conservation plans (CCME, 2002). The American Water Association (AWWA) has also developed water conservation management standards to guide municipalities to develop water conservation plans (AWWA, 2014). Therefore, from the perspective of policy, the implementation of a non-aqueous asphalt mining technology would significantly reduce the environmental impact by decreasing water consumption and delimitation of tailings ponds (Mukherjee et al., 2018).

Compared with the traditional HWEP, the non-aqueous bitumen extraction process can significantly reduce the use of water, eliminate tailings pond and reduce the intensity of greenhouse gases. Thus, the implementation of a non-aqueous bitumen extraction technology would significantly decrease the environmental impacts due to improvement in water management (Mukherjee et al., 2018). Yet, studies on bitumen extraction based on non-aqueous solvents are few and done only most recent(Liu et al., 2019; Pal et al., 2015; Andersson et al., 2016; Andy Hong et al., 2013). Since the beginning of 2012, Hooshiar et al. have investigated non-aqueous asphalt extraction processes as substitution of the CHWE system (Hooshiar et al., 2012). Although Moldrup et al. (2000) used solvents in their study; their method could not be considered as a completely non-aqueous extraction because the water was used in the agglomeration phase which has been common in traditional HWEP. Some other similar non-aqueous extraction methodology has been developed as well (Sitnikov et al., 2016). On this basis, a new bench-scale solvent extraction process has been developed in this study, where the effects of different grades of ores and solvents on non-aqueous bitumen in oil sands are investigated.

Precisely, this study aimed to develop an environmental mitigating technology for non-aqueous extraction of bitumen from oil sands, and to prove to government and industry the feasibility of its implementation. Furthermore, define main factors influencing the successful bitumen recovery. The strategic significance of this new desktop non-water asphalt mining process is reduction of water consumption in mining and economic water management, leading also to elimination of environmental impacts from tailings ponds.

Three grades of oil-sand ores were studied: i) high-grade ores containing 12.2 wt% (by weight %) bitumen, ii) middle-grade ores containing 7.4 wt% bitumen and 8.9 wt% water, iii) low-grade ores containing 6.8 wt% bitumen and 11.4 wt% water. Five different light hydrocarbon solvents were then mixed with three different ores at a relatively low solvent/asphalt ratio (S/Bs). The supplied by Fisher Scientific solvents, such as n-pentane, toluene, and n-hexane were certified by American Chemical Society (ACS). Cyclopentane (reagent grade) and octane (High-Performance Liquid Chromatography, HPLC grade) were purchased from Sigma-Aldrich Company. Centrifugal filtration and conventional pressure filtration are used to separate solute bituminous solutions from solids (Bell, 1967). The residual solvent in the filter cake was recovered at room temperature by vacuum evaporation, and then, the solvent vapor was discharged through the condenser (Brandl et al., 2015). Finally, the recovery rate of bitumen as well as a quality of recovered bitumen under different solvent and ore conditions were evaluated and compared. The effects of the mineral grade and type of solvent on the non-aqueous extraction of bitumen from oil sands were discussed.

Oil sand, being a viscous semisolid substance, consists of a bitumen, water, and mineral phase such as sand and fine crystalline (e.g., clay) or/and non-crystalline forms. The composition of adopted oil sand ores originated from Athabasca deposits are shown in Table 1. The high-grade oil sands ores usually have higher bitumen content and lower clay content. The ratios of sand-to-fines obtained by Dean-Stark extraction (Robin, 1974) are also listed in Table 1.

To separate bitumen from ore, the solvent extraction method was adopted. Solvent extraction methods have been used in the industry for many years and ASTM (ASTM D5765-16) has established test methods in place. According to the standard, solvent extraction method to get the bitumen froth is based on mixing of ore (containing different bitumen content) with a particular solvent. Such mixture undergoes dilution, separation in settler, centrifuging, and often trough additional processes using a series of equipment. In this study a special device was adopted to the non-aqueous extraction. Ideal solvent properties for bitumen extraction include solubility and low boiling point, which are convenient for the solvent recovery.

