Detection of Volatile Organic Compounds in Froth ... - ACS Publications


Detection of Volatile Organic Compounds in Froth...

0 downloads 129 Views 1MB Size

Article pubs.acs.org/EF

Cite This: Energy Fuels XXXX, XXX, XXX-XXX

Detection of Volatile Organic Compounds in Froth Multiphase Systems from Oil Sands Operations Using a Headspace GC−MS Method Lisa Duffin, Xiaomeng Wang,* and Justin Stoesz Natural Resources Canada, CanmetENERGY Devon, 1 Oil Patch Drive, Devon, Alberta, T9G 1A8, Canada S Supporting Information *

ABSTRACT: Degradation of air quality due to oil sands operations is one of the largest concerns for stakeholders and regulators. Volatile organic compounds (VOCs) released from tailings ponds are an important contributor to poor air quality. Current government regulations impose a limit on hydrocarbon losses to the froth treatment tailings at 4 barrels per 1000 barrels of dry bitumen produced. However, considering the scale of bitumen production, atmospheric pollution from allowable VOC emissions is still problematic. One source of solvent loss to tailings ponds is solvent trapped in rag layers formed during froth treatment (a multiphase system that sometimes develops at the interface between the diluted bitumen and water). It would be useful to have a method for directly determining solvent loss in rag layers as support to efforts to optimize solvent recovery from froth treatment tailings. In this paper, analytical methods for the direct determination of solvent content in multiphase waste streams from oil sand froth treatment have been developed using headspace sampling combined with gas chromatographic separation and mass spectroscopic detection. The respective detection limits for heptane, toluene, octane, and p-xylene in the water layer are 0.1, 0.4, 0.03, and 0.4 ppm. The detection limits for heptane, toluene, octane, and p-xylene in the rag layer and oil are all approximately 1 wt %. The respective detection limits for naphtha in water, rag layer, and oil are 0.5 ppm, 6 wt %, and 6 wt %.

1. INTRODUCTION Oil sands operations are a large part of Alberta’s overall energy production. The most commonly used methods for oil sands extraction are surface mining and in situ operations. In 2017, surface mining production is at 1.1 million bbls/d and in situ production is at 1.4 million bbls/d.1 According to the most recent forecast from the Canadian Association of Petroleum Producers, by 2030, mining production will grow to 1.5 million bbls/d, and in situ production will increase to 2.1 million bbs/d. During surface mining operations, water-based extraction is used to recover bitumen from surface-mined oil sand ores. Specifically, mined ore is added to hot water during the extraction process. Bitumen separates from the sand and flows around air bubbles to the surface of the settling vessel to form a bitumen-enriched froth.2 The sand sinks to the bottom and is removed as the primary tailings stream. Bitumen froth is made up of approximately 60% bitumen, 30% water, and 10% solids.3,4 To reduce residual water and solid contents, the froth is diluted with a hydrocarbon solvent (i.e., diluent) to make it less dense and less viscous. As a result, solids and water can be removed more effectively by subsequent procedures. The diluted bitumen is recovered for further processing. The water and solids removed from the froth (froth treatment tailings) are sent to a diluent recovery unit to remove as much solvent as possible before the froth treatment tailings are sent to the tailings ponds. However, residual solvent now in the tailings ponds is released to the atmosphere, contributing to poor air quality.5,6 Current government regulations impose a limit on diluent losses to the froth treatment tailings at 4 barrels per 1000 barrels of dry bitumen produced.7 However, considering the scale of bitumen © XXXX American Chemical Society

