Improvement of Steam Injection Processes Through Nanotechnology: An Approach through in Situ Upgrading and Foam Injection

: This study aims to evaluate a high-performance nanocatalyst for upgrading of extra-heavy crude oil recovery and at the same time evaluate the capacity of foams generated with a nanoﬂuid to improve the sweeping e ﬃ ciency through a continuous steam injection process at reservoir conditions. CeO 2 ± δ nanoparticles functionalized with mass fractions of 0.89% and 1.1% of NiO and PdO, respectively, were employed to assist the technology and achieve the oil upgrading. In addition, silica nanoparticles grafted with a mass fraction of 12% polyethylene glycol were used as an additive to improve the stability of an alpha-oleﬁn sulphonate-based foam. The nanoﬂuid formulation for the in situ upgrading process was carried out through thermogravimetric analysis and measurements of zeta potential during eight days to ﬁnd the best concentration of nanoparticles and surfactant, respectively. The displacement test was carried out in di ﬀ erent stages, including, (i) basic characterization, (ii) steam injection in the absence of nanoﬂuids, (iii) steam injection after soaking with nanoﬂuid for in situ upgrading, (iv) N 2 injection, and (v) steam injection after foaming nanoﬂuid. Increase in the oil recovery of 8.8%, 3%, and 5.5% are obtained for the technology assisted by the nanocatalyst-based nanoﬂuid, after the nitrogen injection, and subsequent to the thermal foam injection, respectively. Analytical methods showed that the oil viscosity was reduced 79%, 77%, and 31%, in each case. Regarding the asphaltene content, with the presence of the nanocatalyst, it decreased from 28.7% up to 12.9%. Also, the American Petroleum Institute (API) gravity values increased by up to 47%. It was observed that the crude oil produced after the foam injection was of higher quality than the crude oil without treatment, indicating that the thermal foam leads to a better swept of the porous medium containing upgraded oil.


Introduction
One of the biggest problems in the use of heavy (HO) and extra-heavy crude oils (EHO) as an energy source is their low quality [1][2][3], which leads to production, transportation, and refining operations that require high energy and economic investments [4][5][6]. HO and EHO generally have a Methanol (99.8% purity), toluene (99.8% purity), and HCl (37% purity) were used for porous media cleaning, and this was prepared with clean silica sand (Ottawa sand, US sieves 30-40 mesh) purchased from Minercol S.A. (Colombia). All chemical reagents used were purchased from Panreac (Chicago, IL, USA) unless otherwise specified. Finally, steam generation was done using deionized water with a conductivity of 3 µS·cm −1 .

Porous Medium
The porous medium cleaning was carried out using a toluene-methanol solution in a 1:1 ratio, using 2 mL of the solution per 1 g of sand. Subsequently, the sand was rewashed with 10% HCl to remove impurities and residues from the solvents. Finally, the sand was dried in a vacuum oven for 12 h at 120 • C. A total of 100 g of silica sand were used to carry out each displacement test, its porosity was estimated following the saturation method [54], where a low flow rate of 0.1 cm 3 ·min −1 is fixed with the sample holder in a vertical position. The time in which the system saturates itself and the permeability under Darcy's law are recorded [55].

Nanocatalyst Selection
The performance of three bimetallic nanoparticles composed of Fe-Pd, Co-Pd, and Ni-Pd supported in CeO 2 nanoparticles was evaluated through adsorption and catalytic experiments [35]. Batch-adsorption tests were carried out at 25 • C with a fixed amount of nanoparticles, maintaining a ratio of 0.1 with the solvent and varying the initial concentration of asphaltenes from 100 to 1500 mg·L −1 . The supernatant was analyzed with a UV-Vis spectrophotometer Genesys 10S (Thermo Scientific, Waltham, MA, USA) by which the adsorbed amount of asphaltenes in the nanoparticles was calculated. As for the precipitate, it was subjected to catalytic experiments using a simultaneous thermogravimetric analyzer Q50 (TA Instruments, Inc., New Castle, DE, USA) coupled to an IR-Affinity-1 FTIR device (Shimadzu, Kyoto, Japan) to analyze the produced gases during the gasification process. In addition, two types of thermogravimetric analysis were evaluated, one under isothermal conditions at 230, 240, and 250 • C and another test, where the samples were heated from 100 to 600 • C at a heating rate of 20 • C·min −1 in an atmosphere of N 2 saturated with H 2 O (g) . For the best of the three systems evaluated, a simplex-centroid mixture design [56] was employed in order to optimize the metal oxides concentration over CeO 2 support and obtain a maximization in the response variable [35].

