An Enhanced-Solvent Deasphalting Process : Effect of Inclusion of SiO 2 Nanoparticles in the Quality of Deasphalted Oil

In this work, the effect of nanoparticles in deasphalting heavy oil and extra-heavy oil process at laboratory-scale based on the conventional solvent deasphalting process was studied and named enhanced-solvent deasphalting (e-SDA) process. This work evaluated the effect of the nanoparticles of SiO 2 in the separation efficiency based on deasphalted oil (DAO) fraction quality compared to the conventional process of deasphalting (SDA). Different effects have been assessed such as solvent to oil ratio, operating temperatures, type of solvent, and SiO 2 nanoparticles dosage. The DAO quality was based on the asphaltene and sulfur contents, API gravity, distillable fraction, and rheological properties.The improvement of the process from the use of nanoparticles was confirmed with important reductions in the asphaltene and sulfur contents in the DAO of up to 24% and 23%, respectively, in comparison with the SDA process. Also, the API gravity can be increased by approximately 14% with the e-SDA process. The rheological properties of the DAO were improved by the inclusion of nanoparticles showing reductions in the viscosities of the DAO greater than 50% in comparison with the conventional process. These results lead to the conclusion that the e-SDA process improves the DAO quality when compared with the typical deasphalting process.


Introduction
Growth in global energy demand of approximately 50% is expected by the year 2040, reaching approximately 400 millions of barrels of oil equivalent per day (mboe/d) [1].Because of this growth, nonconventional sources of hydrocarbons, such as heavy oil (HO) and extra-heavy oil (EHO), with nearly 70% of world oil reserves [2], of which EHO reserves correspond to 32% of the global oil reserves [3], have much potential.
HO and EHO are characterized by high specific gravities and high viscosities [4].According to the American Petroleum Institute (API) gravity, HO is defined by having an API gravity between 10 ∘ and 22.3 ∘ [4].EHO has an API gravity less than 10 ∘ and a viscosity at reservoir conditions higher than 10000 cP [4][5][6].HO and EHO reservoirs are widely distributed throughout the world in countries such as Venezuela, Canada, and Colombia [5,[7][8][9], imposing several technological challenges in their production, transportation, and refining in these countries [4,10,11].One of the main characteristics of HO and EHO is the high concentration of high molecular weight asphaltenes that are complex and highly polar compounds with large amounts of heteroatoms containing O, N, and S, as well as metals such as Ni, Fe, and V [12][13][14][15].These heteroatoms and their location in the asphaltene structure cause these molecules to be the most polar constituents in crude oil, leading to their selfassociation and formation of large asphaltic flocs [16,17].For example, for HO and EHO, in which asphaltenes are in high concentrations [18,19], the viscosity increases due to the significant influence of these molecules on the rheology resulting from the formation of a viscoelastic network [20][21][22][23].Also, high concentrations of asphaltenes in HO and EHO that contain elevated concentrations of sulfur with strong C-S and C=S bonds drastically increase the viscosity [24,25].
Because of these adverse factors, technologies that could be applied at the wellhead would be an efficient alternative to the current technologies.Several techniques have been used for HO and EHO viscosity reduction at the surface, such as dilution with solvents and light hydrocarbons [39,40] or solvent de-asphalting (SDA) [5].SDA is one of these technologies which is often used by the oil and gas industry for on-site upgrading.It is considered to be a viable option because operating costs can be reduced by the recovering and recycling of the solvents [70].SDA is an economical process to remove concentrated asphaltene or pitch, the dirtiest part of HO and EHO [70], and has been widely studied since the first attempts to separate or distilling oils in the 1920s [71,72].Since then, many efforts have been made to understand this process, the effect of the variables present involved, and how to improve it [73][74][75][76][77].SDA is a physical process in which there is a separation of two phases, the deasphalted oil (DAO) and the residue or pitch [70].The DAO is removed from the HO or EHO by solvent extraction using paraffinic hydrocarbons in which there is no chemical reaction between the crude oil feed and the paraffinic solvent [70,[78][79][80][81][82].Researchers have evaluated the effects of the chemical nature of the solvent [70,[78][79][80][81][82][83][84][85][86][87][88][89], temperature [82,90,91], operating pressure [83-85, 90, 92, 93], and solvent to oil ratio [77,79,81,[94][95][96] on the yield of DAO and/or pitch, as well as the DAO quality in some studies.Improvements of the SDA through the addition of macro-and microadsorbent materials have been reported in the literature, showing improvements in the DAO yield and quality [97,98].In some patents, the use of microscale solids such as clay, silica, alumina, and zeolite materials have been reported, reaching a higher quality of the DAO [99,100].Regarding Si-based materials, Ikematsu et al. [101] proposed an improved SDA process by using SiO 2 microparticles.The authors proposed the use of amorphous SiO 2 with a preferable particle size between 0.5 and 1.0 m and surface area of 100-800 m 2 /g [101].
Nevertheless, there have not yet been studies reported in the scientific literature about the role of nanoparticles (NPs) and their impact on the separation efficiency of the SDA process.These nanomaterials may offer special advantages for the SDA process due to their high surface area to volume ratio and hence a large number of available active sites, making them capable of selectively adsorbing asphaltenes onto their surfaces and enhancing the removal of the asphaltenes [31,[102][103][104][105][106][107][108].In previous studies [107,109] our research group has found that silica nanoparticles have a high affinity for asphaltene molecules, which could impact the efficiency directly of the SDA process.Franco et al. [107] showed that silica-based nanoparticles adsorbed up to 5.5 mg/m2 of asphaltenes in comparison with microparticulate silica.Therefore, this study focused on evaluating the effects of adding nanoparticles to a typical SDA process with the purpose of enhancing the quality of the DAO and optimizing this procedure using nanotechnology in process of designated enhancedsolvent deasphalting (e-SDA).Lee et al. [70] provided a typical SDA process scheme that includes a solvent extractor, DAO/solvent separator, and pitch stripper.The SDA process involves injecting an alkane into the crude oil to disrupt and to disperse its components causing the polar components to precipitate [5].The present paper is the first step for proposing the e-SDA technology shown in Figure 1.The experimental tests carried out at the laboratory-scale lead to the proposed e-SDA process to include the cyclic addition and recuperation of nanoparticles and the catalytic conversion of the pitch for energy production.The e-SDA process produces a DAO having enhanced quality and CO 2 that could be used in the EOR process [65,66].
In this study, laboratory-scale experimentation was made where the DAO and pitch yields are presented while focusing on the DAO quality by evaluating its asphaltene and sulfur contents, API gravity, distillable fraction, and rheological properties.The rheological properties were described using a power law model commonly employed for this type of fluid.