production, atmospheric pollution from VOC emissions is still problematic. In addition, during storage of diluted froth, froth treatment, or when handling the froth treatment tailings, a viscous layer may form that can affect oil−water separation. This viscous layer, often called a rag layer, is a complex mixture made up mostly of water, diluted bitumen, and solids.8 Rag layers are typically stable water-in-oil emulsions or multiple emulsions that interfere with separation of the water and solids from the bitumen.9 As a result, rag layer formation and accumulation can seriously affect bitumen processing. The mechanism by which rag layers form and their gradual growth by “trapping” of water, solids, and diluted bitumen is debated in the literature.10,11,8,12 Since these complex emulsions often end up in the tailings ponds, solvents trapped in rag layers contribute to the total solvent loss from the pond,13 and since solvent loss to the atmosphere degrades air quality, it would be useful to have a method for directly and accurately determining the solvent content in rag layers. An easy-to-use analytical technique could also help in the study of ways to optimize solvent recovery from froth treatment tailings. In 2012, Gupta et al. proposed a few workable methods for residual solvent estimation in the produced heavy oil or bitumen, including use of maltenes/metals content as oil markers as well as the use of boiling-point curves of the produced blend compared with the boiling point of the base oil.14 However, these methods were all indirect measurements Received: August 5, 2017 Revised: October 3, 2017 Published: October 3, 2017 A

DOI: 10.1021/acs.energyfuels.7b02286 Energy Fuels XXXX, XXX, XXX−XXX

Article

Energy & Fuels

2.1. Major Component Analysis in Naphtha. Naphtha is widely used as a diluent in the oil sand industry. In the present study, it was analyzed by GC with a flame ionization detector (FID) using an ASTM method.18 The GC procedure was as follows: A fixed volume (1 μL) of naphtha is injected into a gas chromatograph equipped with a programmed temperature vaporized injector (200:1 split ratio), and a methyl siloxane bonded phase fused silica capillary column. The components are separated through the column and are detected by FID. Total run time is 3 h. 2.2. Rag Layer Formation Procedure. A bitumen sample was weighed into a separatory funnel, and naphtha was weighed and added to the bitumen. The solution was then thoroughly mixed by hand for 30 min so that there was a consistent homogeneous oil phase and all of the bitumen was dissolved in the naphtha. Saline water was then added to the diluted bitumen. The separatory funnel was stoppered and vigorously shaken by hand for 2 min to mix the water and diluted bitumen. The separatory funnel was then left overnight to allow the phases to separate. The three phases (water, rag layer, diluted bitumen) were removed in turn from the separatory funnel and weighed. Data are given in Table 2. 2.3. Calibration Standards for Multiphase Systems. For oil and rag layers, a standard addition method was employed due to the complex matrix effects. However, instead of using naphtha for the standard addition method for the rag layer and the diluted bitumen layer analysis, four hydrocarbons found in naphtha (toluene, heptane, octane, and p-xylene) were used to check the distribution of various VOCs among the different layers of the multiphase systems. A multiphase system made by employing a solvent-to-bitumen ratio of 0.7 was chosen for the calibrations to match industry practice of using a naphtha-to-bitumen ratio of approximately 0.6−0.7.19 Moreover, it is well-known that a solvent-to-bitumen ratio greater than 4 results in precipitation of asphaltenes.20−22 A set amount (0.3 g) of the sample was transferred into five 20 mL vials, respectively, and increasing amounts of four hydrocarbon mixtures (each 25% by weight) were added to each vial at different weights from 0 to 0.06 g. Chloroform (0.1 g) was added into the mixture as an internal standard since it is not a component of naphtha and its GC retention time is well separated from most major components of naphtha and the selected hydrocarbon standards. By using the combined standard addition and internal standard methods, we were able to examine which hydrocarbons preferentially partition into the rag layer, diluted bitumen layer, and water layer. Mass changes introduced by solvent addition (chloroform) are ignored in the present study. In order to analyze the hydrocarbon content of the water layers of the multiphase systems at a solvent-to-bitumen ratio of 0.7, standards for the water layers were made by diluting toluene-saturated water, heptane-saturated water, octane-saturated water, and p-xylenesaturated water using saline water. The concentrations of toluene, heptane, octane, and p-xylene in saturated saline water were assumed to be 455.86, 2.18, 0.47, and 123.07 ppm by weight, respectively, based on literature calculations.23,24 While these numbers are only an approximation for the solubility, this is sufficient for comparing results within the present study.

of the residual solvents and errors are often introduced in solvent estimation. In the present study, a headspace (HS) gas chromatography−mass spectrometry (GC−MS) method has been developed that can directly measure solvent content in the rag layer with minimal sample preparation. In addition, the same method was applied to measure solvent content in bitumen. Furthermore, the solvent content in water was measured by a modified HS GC−MS method.