Nanofluid Formulation for in Situ Upgrading
The formulation of the nanofluid for the in situ upgrading of the employed EHO was carried out using a Tween 80 surfactant and CeNi0.89Pd1.1 nanoparticles in a NaCl-based brine solution.
To obtain the best formulation of surfactant and nanoparticles, a sweep of concentrations of each component was made between 0.1% and 0.5% in mass fraction. The best nanoparticle concentration was determined based on its catalytic activity to decompose heavy oil compounds in a steam atmosphere [57]. The nanoparticles were mixed with the EHO and the solutions were stirred magnetically for 24 h at 80 • C. By means of centrifugation, filtering and washing with n-heptane, the nanoparticles are separated with the adsorbed fractions from the solutions. Nanoparticles are dried at 100 • C to remove solvent residues and finally the catalytic steam gasification of the samples was carried out using a thermogravimetric analyzer Q50 (TA Instruments, Inc., New Castel, DE, USA) at a heating rate of 20 • C·min -1 from 100 to 600 • C under an airflow of 100 mL·min −1 and a H 2 O (g) flow of 6.30 mL·min −1 [35].
On the other hand, the appropriate surfactant concentration was determined based on nanofluid stability. For this, the nanoparticles concentration was fixed, the nanofluids were subjected to an ultrasound bath for 4 h, and then the zeta potential was measured during 8 days with a Nanoplus-3 (Micromeritics, GA, USA) [58].
The nanofluid was characterized by viscosity at 25 • C using a Kinexus Pro+ (Malvern Instruments, Worcestershire, UK) rheometer, and the density as measured using an Anton Paar Stabinger SVM 3000 (Madrid, Spain).

Selection of Foaming Nanofluid
The nanofluid for the foam was prepared by adding a specific amount of 0.5% (by mass) of alpha-olefin sulfonate surfactant to a NaCl brine of 0.2% (by mass). Then, SPG12 nanoparticles were added, and the suspension was stirred at 600 rpm for 2 h. After, the suspension was placed in an ultrasound probe for 30 min and then left untouched for 24 h. The foaming nanofluid was also characterized through viscosity and density as explained in the previous section.
The test for foaming agent selection consists of determining the half-life (t 1/2 ) and durability (t ∞ ) of the foam generated by nitrogen injection to the nanofluid. The half-life is the time when only 50% of the initial height of the foam remains and durability when 10% remains. Both parameters were taken as an indication of stability. Foam half life and durability of the nanofluids was evaluated by placing 10 mL in a graduated glass cylinder that consists of two concentric cylinders. In the internal cylinder, the foam is generated by injecting N 2 (flow rate of 0.4 L·min −1 ) through a porous disc located at the bottom of the cylinder. Hot water is circulated in the outer cylinder for temperature control. A schematic representation of the experimental set up employed is shown in Figure S1 of the Supplementary Material. When the foam filled the internal cylinder, gas injection was stopped. The foam height was recorded as a function of time. The half-life and durability, which is the time when only 50% and 10% of the initial height of the foam remains, were taken as an indication of stability. The normalized foam height was defined as [47]: where H N (dimensionless) is the normalized foam height, H t=0 (mL) the initial foam volume, and H t (mL) is the foam volume at a determined time "t" (min). Experiments were performed by triplicate to confirm reproducibility and the average values are reported here.