Materials and Chemicals.
A Colombian extra-heavy crude oil obtained from a reservoir located in the Department of Meta in Colombia's central region was used in these experiments.This EHO of 6.4 ∘ API had a content of asphaltenes and sulfur of 20.3 and 4.5 wt%, respectively.Its distillable fraction was approximately of 56% until 750 ∘ C. Two different paraffinic solvents were used in the deasphalting process: n-heptane (99%, Sigma-Aldrich, St. Louis, MO) and npentane (99%, Panreac, Barcelona, Spain).In this sense, the commercial fumed silica (SiO 2 ) nanoparticles (mean particle size of 7 nm and a surface area "S BET " of 389 m 2 /g) that were used to enhance the SDA process were obtained from Sigma-Aldrich (St. Louis, USA).

DAO Separation and Characterization.
The experimental procedure begins with the addition of the paraffinic solvents to the oil at a defined solvent to oil mass ratio (SOR) for the SDA process.For the e-SDA process, an identical test was made using SiO 2 nanoparticles at different dosages of 2.5, 5.0, and 10 wt% of the crude oil mass based on the maximum amount of asphaltenes that can be adsorbed by the nanoparticles [109], such that the asphaltenes adsorbed onto the nanoparticles can be decanted in the separation facilities of the e-SAD process.For this, the selected amount of nanoparticles was premixed with selected solvent and sonicated at 25 ∘ C for 1 h.The mixture was stirred at 300 rpm for 2 h at 25 ∘ C. At this point, the precipitation of the pitch was achieved, and this was separated from the DAO by filtration with an 8 m Whatman filter paper.At the end of the experiment, the solvent was separated from the DAO, and the solvent loss was calculated.
The DAO and pitch weight were determined to establish the SDA and e-SDA yields.On quality, the asphaltene and sulfur contents in the DAO fractions were determined, as well as their API gravities, rheological properties, and distillable fractions.The remaining asphaltenes in the DAO are a critical parameter because these are one of the responsible for the high viscosity values [16,17,[20][21][22][23], just as the sulfur content and their bonds with carbon [24,25].In this sense, DAO asphaltene content was determined by extraction using an excess amount of n-heptane according to a standard procedure having a measured deviation of ±0.01% that was described in previous studies [31,105].The sulfur content of the DAO and its API gravity were measured according to the procedures established by ASTM D-7220-12 [110] and ASTM D-1298-12 [111], respectively.The sulfur content was determined with an error in measurements of ±0.01%, and those of the API gravity measurements were within ±0.02 ∘ .DAO rheological measurements were conducted using a Kinexus rotational rheometer C-VOR 200 (Malvern Instruments, Worcestershire, UK), with a Peltier cell to maintain temperature control.The standard plate to plate geometry with a GAP of 300 microns was used.Rheological data were fitted to the model specified below.Finally, for each sample, the distillable fraction of the DAO was determined by gas chromatography following the ASTM D-7169-11 procedure [112], using an Agilent 7890A gas chromatograph (Agilent Technologies, Santa Clara, USA).The analytical tests for the DAO quality determinations were repeated three times.
It is worth noting that different factors were assessed in the tests: SOR values of 4, 8, and 12; temperatures of 25 and 70 ∘ C; type of solvent (n-pentane or n-heptane); and nanoparticles dosage.The experimental base scenario used an SOR = 8 using a dosage of 5 wt% of nanoparticles in the crude oil mass and n-heptane at 25 ∘ C.

Modelling
The Ostwald-de Waele model was used to fit the experimental data.This model is determined by the following mathematical expression [113,114]: where  (cP) represents viscosity,  (s −1 ) is the shear rate,  (dimensionless) is called power law index in which  < 1 correspond to a shear-thinning non-Newtonian liquid or pseudo-plastic fluids,  > 1 is characteristic of a shearthickening non-Newtonian fluid, and finally  = 1 is typical of Newtonian fluids [113,114].The  (cP⋅s −1 ) parameter is called the consistency coefficient.It is associated with the viscosity at a shear rate of 1 s −1 .On this parameter, it can be noticed that when n is equal to 1 (a Newtonian fluid), this parameter has units of cP and can be interpreted as the fluid viscosity independent of shear rate.For determining the goodness of fit of the employed model, the nonlinear chisquare ( 2 ) analysis and root-mean-square error (RMSE%) were used [115].