2. EXPERIMENTAL SECTION Bitumen, solvent, and water were mixed in the amounts shown in Table 1 to create a multiphase (three-phase) system consisting of oil

Table 1. Composition of Multiphase Mixturesa composition (wt %) solvent/bitumen

saline water

solvent

bitumen

0.5

10 25 50 75 10 25 50 75

30 25 17 8 37 31 21 10

60 50 33 17 53 44 29 15

0.7

a

Naphtha used as solvent.

phase, water phase, and rag layer. Generally, the solvent-to-bitumen ratio is very important since it has been shown to influence the rigidity or flexibility of the oil/water interfacial film,15 as indicated by film rigidity K.16 A critical solvent-to-bitumen ratio is often described where values below the critical ratio produce films that are rigid, whereas, at higher ratios, the interfacial film is observed to be flexible. In the present study, four different mixtures were tested to account for different solvent/bitumen and different amounts of water in each system. The solvent used in the present study is naphtha obtained from an oil sands operation site.27 The experimental procedure was described previously in the literature.17 The saline water solution was made by dissolving 25 mmol/L NaCl, 15 mmol/L NaHCO3, 2 mmol/L Na2SO4, 0.3 mmol/L CaCl2, and 0.3 mmol/L MgCl2 in deionized (DI) water to mimic water compositions found in the oil field operations. Bitumen was extracted from a froth sample (53.43% bitumen, 21.98% water, 22.57% solids) using hot toluene extraction (Dean−Stark method), followed by rotary evaporation to remove the toluene. Bitumen was used in place of froth so that the composition could be better controlled during rag layer formation. The asphaltene content in the bitumen was determined to be 14.3 wt % by n-heptane precipitating. The detailed procedure is provided in the Supporting Information (section 1.1).

Table 2. Composition of Multiphase Layers and Masses of Each Layer mass in (g) solvent/bitumen, wt % water 0.5, 0.5, 0.5, 0.5, 0.7, 0.7, 0.7, 0.7,

10 25 50 75 10 25 50 75

mass out (g)