Displacement Tests
The main objective of the coreflooding test was to evaluate a novel technology for increasing the oil recovery based on the combination of the steam injection-foam-steam injection assisted with nanoparticles. The injection of the foam is looking to improve the sweep of the porous media by blocking the zones of high permeability, favoring the sweep of zones of low permeability. Figure 1 shows a representative diagram of the experimental assembly used for steam injection tests. The set-up is composed of three main systems namely: (i) the steam generation system, (ii) the fluids injection system, and (iii) the pressure and condensation system. The steam generation system includes positive displacement pumps (DB Robinson Group, Edmonton, AB, Canada), two cylinders (Max Servicios S.A.S., Colombia) that contain the water for steam generation, and another to take control of the quality of the steam and nanofluid injection, and a tubular furnace (Thermo Scientific Waltham, MA, USA). The fluid injection system consists of an oil-containing and a nanofluid-containing cylinder, a pressure transductor (Rosemount, Emerson, MO, USA), a core holder along with the sand pack (3.8 cm and 60 cm long effective), and six thermocouples (Termocuplas, S.A.S., Medellín, Colombia) throughout the assembly. Finally, the pressure and condensation system consists of a hydraulic pump (Enercap, Actuant Corporation, WI, USA), a pressure multiplier, a separator, and liquid sample collectors. Additionally, each system has its respective manometers (Rosemount, Emerson, MO, USA) and valves (Swagelok, OH, USA).
Energies 2019, 12, x FOR PEER REVIEW 6 of 21 Figure 1 shows a representative diagram of the experimental assembly used for steam injection tests. The set-up is composed of three main systems namely: (i) the steam generation system, (ii) the fluids injection system, and (iii) the pressure and condensation system. The steam generation system includes positive displacement pumps (DB Robinson Group, Edmonton, AB, Canada), two cylinders (Max Servicios S.A.S., Colombia) that contain the water for steam generation, and another to take control of the quality of the steam and nanofluid injection, and a tubular furnace (Thermo Scientific Waltham, MA, USA). The fluid injection system consists of an oil-containing and a nanofluidcontaining cylinder, a pressure transductor (Rosemount, Emerson, MO, USA), a core holder along with the sand pack (3.8 cm and 60 cm long effective), and six thermocouples (Termocuplas, S.A.S., Medellín, Colombia) throughout the assembly. Finally, the pressure and condensation system consists of a hydraulic pump (Enercap, Actuant Corporation, WI, USA), a pressure multiplier, a separator, and liquid sample collectors. Additionally, each system has its respective manometers (Rosemount, Emerson, MO, USA) and valves (Swagelok, OH, USA).

Steam Injection Tests in the Presence and Absence of Nanofluids
The evaluation of nanofluid at dynamic conditions was carried out in five stages including (i) basic characterization, (ii) oil recovery after steam injection in the absence of nanofluids, (iii) oil recovery after the injection of nanofluids for in situ upgrading, and (iv) N2 injection to obtain gas permeability baseline, and (v) oil recovery after the injection of the foaming nanofluid. The tests were conducted at a fixed back pressure of 110 psi, overburden pressure of 700 and 800 psi for basic characterization and steam injection tests, respectively and a temperature of 80 °C. For the basic characterization of the absolute and effective permeabilities, (Kabs and ko) are estimated applying Darcy's law [55] by injecting water and oil at 0.5 mL 3 •min −1 according to previous studies [59]. Then, the steam was injected into the porous medium between 3 and 5 mL·min −1 at 210 °C and 210 psi and a quality of 70% (X = 70%) and the oil recovery curves in the absence of the nanofluids were obtained. After, 0.5 pore volumes (PV) of the CeNi0.89Pd1.1 nanoparticles-based nanofluid for in situ upgrading were injected at 0.5 mL•min −1 and left for a soaking time of 12 h. Then, steam was injected again to measure the oil recovery after the upgrading of the heavy crude oil. After that, 0.4 PV of the foaming nanofluid was injected at 3 mL•min −1 to block preferential channels and allow the steam contact with the non-invaded area. The foam was generated in the porous medium trough the injection of 1.5 equivalent pore volumes equivalents (EPV) of N2 at a rate of 3 L•min −1 to obtain a gas permeability baseline and verify, after divergent fluid injection, the foam in situ formation. The formation of the foam in situ is validated by the conductivity reduction of the system (Q•Δ −1 ) and pressure changes. Finally, steam was injected to measure oil recovery after injection. In all cases the pressure drop was recorded, and the steam was injected until no more oil is produced.

Steam Injection Tests in the Presence and Absence of Nanofluids
The evaluation of nanofluid at dynamic conditions was carried out in five stages including (i) basic characterization, (ii) oil recovery after steam injection in the absence of nanofluids, (iii) oil recovery after the injection of nanofluids for in situ upgrading, and (iv) N 2 injection to obtain gas permeability baseline, and (v) oil recovery after the injection of the foaming nanofluid. The tests were conducted at a fixed back pressure of 110 psi, overburden pressure of 700 and 800 psi for basic characterization and steam injection tests, respectively and a temperature of 80 • C. For the basic characterization of the absolute and effective permeabilities, (K abs and k o ) are estimated applying Darcy's law [55] by injecting water and oil at 0.5 mL 3 ·min −1 according to previous studies [59]. Then, the steam was injected into the porous medium between 3 and 5 mL·min −1 at 210 • C and 210 psi and a quality of 70% (X = 70%) and the oil recovery curves in the absence of the nanofluids were obtained. After, 0.5 pore volumes (PV) of the CeNi0.89Pd1.1 nanoparticles-based nanofluid for in situ upgrading were injected at 0.5 mL·min −1 and left for a soaking time of 12 h. Then, steam was injected again to measure the oil recovery after the upgrading of the heavy crude oil. After that, 0.4 PV of the foaming nanofluid was injected at 3 mL·min −1 to block preferential channels and allow the steam contact with the non-invaded area. The foam was generated in the porous medium trough the injection of 1.5 equivalent pore volumes equivalents (EPV) of N 2 at a rate of 3 L·min −1 to obtain a gas permeability baseline and verify, after divergent fluid injection, the foam in situ formation. The formation of the foam in situ is validated by the conductivity reduction of the system (Q·∆ −1 ) and pressure changes. Finally, steam was injected to measure oil recovery after injection. In all cases the pressure drop was recorded, and the steam was injected until no more oil is produced.