Solvent to Oil Ratio Effects.
In this section, the results of solvent to oil ratio (SOR) effect on the DAO yield and the quality of SDA and e-SDA processes are presented.Figure 2 shows the DAO and pitch yields for both processes using n-heptane at a nanoparticles dosage of 5 wt% and a fixed temperature of 25 ∘ C. Increasing the SOR reduces the DAO yield (Figure 2(a)) and consequently increases the pitch yield (Figure 2(b)).These results are in agreement with those reported by several authors [77,79,81,[94][95][96] in which this behavior is explained by the solvating power of solvent [81].To further understand this phenomenon, it is necessary to consider resins in addition to asphaltenes that are responsible for considerable changes in the solvent power [81].As the SOR increases, more resins are solubilized by the solvent and these are unable to act as peptizing (or interfacial) agents of asphaltenes [116][117][118][119] leading to their aggregation and precipitation [81,[116][117][118][119].Hence, as the SOR increased, the pitch yield increased while the DAO yield decreased proportionally [79,82].
On the other hand, the e-SDA process using SiO 2 nanoparticles has considerable effects on the DAO and pitch yields compared with the SDA process.It is worth mentioning that the silica nanoparticles selectivity to asphaltenes due to the presence of acid centers such as hydroxyl silanol (Si-OH) on their surface [119][120][121] and their interaction with the polar groups of asphaltenes are expected.Also, the DAO yields in the e-SDA process are lower in comparison with traditional SDA, and consequently the pitch yields are larger in the e-SDA process (Figure 2(b)).This increase of pitch can be explained by the adsorption phenomena between the silica nanoparticles and asphaltenes that has been widely studied and is determined by the affinity between adsorbent (SiO 2 nanoparticles) and adsorbate (asphaltenes), the asphaltene self-association over the surface of the nanoparticles, and their maximum adsorption capacity [122].Franco et al. [107] studied asphaltene adsorption in different nanoparticles and found that the process strongly depends on of the surface structure and chemistry of the nanomaterial.In this way, SiO 2 nanoparticles added to the deasphalting process adsorb asphaltenes on their surface and enhance their precipitation and separation from the crude oil matrix.However, the trend obtained by DAO and pitch yields over the wide range of SOR evaluated can be considered constant, as seen in Figure 2(a), indicating that the amount of solvent needed for the de-asphalting process can be reduced significantly by the addition of nanoparticles.Additionally, it has been demonstrated that nanoparticles impact the viscosity of oil by altering the aggregation system of asphaltenes [123,124].This may modify the efficiency of the deasphalting process by lowering the system viscosity, reducing the mass-transfer resistance between fractions, and thereby increasing the asphaltene extraction efficiency [98].
Figure 3 presents the DAO quality in three panels related to (a) asphaltene content, (b) sulfur content, and (c) ∘ API for both SDA and e-SDA processes.For the SDA process increasing the SOR decreased the asphaltene content in the DAO, as shown in Figure 3(a).This situation can be explained because increasing the SOR leads to reducing the asphaltene solubility causing them to precipitate [81,[116][117][118][119].For this reason, as the SOR increased, more asphaltenes were precipitated [83,84] and therefore the DAO quality increased.These results are widely confirmed by previous studies that measured the content of asphaltene in DAO [70,82,85].Regarding the sulfur content of the DAO, this is directly related to asphaltene content [82], which can be seen in Figure 3(b).As the asphaltene content of the DAO decreased, the sulfur content decreased, showing the same trend as the asphaltene content as SOR increased.Thus, Figure 3(c) shows that increases of the SOR increased the API gravity of the DAO.This result is consistent with the reductions in the asphaltene and sulfur contents [125].