bitumen

naphtha

water

oil layer

rag layer

water layer

percent recovery of total system

19.71 20.77 20.22 10.42 9.97 10.05 10.05 10.00

10.17 10.09 10.37 5.29 7.04 7.03 6.88 7.30

2.94 8.95 30.09 45.60 1.87 5.69 17.02 51.01

29.54 29.52 15.42 12.58 15.88 15.02 11.85 15.48

2.58 8.12 40.98 17.77 1.79 6.20 19.29 5.02

0 0 0 30.90 0 0 0 45.79

98 95 93 100 94 93 92 97

B

DOI: 10.1021/acs.energyfuels.7b02286 Energy Fuels XXXX, XXX, XXX−XXX

Article

Energy & Fuels

Figure 1. Major components of naphtha by mass % measured by GC-FID. Some fractions were not labeled due to space limitation in the figure. For details, refer to the Supporting Information section on DHA analysis. In addition, for the solvent-to-bitumen ratio of 0.5, standard addition was also done by adding naphtha directly into the rag layer and oil layer. For the water layer, a naphtha-saturated water was created. The naphtha solubility was assumed to be 7 ppm based on the literature.25 2.4. Headspace Analysis by GC−MS. The different layers (oil, rag, water) in all of these samples were analyzed by headspace sampling using a CombiPAL system. This method requires that an analyte has significant vapor pressure between 30 and 300 °C. However, identification based only on retention time may be inaccurate or misleading. MS represents the ratio of the mass of a given particle (Da) to the number (z) of electrostatic charges (e) that the particle carries. GC−MS improves confidence in sample identification, significantly increases the range of thermally labile and low-volatility samples amenable to analysis, provides much faster analysis, and improves sensitivity. In the present study, the scan mode of GC−MS was used. For oil and the rag layer, a vial containing 0.1 g of sample was heated to 30 °C for 30 s with the agitator set to 250 rpm. The syringe was heated to 35 °C and 0.1 mL of headspace was injected to the GC. The split/splitless injector was heated to 210 °C; a split ratio of 500:1 was used; the DB5 column was held at 50 °C for 1 min and then heated at 10 °C/min to 100 °C; and the flow rate through the column was 0.8 mL/min. The MS detector scanned 50−200 amu at a rate of 8.4 scans per second; a 0.5 min solvent delay was applied. For water, 0.25 g of sample was placed in a 20 mL septa-sealed vial. The vial was heated at 80 °C for 120 s with the agitator set to 250 rpm. The syringe was heated to 50 °C and 1.0 mL of headspace was injected to the GC. The split ratio was set to 20:1. All other parameters were the same as those for the rag layer analysis. The water samples contained low levels of solvents, so the volume of headspace injected onto the GC was set to the maximum value (1.0 mL). The incubation temperature (80 °C) and duration (2 min) were set to allow a large portion of the solvent to evaporate while minimizing water vapor formation. The split ratio was reduced to 20:1 to allow as much solvent onto the column as possible without compromising the reproducibility, which degrades at low split ratios.

3. RESULTS AND DISCUSSION The aqueous phase (bottom layer) in the separatory funnel was clear and colorless. However, in six out of eight tests in this study, there was no water-continuous phase present after the system was left to separate. This was probably due to the fact that all the water in the system was trapped in the rag layer. An example of the multiphase system is shown in Figure S1 of the Supporting Information. The middle layer was the rag layer and was dark brown in color. It was the most viscous layer of the three layers formed. When a water layer was present, there was a well-defined phase separation between the water and rag layer due to the hydrophobicity of the rag layer. The top layer was the diluted bitumen or oil phase, which was black. The interface separating the oil and rag layer was not as well-defined as that between the water and rag layer, but could be seen. Karl Fischer titrations for water content were performed on selected diluted bitumen samples. The water content was found to be less than 1% by mass. Because the errors in weighing the resulting layers were much greater than this (as evidenced by the recovery percentages), the water content in the diluted bitumen phase was determined to be insignificant and no further Karl Fischer titrations were performed. 3.1. Naphtha Composition. The major components in naphtha are shown in the Figure 1. Further detailed hydrocarbon analysis (DHA) test results are given in the Supporting Information (section 1.3). Each eluting peak was identified by comparing its retention index to a library of retention indices of known standards, which has been established by running reference compounds under identical conditions. In the present study, octane, toluene, p-xylene, and heptane were selectively monitored to study their distribution in the different layers of the multiphase system. C

DOI: 10.1021/acs.energyfuels.7b02286 Energy Fuels XXXX, XXX, XXX−XXX

Article

Energy & Fuels

3.3. Hydrocarbon Determination for Different Phases for Solvent/Bitumen Ratio 0.7. On the basis of the standard addition curves (as shown in Figures 2 and 3 as well as in the Supporting Information) of the oil layer and rag layer, the concentrations of four hydrocarbons in the oil and rag layers can be determined. Then, knowing the total weight of each layer, the total mass of the four hydrocarbons in the two layers of the sample can be calculated. The results are shown in Table 4. On the basis of the water calibration curves, the concentrations of the four hydrocarbons in the water layer can be determined. In this study, only one system (solvent/ bitumen 0.7, 75 wt % water) had a water layer. The concentrations of heptane, octane, toluene, and p-xylene in the water layer of this multiphase system were determined to be 0.05, 0.31, 11.35, and 1.62 ppm. However, since the concentrations of these hydrocarbons in the water layer will always be at ppm levels, considering the larger amount of hydrocarbons in the oil and rag layers, the amount in the water layer is insignificant when conducting mass balance calculations on the hydrocarbons. Figure 6 shows the distribution of individual hydrocarbons between the oil and rag layers as a function of solvent/bitumen and water added when making the rag layers. The higher the ratio along the y-axis, the greater the degree to which solvent partitioned into the oil; the lower the ratio, the greater the degree to which solvent partitioned into the rag layer. Ignoring the water layer, with one exception (solvent/bitumen 0.7, 25 wt % water), it appears that more p-xylene partitioned into the rag layer compared to the other hydrocarbons (also as shown in Figure S8, Supporting Information). Considering that both pxylene and toluene are aromatic hydrocarbons, it is interesting to note that they partitioned differently between the oil and rag layers. As shown in Table 4, the total masses of all added hydrocarbons in the oil layers are quite similar, as are those in the rag layers for the four systems tested (0.7, 10; 0.7, 25; 0.7, 50; and 0.7, 75), showing that the amount of water added did not greatly affect the partitioning of the added solvents, and interestingly, the concentration of solvent in the rag layers formed was quite consistent at this solvent/bitumen ratio.