Effluent Analysis
In order to evaluate the effect of the CeNi0.89Pd1.1 nanoparticles assisting steam injection processes, the changes of the EHO on its API gravity, viscosity, the content of saturates, aromatics, resins, and asphaltenes (SARA), and residue content (> 620 • C + ) were estimated. The API gravity was measured using an Anton Paar Stabinger SVM 3000 rotating cylinder viscometer (Madrid, Spain). For the SARA analysis, micro de-asphalting was done to the extra-heavy oil and the upgraded oil using n-heptane and using the IP 469 method in an IATROSCAN MK6 TLC-FID/FPD equipment through hydrogen injection [36,60]. The viscosity measurements were made using a Kinexus Pro rheometer (Malvern Instruments, Worcestershire, UK). A parallel plate-plate geometry [61] (optimal for highly viscous fluids) was employed with a GAP of 0.3 mm, for different shear rates between 0 s −1 to 100 s −1 at 25 • C. The viscosity reduction degree (VRD) was calculated following the Equation (2) [61]: where µ v,EHO and µ treat are the viscosity of the virgin EHO and its viscosity after steam injection assisted by a nanofluid, respectively. Finally, through high temperature simulated distillation (HTSD), the residue content was estimated following the American Society for Testing and Materials (ASTM) D-7169 standard in an Agilent 7890 chromatographer (Agilent Technologies, Wilmington, DE, USA) [62]. The residue content was calculated with Equation (3) [63]: where R v,EHO and R treat are the residue content of the virgin EHO and EHO after the steam and nanofluid injection, respectively. Also, to describe the oil rheological behavior in the different stages of the oil recovery, the Cross model is proposed and showed in Table S1 of the Supplementary Material Document.