The DAO quality was also affected by the addition of nanomaterials in the e-SDA process.As seen in Figure 3(a), the asphaltene content of the DAO was lower for all SOR values in comparison with the system without nanoparticles.This improvement in quality can be mainly explained by the interaction between the asphaltenes and the nanoparticles that remain in the pitch fraction, which also explains the sulfur content values of the e-SDA process that were clearly lower than those of the SDA process because of the direct relationship between the asphaltene and sulfur contents [82].The sulfur content is nearly constant as the SOR increased and could be due to the fact that O-and N-containing asphaltenes are more prone to get adsorbed on the silica nanoparticles than those containing sulfur [119,121].Additionally, Figure 3(c) shows that the e-SDA process improved the ∘ API values of the DAO for all of the SOR values.This remarkable result is also related to the reduction of the DAO asphaltene content values in e-SDA versus the traditional SDA process [125].This result can be contrasted with the simulated distillation results that show that the distillable fractions for all DAO samples (either SDA or e-SDA) were approximately 100% up to 750 ∘ C. Rheological measurements were conducted by triplicated to evaluate the DAO quality.Figure 4 shows the viscosity, shear stress, and the fitting with the Oswald-de Waele model as a function of shear rate for (a) crude oil and for DAO obtained from tests made with SOR values of (b) 4, (c) 8, and (d) 12.The estimated parameters of the Oswaldde Waele model are summarized in Table 1.As shown in Figure 4(a), the crude oil had the expected shear-thinning behavior frequently known as pseudo-plastic behavior [114].This is confirmed by the  parameter value in Table 1, which is clearly less than 1.
The DAO obtained from the SDA process with an SOR of 4 (Figure 4(b)) also exhibited a shear-thinning behavior.However, in this case, the  value is 0.906 and, being almost 1, had a nearly Newtonian behavior with a  value of 73321 cP⋅s −1 that is related to the fluid viscosity at a shear rate value of 1 s −1 .For the same SOR = 4, the e-SDA process considerably improved the DAO rheological properties as can be seen clearly from the results shown in Figure 4 and Table 1.The parameters of the Ostwald-de Waele model ( of 0.979 and  of 35522 cP⋅s −1 ) indicate an approximately Newtonian behavior with a viscosity of 35522 cP.The last parameter indicates that adding SiO 2 nanoparticles reduces expected viscosity of the DAO by more than 50%.For an SOR value of 8 in the SDA process (Figure 4(c)), the DAO rheological characteristics presented a similar trend with an approximately Newtonian behavior and viscosity of 19552 cP.For the same SOR, when the e-SDA process was conducted with SiO 2 nanoparticles, the rheological properties of the DAO were improved, having a viscosity of 11883 cP that indicates a viscosity reduction of 42% on the typical SDA process conducted without nanoparticles.The same performance was observed for the DAO with an SOR of 12 in which the viscosity was reduced from 10670 to 8504 cP in the SDA and e-SDA processes, respectively.These rheological results are interesting because these fluid  properties determine the crude oil and DAO transporting conditions.In Figure 4 it is observed that the inclusion of SiO 2 nanoparticles significantly reduced the shear stress and the yield point in comparison with the SDA process in the absence of nanoparticles.Finally, for the deasphalting process conducted with and without nanoparticles, at all of the assessed SOR values, the estimated solvent losses were lower than 1.0 wt%.