3.2. Calibration Curves. As mentioned in the Experimental Section, for both oil and rag layer analysis, a standard addition method was used for VOC analysis because of the complex sample matrix for these two layers. The details of the method are shown in Table 3. Table 3. Standard Addition Method for Solvent Determination in Oil and Rag Layers

a

mass of sample (g)

total mass of added solvent (g)

mass of individual solventa added (g)

mass of chloroform (internal standard) (g)

0.3 0.3 0.3 0.3 0.3

0 0.0125 0.025 0.04 0.055

0 0.00313 0.00625 0.01 0.0138

0.1 0.1 0.1 0.1 0.1

Among four solvents (toluene, heptane, octane, and p-xylene).

The calibration curve is a graph of GC-peak areas (y-axis) graphed against the masses of the individual hydrocarbons in the four-hydrocarbon mixture added into the sample (x-axis). The x-intercept was taken from the extension of the calibration curves. The absolute value of this intercept gives the mass of an individual hydrocarbon in an unknown sample in a GC vial. To obtain the total individual hydrocarbon content in an unknown sample, one needs to convert the value obtained based on the total mass of the sample. Examples of the calibration curves for oil and rag layers are shown in Figures 2 and 3, respectively. As mentioned previously in the Experimental Section, standards for the water analysis were made by diluting toluene-saturated water, heptane-saturated water, octanesaturated water, and p-xylene-saturated water with saline water. On the basis of the literature, the respective concentrations of toluene, heptane, octane, and p-xylene in saturated saline water were assumed to be 455.86, 2.18, 0.47, and 123.07 ppm by weight.23,24 Due to the ionic strength of the saline water, the solubilities of the hydrocarbons were reduced compared to their solubilities in DI water because of the “salting-out” effect.26 The calibration curves for the four hydrocarbons in water are shown in Figures 4 and 5.

Figure 2. Calibration curves of oil layer at solvent/bitumen 0.7 and 10% water. D

DOI: 10.1021/acs.energyfuels.7b02286 Energy Fuels XXXX, XXX, XXX−XXX

Article

Energy & Fuels

Figure 3. Calibration curves of rag layer at solvent/bitumen 0.7 and 10% water.

naphtha in an unknown sample in a vial. The total naphtha content in an unknown sample is calculated knowing the total mass of the sample. Standards were made by diluting naphtha-saturated water. The following concentrations were used for naphtha standards: 0, 0.7, 1.75, 3.5, 5.25, and 7 ppm. The calibration curve is shown in Figure 8. 3.5. Naphtha Partitioning in Multiphase Systems. Using the analytical method of naphtha addition, the multiphase systems described in Table 2 were studied. The amounts of solvent in the oil and rag layers are given in Table 5. The concentration of solvent in the water layer of sample (0.5, 75) is 0.3 ppm. Compared to the amount of solvent in the oil and rag layers, the amount of solvent in the water layer is negligible and is therefore not included in the recovery calculation. On the basis of the analytical methods developed in this study, it was found that most of the solvent was present in the oil phase (68−98%) and the rest of the solvent was trapped in the rag layer (2−42%). The amounts of solvent in the water were minimal (ppm levels). The recoveries for most of the runs were greater than 100% (Table 5), which indicates a significant amount of uncertainty in this analytical method. However, this analytical method is the first reported method to directly detect VOCs in the multiphase systems for oil sands samples without sample manipulation and has great potential to be applied in the future studies. The method of preparing the rag layers used here though has its limitations. Some duplicates did not have consistent compositions in terms of the weight of oil, rag, and water layers produced. This inconsistency could add uncertainty to the final results. The homogeneity of the samples also needs to be investigated as this method only uses a small amount of sample. Comparing the two standard addition methods, we conclude that the four-hydrocarbon addition method is preferred if one needs to find the individual hydrocarbon concentrations in different phases. However, if the ultimate purpose is to determine the concentrations of naphtha in the phases, the naphtha addition method should be used. Theoretically, one could determine the concentrations of the four hydrocarbons in the system based on the naphtha addition method and, in return, one could determine the concentrations of naphtha