Selection of Nanocatalyst
The nanocatalyst selection was carried out through batch adsorption and thermogravimetric experiments. Initially, three bimetallic systems composed of the couples Ni-Pd, Fe-Pd, and Co-Pd, all supported over CeO 2 nanoparticles, were subjected to an asphaltenes adsorption test. Subsequently, the decomposition of the adsorbed fractions was monitored through a gasification process under steam atmosphere [35]. The system with Ni-Pd nanocrystals, showed a greater adsorption capacity and a high affinity towards heavy hydrocarbons, over the other systems, favoured by Lewis acidity and effective nuclear charge that increases in the order Fe < Co < Ni [64]. In addition, this system has the ability to decompose the asphaltenes molecules at 220 • C, while the Co-Pd and Fe-Pd systems lead to decomposition temperatures around 230 and 240 • C, respectively. Further, a simplex-centroid mixture design was employed to optimize de asphaltenes conversion over the surface of Ni-Pd nanoparticles [35]. The statistical method showed that with a fraction mass of 0.89% and 1.1% of NiO and PdO, it is possible to convert 100% of the asphaltenes adsorbed in less than 80 min. A detailed description of these experiments is shown in previous studies [35,65].  Figure 2 shows the tests carried out for the formulation of the nanofluid to select the best concentration of CeNi0.89Pd1.1 nanoparticles and Tween 80 surfactant through (a) thermogravimetric and (b) zeta potential analysis, respectively. The nanoparticles and surfactant concentrations evaluated were 0.1%, 0.2%, 0.3%, and 0.5% in a mass fraction. From the panel (a) of Figure 2, it is possible to observe that the dosage of the treatment has a significant effect on the decomposition of the extra-heavy oil compounds, which is reflected in the intensities and temperature peaks for each sample. For the particular case of 0.1% of CeNi0.89Pd1.1 nanoparticles, the peaks around 470 • C and 540 are predominant, indicating that with this amount of nanoparticles, the decomposition temperature of the heaviest hydrocarbons occurs in this region and that a higher amount of nanoparticles is needed [65,66] as a consequence of the high content of heavy compounds in the crude oil. For concentrations of 0.2%, 0.3%, and 0.5% in a fraction mass of the nanoparticles, an increase the intensities of the peaks located around 210 and 280 • C is observed. This means that the catalytic effect of the nanoparticles starts to be significant regarding the amount of heavy compounds in the system from the dosage of 0.2%. At this concentration, the intensities of the peaks at low temperatures (<250 • C) is higher than for the other concentrations employed. Also, at high temperatures (>450 • C) the intensity of the peak for a concentration of 0.2% is the lowest in comparison with the other concentrations evaluated. These results could be due to concentrations higher than 0.2% lead to a higher nanoparticle aggregation and a reduction of the available surface area [57]. Hence, a nanoparticle dosage of 0.2% was selected for injection into the porous medium. concentrations evaluated were 0.1%, 0.2%, 0.3%, and 0.5% in a mass fraction. From the panel (a) of Figure 2, it is possible to observe that the dosage of the treatment has a significant effect on the decomposition of the extra-heavy oil compounds, which is reflected in the intensities and temperature peaks for each sample. For the particular case of 0.1% of CeNi0.89Pd1.1 nanoparticles, the peaks around 470 °C and 540 are predominant, indicating that with this amount of nanoparticles, the decomposition temperature of the heaviest hydrocarbons occurs in this region and that a higher amount of nanoparticles is needed [65,66] as a consequence of the high content of heavy compounds in the crude oil. For concentrations of 0.2%, 0.3%, and 0.5% in a fraction mass of the nanoparticles, an increase the intensities of the peaks located around 210 and 280 °C is observed. This means that the catalytic effect of the nanoparticles starts to be significant regarding the amount of heavy compounds in the system from the dosage of 0.2%. At this concentration, the intensities of the peaks at low temperatures (<250 °C) is higher than for the other concentrations employed. Also, at high temperatures (>450 °C) the intensity of the peak for a concentration of 0.2% is the lowest in comparison with the other concentrations evaluated. These results could be due to concentrations higher than 0.2% lead to a higher nanoparticle aggregation and a reduction of the available surface area [57]. Hence, a nanoparticle dosage of 0.2% was selected for injection into the porous medium. From Panel b in Figure 2 it can be observed that the surfactant concentration changes the nanofluid stability for a fixed concentration of CeNi0.89Pd1.1 nanoparticles of 0.2%. For surfactant concentrations ≤ 0.2%, the Z potential (ζ) of the system has fluctuations in both the stability zone (> 20 and < −20 mV) and the non-stability zone (−20-20 mV) [67]. On the other hand, the concentrations of 0.3% and 0.5% maintain the value constant of the ζ up to day 6 at 25 mV and −5 mV, respectively. The nanoparticles absorb surfactant and consequently, there are variations in the ζ. Probably for concentrations below 0.02%, the adsorption equilibrium is never reached, so the z potential is not stable. In this order, a dosage of 0.3% of Tween 80 surfactant was selected for nanofluid preparation. It is worth to mention here, according to previous studies, Tween 80 was selected as it has null influence on the heavy oil recovery and the main mechanisms are controlled by the nanoparticles [68]. The density and viscosity of the selected nanofluid were 1.12 cP and 0.998 g•cm −3 at 25 °C.  Figure 2 it can be observed that the surfactant concentration changes the nanofluid stability for a fixed concentration of CeNi0.89Pd1.1 nanoparticles of 0.2%. For surfactant concentrations ≤ 0.2%, the Z potential (ζ) of the system has fluctuations in both the stability zone (> 20 and < −20 mV) and the non-stability zone (−20-20 mV) [67]. On the other hand, the concentrations of 0.3% and 0.5% maintain the value constant of the ζ up to day 6 at 25 mV and −5 mV, respectively. The nanoparticles absorb surfactant and consequently, there are variations in the ζ. Probably for concentrations below 0.02%, the adsorption equilibrium is never reached, so the z potential is not stable. In this order, a dosage of 0.3% of Tween 80 surfactant was selected for nanofluid preparation. It is worth to mention here, according to previous studies, Tween 80 was selected as it has null influence on the heavy oil Energies 2019, 12, 4633 9 of 21 recovery and the main mechanisms are controlled by the nanoparticles [68]. The density and viscosity of the selected nanofluid were 1.12 cP and 0.998 g·cm −3 at 25 • C. Table 2 shows the values of the half-life time and durability of the foam generated with nanofluids of alpha-olefin sulfonate with a mass fraction of 0.5% and SPG12 nanoparticles. The nanoparticle concentration was varied between 0.01% and 1% in fraction mass. The best performance was obtained for the nanofluid with a concentration of 0.01% in a mass fraction of nanoparticles. The experimental results show that the addition of nanoparticles helps to double the half-life and triple the durability compared with the surfactant in the absence of the nanomaterial. Table 2. Stability parameters of foam generated using alpha-olefin sulfonate surfactant (5000 mg·L −1 ) and polyethylene glycol grafted silica nanoparticles. Experimental conditions: 90 • C and atmospheric pressure. The measured density and viscosity of the nanofluid composed of the surfactant an 0.01% of SPG12 nanoparticles were 0.997 g·cm −3 and 1.01 cP, respectively.