Temperature Effects.
Procedures were conducted with nheptane as the solvent at an SOR of 8 with a SiO 2 nanoparticles dosage of 5 wt% and a fixed temperature of 70 ∘ C to assess the temperature effects.The results are summarized in Table 2 and are in agreement with the literature reports, which found that, within the range evaluated, increasing the temperature increases the pitch yield and reduces the DAO yield [82,90,91].This can be seen for both the SDA and e-SDA processes and can be explained by increases in the differences in the solubility parameters and molar volumes between the solvent and crude oil as temperature rises, leading to a larger region of immiscibility in a solventpitch-DAO system [82].In this sense, increasing system temperature has a similar effect compared to that when the molecular weight of used solvent is decreased.For the SDA, the DAO yield decreased from 54.3% at 25 ∘ C to 48.3% at 70 ∘ C. Increasing the temperature in the e-SDA process improved the asphaltene separation as indicated by a decrease of the DAO yield from 44.6% to 39.5%.This situation suggests that there is a synergistic effect between the nanoparticles and temperature that improved the DAO and pitch separation that is controlled by adsorption on the nanoparticles and the increased diffusion of asphaltenes through the feedstock to the pitch by increased thermal motion [98].
The temperature effects on DAO quality are shown in Table 2.For the SDA process, the temperature improved the DAO quality, showing that asphaltene content was reduced from 5.14% at 25 ∘ C to 3.43% at 70 ∘ C.This type of behavior was also noticed in sulfur content that was reduced from 2.33% at 25 ∘ C to 2.04% at 70 ∘ C. The DAO quality was especially related to the asphaltene content.It increased as the temperature is increased, as expected for the typical SDA process [82,90,91].As noted above, the sulfur content and ∘ API are also related to the asphaltene content, and this explains their behavior in this investigation [78,82,125].For the e-SDA process, temperature plays the same role in improving DAO quality.The DAO asphaltene content was reduced from 4.24% at 25 ∘ C to 2.96% at 70 ∘ C. Consequently, the DAO sulfur content was also improved, decreasing from 2.02% to 1.56%.Table 2 shows that the physicochemical properties of the DAO were enhanced by the e-SDA process in comparison with the SDA process by increasing API gravity and by reducing both the asphaltene and sulfur contents at 70 ∘ C.Meanwhile, the solvent losses were also lower than 1.0 wt% in the processes conducted at 70 ∘ C.
Figure 5 shows the rheological properties of the DAO obtained from SDA and e-SDA processes at 70 ∘ C, and the estimated parameters for the Oswald-de Waele model are presented in Table 3.These results show that, at 70 ∘ C, the DAO samples obtained from both processes exhibited  approximately Newtonian behavior.For SDA, the Ostwaldde Waele estimated parameters were  = 1.005 and hence  can be considered as sample viscosity equal to 18410 cP.These results from the SDA process at 70 ∘ C denote a marked improvement over those at 25 ∘ C (Table 1).The same type of response was observed for the e-SDA process and can be noted in estimated parameters:  = 0.998 and  = 11699 cP.These results demonstrate that nanoparticles can improve the deasphalting process at 70 ∘ C by reducing the resulting DAO viscosity by approximately 37%.