Figure 4. Calibration curves for octane and heptane in saline water.

Figure 5. Calibration curves for toluene and p-xylene in saline water.

3.4. Calibration Curves Using Naphtha. Due to the different partitioning behaviors of the selected hydrocarbons, the weight of naphtha in the multiphase system cannot be easily extrapolated from these four hydrocarbons. As a result, using chloroform as an internal standard, examples of an oil layer standard addition curve and a rag layer standard addition curve using naphtha are shown in Figure 7. The y-axis is the ratio of the total GC peak area minus the peak area of chloroform, to the peak area of chloroform. The x-axis is the added mass of the naphtha in each of the GC vials. For these curves, it was found that a second-order polynomial fit was better than a linear fit. The intercept at the x-axis was obtained from the calibration curves. The absolute value of this intercept gives the mass of E

DOI: 10.1021/acs.energyfuels.7b02286 Energy Fuels XXXX, XXX, XXX−XXX

Article

Energy & Fuels Table 4. Solvent Concentrations in Oil and Rag Layers Based on Standard Addition Method oil layer solvent/bitumen, wt % water 0.7, 0.7, 0.7, 0.7,

10 25 50 75

solvent/bitumen, wt % water 0.7, 0.7, 0.7, 0.7,

10 25 50 75

wt % heptane

wt % toluene

wt % octane

3.94% 4.32% 4.19% 3.74%

3.68% 4.28% 4.37% 3.51%

5.80% 7.86% 6.83% 6.09%

wt % heptane

wt % toluene

wt % octane

1.48% 1.55% 2.01% 1.68%

1.21% 1.36% 1.97% 1.58%

2.46% 2.67% 3.67% 3.09%

wt % p-xylene

total mass of four hydrocarbons (g)

1.20% 1.95% 1.53% 1.13% rag layer

Figure 6. Distribution of four hydrocarbons between oil and rag layers giving the ratio of hydrocarbon in the oil/hydrocarbon in the rag layer as a function of solvent/bitumen and water added when making the rag layers.

ratio of four solvents

2.2 2.77 2.54 2.17

3.3:3.1:4.8:1 2.2:2.2:4.0:1 2.7:2.9:4.5:1 3.3:3.1:5.4:1

wt % p-xylene

total mass of four hydrocarbons (g)

ratio of four solvents

0.59% 0.62% 1.01% 0.76%

0.86 0.93 1.3 1.07

2.5:2.1:4.2:1 2.5:2.2:4.3:1 2.0:2.0:3.6:1 2.2:2.1:4.1:1

Figure 8. Calibration curve of naphtha in salt water.

and p-xylene in water were found to be 0.1, 0.4, 0.03, and 0.4 ppm, calculated from the calibration curves in the water phase. The detection limits for individual heptane, toluene, octane, and p-xylene in rag layers and oil were all approximately 1 wt %, estimated from the experiments. The detection limit for naphtha in water layers was 0.5 ppm, calculated from the naphtha calibration curve in the water phase. The detection limits for naphtha in rag layers and oil were both approximately 6 wt % estimated from the experiments. In addition, it was found (with one exception) that more p-xylene partitions into the rag layer than the other hydrocarbons. This analytical method is the first reported method to detect VOCs in the multiphase systems for oil sands samples without sample manipulation and has great potential to be applied in the future studies. Though it has some limitations, because of the sensitivity of the developed methods, with some modification, it is potentially applicable for the determination of VOCs in any environmental samples. For example, the method developed has been modified to measure the VOC content in the water phase of oil spill samples. It could also be used to quantify the solvent content in diluted bitumen. The advantage of the developed method is its ease of use and the absence of a sample extraction step, which would reduce human error in sample handling and the experimental procedure.