Table 2 lists properties of adopted solvents such as toluene, pentane, n-hexane, cyclopentane and a mixture (1:1 wt/wt) of n-C5 and cyclopentane. Those solvents are certified by the American chemical society (ACS) and provided by Fisher. The solvents characters were determined by standard pentane precipitation and filtration methods (ASTM D2007-80) (Yuan et al., 2021). The boiling point of cyclopentane and toluene were found to be 49 and 110−111 °C, respectively. Toluene had higher water solubility than cyclopentane.

Based on the previous research work of Canmet ENERGY (Wu and Dabros, 2012), a design schematic of the device was developed. In this research, a real apparatus was designed and built to mix the solvent with oil sand and separate the solvent asphalt solution from minerals in Canmet ENERGY (Yuan and Dabros, 2013).

The schematic diagram of a device, shown in Fig. 1, consisted of two chambers (upper and bottom chamber) with a screen of 12.5 cm high and 6 cm diameter in the middle.

The filter paper was fixed on the top of the filter screen. The container was well sealed to prevent the solvent loss. The valve at the bottom of the device could be used to add solvents and extract solvent-based bituminous solutions. Pressure can also be applied through the valve on the cover. The entire device can be installed in commercial laboratory centrifuges so that solvent-based bituminous solutions can be separated from solids using controlled gravity.

The solvent was mixed with oil sand. Gravity was applied to the container using centrifuging force. The asphaltene content in asphalt was determined by Near-Infrared Reflectance Spectroscopy (NIR) probe (Zhang et al., 2016). The weight of filter cake was analyzed by GC-FID (Flame Ionization Detector), and the content of the final residual solvent in mineral solids was determined (Connor et al., 1998). The bitumen content in dried extracted solids was determined by TGA (thermogravimetric analysis) method.

The experimental procedure comprised detailed method for determine solvent content in Apparatus listed in Fig. 2. Initially, a filter paper with a pore size of 40 mm was installed on the top of the screen in the middle of the vessel. About 50 g(grams) of oil sands were loaded onto the top of the vessel, which was subsequently sealed and turned upside down. A syringe was used to fill the container with solvent through the bottom valve. A requested amount of solvent was calculated according to the target S/Bs. The solvent could penetrate the sand matrix through the separator within minutes. The container was then placed on a wrist-action shaker and stirred for about 1 h to ensure an adequate mixing of the solvent in the oil sand matrix. After stirring, the container was placed in the centrifuge and rotated for 10 min with a force of 1000 g. The solvent-based bituminous solution entered the lower part of the container through the filter paper. Subsequently, the solvent was removed from the bitumen by evaporation, then, dry asphalt was collected and weighed (Turhanen et al., 2015; Korkisch, 1966). The weight of solvent asphalt solution and dry asphalt was used to calculate S/Bs ratios at the range of 1–4. The extraction process was repeated twice. The amount of solvent used in the second and third stages was half that in the first stage (Fig. 1). Thus, the content of n-pentane asphaltene was determined by a NIR probe, while the ash content in bitumen was assessed by TGA method.

The content of the final residual solvent in mineral solids was determined by o-xylene extraction followed by GC analysis with FID detector. This method has been developed based on some previous works on solvent extraction for chromatography purpose (Hong et al., 1999; Aeppli et al., 2008). In short, mineral solids were mixed with a half amount of the xylene. After thoroughly mixing and centrifugation, the solvent was extracted into the xylene phase. The concentration of solvent in xylene phase was determined by GC analysis with n-octane as an internal standard. The content of residual bitumen was determined by TGA method. Mineral solids extracted from Dean-Stark extraction were used as reference samples. Then, the recovery rate was calculated according to the amount of recovered diluted bitumen after its separation and the residual bitumen remained in solids.