Effect of the Type of Solvent.
The solvent used in SDA and e-SDA processes has a high influence.Table 4 shows the DAO yield and quality from both processes when conducted with n-pentane and at 25 ∘ C a dosage of 5 wt% of SiO 2 nanoparticles.Comparing these results with those obtained when using n-heptane as the solvent (Figure 2), it is observed that reducing the carbon number of solvent reduced the DAO yield and hence increased the pitch or precipitated fraction yield [78][79][80][81][82]84]. This is because increasing the molecular weight of the solvent allows it to solubilize heavier hydrocarbons [78,84], as reflected in the solubility parameter of asphaltenes in different solvents [80].This situation is similar in both the SDA and e-SDA processes.The DAO quality also followed this same phenomenon, and reducing the solvent molecular weight increases the DAO quality  [79,81,82].For the SDA process, the DAO asphaltene content was reduced from 5.14% with n-heptane to 2.34% with npentane.The same trend occurred with sulfur content, which fell from 2.33% to 1.55%.Finally, the ∘ API of the DAO improved considerably by increasing from 13.19 to 19.38 for n-heptane and n-pentane, respectively.This considerable improvement of the API gravity can be attributed to the resins and their influence on the deasphalting process [81,[116][117][118][119] and how they precipitated in higher proportion when the solvent had a lower molecular weight [82].
The e-SDA process also exhibited improvements regarding the DAO quality as the molecular weight of solvent was reduced.The asphaltene content fell from 4.24% with nheptane (Figure 2) to 2.01% with n-pentane (Table 4) and the same trend was observed for the sulfur content.On the DAO ∘ API, it increased from 15.37 to 21.11 with the use of n-heptane and n-pentane, respectively.The solvent losses were also less than 1 wt%.In this sense, there seemed to be a synergetic effect on the DAO quality between the reduction of solvent molecular weight and the use of SiO 2 nanoparticles.
Figure 6 and Table 5 summarize the rheological properties of DAO that were affected in several ways by the solvent used in the SDA and e-SDA processes.The results indicate nearly Newtonian behavior for the DAO obtained from both processes.These results were confirmed by the  parameters of Ostwald-de Waele models that equal 0.994 and 0.991, respectively.In this sense, the  values, interpreted as the viscosity, were 2500 cP for SDA and 2166 cP for e-SDA; that is, a reduction in resultant DAO viscosity of approximately 14% occurred with the e-SDA process.This viscosity reduction that was exhibited when using nanoparticles could be attributed mainly to the resin fraction of the DAO that played a key role in the formation of the viscoelastic network that drastically increased the EHO viscosity [20,124].It is known that resins are adsorbed simultaneously with asphaltenes on the nanoparticles [109] implying that a higher amount of the resins could remain in the pitch fraction.Hence, nanoparticles may have a synergistic effect with the solvent employed, n-pentane in this case, by enhancing the amount of resins adsorbed and thereby inhibiting the formation of the viscoelastic network, which translates into lower energy demand for fluid transportation and processing.