Figure 7. Calibration curves for oil layer and rag layer by naphtha standard addition method at solvent/bitumen 0.7 and 25% water.

based on the four-hydrocarbon addition method. However, due to the different partitioning behaviors of the various hydrocarbons in the multiphase system, this may not hold true. As a result, employing a respective standard addition method for the analytical objective is recommended.

4. CONCLUSIONS Analytical methods for detecting hydrocarbons in a multiphase system were developed in the present study. The methods developed employ standard addition and an internal standard. The respective detection limits for heptane, toluene, octane, F

DOI: 10.1021/acs.energyfuels.7b02286 Energy Fuels XXXX, XXX, XXX−XXX

Article

Energy & Fuels Table 5. Summary of Percent Solvent in Diluted Bitumen and Rag Layer Obtained by Standard Addition Method solvent/bitumen, wt % water



0.5, 0.5, 0.5, 0.5,

naphtha in oil layer (wt %)

naphtha in rag layer (wt %)

mass naphtha in oil (g)

mass naphtha in rag (g)

total mass recovered (g)

input mass (g)

percent naphtha recovery

46.7% 32.7% 56.7% 41.0%

8.3% 11.7% 16.0% 13.3%

13.7 9.6 8.7 5.1

0.2 0.9 6.5 2.4

13.9 10.5 15.2 7.5

10.17 10.09 10.37 5.29

137% 104% 146% 142%

10 25 50 75

(9) Kupai, M. M.; et al. Characterising rag-forming solids. Can. J. Chem. Eng. 2013, 91 (8), 1395−1401. (10) Varadaraj, R.; Brons, C. Molecular origins of crude oil interfacial activity part 3: Characterization of the complex fluid rag layer formed at crude oil-water interfaces. Energy Fuels 2007, 21 (3), 1617−1621. (11) Czarnecki, J.; Moran, K.; Yang, X. On the “rag layer” and diluted bitumen froth dewatering. Can. J. Chem. Eng. 2007, 85 (5), 748−755. (12) Gu, G.; et al. Novel bitumen froth cleaning device and rag layer characterization. Energy Fuels 2007, 21 (6), 3462−3468. (13) Kasperski, K. L.; Munoz, V.; Mikula, R. Naphtha Evaporation from Oil Sands Tailings Ponds. In Proceedings of the 2nd International Oil Sands Tailings Conference, University of Alberta Geotechnical Center and Oil Sands Tailing Research Facility: Edmonton, Alberta, Canada, Dec 2010. (14) Gupta, S. C.; Gittins, S.; Canas, C. Methodology for Estimating Recovered Solvent in Solvent-Aided Process. J. Can. Pet. Technol. 2012, 51, SPE-136402-PA. (15) Madjlessikupai, M. A. Study of the Rag Layer: Characterization of Solids. Ph.D. Thesis, University of Alberta, Edmonton, Alberta, Canada, 2012; p 77. (16) Binks, B. P.; et al. Measurement of film rigidity and interfacial tensions in several ionic surfactant-oil-water microemulsion systems. Langmuir 1989, 5 (2), 415−421. (17) Kiran, S. K.; Acosta, E. J.; Moran, K. Study of Solvent-BitumenWater Rag Layers. Energy Fuels 2009, 23, 3139−3149. (18) ASTM International. Standard Test Method for Determination of Individual Components in Spark Ignition Engine Fuels by 100−Metre Capillary (with Precolumn) High-Resolution Gas Chromatography; ASTM D6730-01(2016); West Conshohocken, PA, 2016. (19) Long, Y.; Dabros, T.; Hamza, H. Selective Solvent Deasphalting for Heavy Oil Emulsion Treatment. In Asphaltenes, Heavy Oils, and Petroleomics; Mullins, O., et al., Eds.; Springer: New York, 2007; pp 511−547. (20) Wu, X. Investigating the stability mechanism of water-in-diluted bitumen emulsions through isolation and characterization of the stabilizing materials at the interface. Energy Fuels 2003, 17 (1), 179− 190. (21) Yang, X. L.; Czarnecki, J. The effect of naphtha to bitumen ratio on properties of water in diluted bitumen emulsions. Colloids Surf., A 2002, 211 (2−3), 213−222. (22) Yeung, A.; et al. Micropipette: a new technique in emulsion research. Colloids Surf., A 2000, 174 (1−2), 169−181. (23) Mackay, D.; et al. Illustrated Handbook of Physical-Chemical Properties and Environmental Fate for Organic Chemicals, 2nd ed.; CRC Press: Boca Raton, FL, 2006; p 4216. (24) Yaws, C. L.; Narasimhan, P. K. Solubility of Hydrocarbons in Salt Water. In Water Encyclopedia; John Wiley & Sons, Inc.: Hoboken, NJ. 2005. (25) Uglev, N. P.; Ryabov, V. G.; Uraskin, S. I. Solubilities of naphtha, kerosine, and ethylbenzene in water. Chem. Technol. Fuels Oils 1986, 22 (4), 199−200. (26) Rossi, S. S.; Thomas, W. H. Solubility behavior of three aromatic hydrocarbons in distilled water and natural seawater. Environ. Sci. Technol. 1981, 15 (6), 715−716. (27) Confidential information.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.energyfuels.7b02286. Procedure for asphaltene content measurement, example of the multiphase system, detailed hydrocarbon analysis, additional four-solvent standard addition calibration curves used in the present study, alternative representation of the partitioning of four hydrocarbons between oil and rag layers giving the ratio of hydrocarbon in oil to hydrocarbon in rag layer as a function of solvent/ bitumen and water added, and additional naphtha standard addition calibration curves used in the present study (Figures S1−S12) (PDF)