Effects of Nanoparticle Dosage.
Using SiO 2 nanoparticles in the e-SDA process in dosages of 2.5 and 10 wt% of crude oil mass generated differences with the same process conducted with a fixed dosage of 5 wt%, in both DAO yield and quality.Figure 7 shows the yield of (a) DAO and (b) pitch as a function of the nanoparticles dosage.As can be noticed, increasing nanoparticles dosage decreases DAO yield and increases pitch yield.This behavior can be explained by the relation between adsorbate and adsorbent.Increasing the mass of the adsorbent and hence the total surface area available for adsorption by mass unit generates that larger amounts of asphaltenes that could be adsorbed on the nanoparticles surface.This situation leads to the fact that as SiO 2 nanoparticles dosage increases, more asphaltenes are adsorbed, and DAO yield decreases while pitch yield increases.These results are in agreement with those reported by several authors [120,122,124,[126][127][128][129], but considering the total mass of adsorbed asphaltenes instead of the relation between the mass of adsorbed asphaltenes and the mass of adsorbent.
Figure 8 shows the DAO quality in three panels related to (a) asphaltene content, (b) sulfur content, and (c) ∘ API for e-SDA process at different SiO 2 nanoparticles dosage.The asphaltene content of the DAO reduces as nanoparticles dosage increases.This situation can be explained by the same situation abovementioned; increasing the total surface area for asphaltene adsorptions leads to their precipitation [ 120,122,124,[126][127][128][129] and lowers the quantity of them present in the DAO.The sulfur content presents the same behavior and the explanation lays in the direct relation between asphaltene and sulfur content [82].Additionally, Figure 8(c) shows that increasing nanoparticles dosage improved the ∘ API values for the DAO.This result is also related to the reduction of DAO asphaltene content with the increment in nanoparticle dosage.Regarding the simulated distillation, results show that the distillable fractions for all DAO samples (at different nanoparticles dosage) were approximately 100% up to 750 ∘ C. Figure 9 and Table 6 summarize the result of the rheological test carried out to the DAO obtained from e-SDA process with SiO 2 nanoparticles at dosages of 2.5 wt% and 10 wt%.Both DAO samples obtained with the 2.5 and 10 wt% dosages exhibited a Newtonian behavior.For a dosage of 2.5 wt%, the  value is 1.001 and  value is 15802 cP⋅s −1 .The last parameter indicates a viscosity of 15802 cP.For a SiO 2 nanoparticles dosage of 10 wt% the DAO shows  value equal to 0.996 and  value of 9982 cP⋅s −1 , showing a viscosity of 9982 cP.It can be noticed that increasing nanoparticles dosage improves the rheological behavior of DAO considering these results and those obtained for a dosage of 5 wt%.Increasing the nanoparticles dosage from 2.5 to 5 wt% generates a viscosity reduction of 25% while raising the dosage from 5 to 10 wt% reduces the viscosity in 16%.Hence, a dosage of 5 wt% can be considered adequate because increasing in the DAO quality by rising nanoparticles dosage to 10 wt% is not as significant as that obtained when the dosage is increased from 2.5 to 5 wt%.

Conclusions
The efficiency of a typical SDA process was successfully increased at laboratory-scale conditions by the inclusion of nanoparticles in systems with different solvent to oil ratios, temperatures, types of solvent, and dosage.It was found that the use of nanoparticles reduces the DAO yield in comparison with the traditional SDA process, resulting in a higher quality product.The reason is the adsorption phenomena of the asphaltenes and the nanoparticles and how the former are transferred from the DAO fraction to the pitch fraction.Regarding the DAO quality, the e-SDA process was demonstrated to be a more suitable alternative regarding the asphaltene and sulfur contents, API gravity, and rheological properties.The improvement of the rheological properties is desirable because this leads to savings of energy and transportation efforts when the DAO has lower viscosities.Finally, it is worth noting that this study was focused on DAO quality in the e-SDA process and there is importance in future investigations to assess the improvements of the catalytic thermal decomposition of the pitch when adding nanoparticles in the e-SDA process.This subject is important because of the energy costs involved in this process and its efficiency, considering that used solvent and nanoparticles can be recycled.It is expected that this knowledge will provide a suitable alternative for the enhancement of heavy and extraheavy oil processing operations, particularly as related to the transportation process.

Figure 1 :
Figure 1: Schematic diagram of the e-SDA process.The red box corresponds to studied part of proposed process in this paper.

Figure 2 :
Figure 2: Effect of the solvent to oil ratio in (a) DAO and (b) pitch yields for the SDA and e-SDA processes using SiO 2 nanoparticles at a dosage of 5 wt% at 25 ∘ C and using n-heptane as the solvent.

Figure 3 :
Figure 3: Effect of solvent to oil ratio on DAO quality related to (a) asphaltene and (b) sulfur content and (c) API gravity for the SDA and e-SDA processes using SiO 2 nanoparticles at a dosage of 5 wt% at 25 ∘ C and using n-heptane as the solvent.

Figure 4 :
Figure 4: Viscosity and shear stress as a function of shear rate for (a) crude oil and DAO after SDA and e-SDA with SOR values of (b) 4, (c) 8, and (d) 12 at 25 ∘ C with a fixed dosage of SiO 2 nanoparticles of 5 wt% and with n-heptane as the solvent.

Figure 5 :
Figure 5: Viscosity and shear stress of the DAO as a function of shear rate for the SDA and e-SDA processes at 70 ∘ C using a SiO 2 nanoparticles dosage of 5 wt% and an SOR of 8 with n-heptane as the solvent.

Figure 6 :
Figure 6: Viscosity and shear stress of the DAO as a function of shear rate for the SDA and e-SDA processes at 25 ∘ C using SiO 2 nanoparticles at a dosage of 5 wt% and an SOR of 8 with -pentane as the solvent.

Figure 7 :Figure 8 :
Figure 7: Effect of the SiO 2 nanoparticles dosage in (a) DAO and (b) pitch yields for the e-SDA process at SOR value of 8 at 25 ∘ C and using n-heptane as the solvent.

Figure 9 :
Figure 9: Viscosity and shear stress as a function of shear rate for SiO 2 nanoparticles dosages of (a) 2.5 and (b) 10 wt% at SOR = 8 at 25 ∘ C and using n-heptane as the solvent.

Table 1 :
Ostwald-de Waele estimated parameters for rheological experiments on crude oil and DAO after SDA and -SDA with SOR values of 4, 8, and 12 at 25 ∘ C with a fixed dosage of SiO 2 nanoparticles 5 wt% and -heptane as the solvent.

Table 2 :
Yield and quality of the DAO after the SDA and -SDA processes at 70 ∘ C using a SiO 2 nanoparticles dosage of 5 wt% and an SOR of 8 with -heptane as the solvent.

Table 3 :
Ostwald-de Waele estimated parameters from rheological experiments for DAO after the SDA and -SDA processes with an SOR of 8 at 70 ∘ C using SiO 2 nanoparticles and -heptane as the solvent.

Table 4 :
Yield and quality of the DAO after the SDA and -SDA processes at 25 ∘ C using SiO 2 nanoparticles at a dosage of 5 wt% and an SOR of 8 with -pentane as the solvent.

Table 5 :
Ostwald-de Waele estimated parameters from rheological experiments for the DAO after the SDA and -SDA processes with an SOR of 8 at 25 ∘ C using SiO 2 nanoparticles at a dosage of 5 wt% and an SOR of 8 with -pentane as the solvent.

Table 6 :
Ostwald-de Waele estimated parameters for rheological experiments on DAO after -SDA with SiO 2 nanoparticles dosage of 2.5 and 10 wt% at SOR = 8 at 25 ∘ C and using -heptane as the solvent.