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Xiaomeng Wang: 0000-0003-2169-3013 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The authors would like to thank H. Yin for conducting asphaltene content measurement and A. Truong for conducting DHA analysis. In addition, the authors would like to acknowledge financial support from the Government of Canada’s interdepartmental Program of Energy Research and Development, PERD.



REFERENCES

(1) Government of Alberta. Alberta Oil Sands Industry Quarterly Update; Edmonton, Alberta, Canada, 2017. (2) Masliyah, J.; et al. Understanding water-based bitumen extraction from athabasca oil sands. Can. J. Chem. Eng. 2004, 82 (4), 628−654. (3) Gu, G.; et al. Influence of water-soluble and water-insoluble natural surface active components on the stability of water-in-toluenediluted bitumen emulsion. Fuel 2002, 81 (14), 1859−1869. (4) Long, Y. C.; Dabros, T.; Hamza, H. Analysis of solvent-diluted bitumen from oil sands froth treatment using NIR spectroscopy. Can. J. Chem. Eng. 2004, 82 (4), 776−781. (5) Liggio, J.; et al. Oil sands operations as a large source of secondary organic aerosols. Nature 2016, 534, 91−94. (6) Simpson, I. J.; et al. Characterization of trace gases measured over Alberta oil sands mining operations: 76 speciated C2−C10 volatile organic compounds (VOCs), CO2, CH4, CO, NO, NO2, NOy, O3 and SO2. Atmos. Chem. Phys. 2010, 10 (23), 11931−11954. (7) Government of Alberta. Commercial Scheme: Approval No. 8512D; Energy Resources Conservation Board: Fort McMurray, Alberta, Canada, 2010. (8) Saadatmand, M.; Yarranton, H. W.; Moran, K. Rag Layers in Oil Sand Froths. Ind. Eng. Chem. Res. 2008, 47 (22), 8828−8839. G

DOI: 10.1021/acs.energyfuels.7b02286 Energy Fuels XXXX, XXX, XXX−XXX