Steam Injection Process Assisted by Nanotechnology
The effect of the nanocatalyst and foaming nanofluid in the steam injection was developed in five main stages including, (i) basic characterization, (ii) steam injection in the absence of nanofluids, (iii) steam injection after soaking with nanofluid for in situ upgrading, (iv) N 2 injection, and (v) steam injection after foaming nanofluid. The (a) oil recovery curves and (b) the system conductivity in each scenario are shown in Figure 3. active phases of the support and metal oxides promotes its conversion into lower molecular weight hydrocarbons [32,38,81,82], and gases such as CO [35], CO2 [35], and CH4, through the development of different reactions such as water-gas shift [83], methanation [84], steam reforming [38], partial oxidation [85], methane combustion, Boudouard, and hydrogenation [86,87]. Thanks to oxygen anions vacancies generated on the support during a redox cycle (Ce 4+ /Ce 3+) [88] and the different interfaces NiO/CeO2±δ, PdO/CeO2±δ, NiO-PdO/CeO2±δ [65], nanoparticles are capable of increasing the interactions between the water vapor molecules (H2O(g)) adsorbed in oxygen vacancies [89] with the hydrocarbons, cracking and avoiding addition reactions [90,91]. It is important to emphasize that no injection of a hydrogen donor agent was made. After, nitrogen injection was carried out in order to obtain the base gas permeability estimated as 771.9 mD. However, at this point, an increase in oil recovery from 72.3% to 75.3% was obtained. This incremental is associated with the steam condensation in the porous media, which worked as a divergent agent for the injected nitrogen, allowing the contact of areas that were not previously invaded by the steam. Finally, the injection of 1 PV of foaming nanofluid was made. The gas permeability at this point was established at 302.2 mD, decreasing approximately 60%. The in situ formation of the foam was carried out successfully and is corroborated by the change in conductivity from 71.15 to 26 cm 3 ·min −1 ·psi −1 for the base system and after the foam injection, respectively. Again, continuous steam injection was made, obtaining an additional recovery of 6.5%.  In the case of the pressure profile in panel b from Figure 3, initially the pressure drops due to the oil displacement by the steam injected into the porous medium. Then, the injection of the nanofluid does not generate a significant change in the system pressure. Besides, the steam injection after nanofluid injection, promotes an additional sudden drop in pressure due to the improvement in oil quality with less restriction to flow.
For nitrogen injection, an increase in pressure is observed because the condensed liquids could cause a divergent effect. When the pressure values stabilize, the foaming agent is injected, and the The absolute and oil effective permeability were estimated as 3627 and 2790 mD, respectively. After the first steam injection in the absence of nanoparticles, an oil recovery of 63.5% was obtained, and a pressure drop of 10 psi. In this case, the recovery is mainly due to the effect of oil viscosity reduction [69], thermal expansion [70], variation in capillary pressure, and relative permeability of the phases [36], volatilization of light hydrocarbons (C 3 -C 5 ) [71], and gravitational segregation [72] caused by the transfer of steam energy to both fluids and to the porous medium. The steam injection without nanoparticles stopped after the injection of 1 equivalent pore volume of water (PVWE).
Then, in the case of the in situ upgrading, a dosage of 0.5 PV of CeNi0.89Pd1.1 nanoparticles-based nanofluid and a soak time of 12 h were employed. In this scenario, additional recovery of 8.8% is observed. The effect of the nanoparticles on the increase in the recovery obtained is mainly explained by for main mechanisms including: viscosity reduction, wettability alteration, enhanced thermal conductivity of the porous medium, and oil upgrading. First, the nanoparticles lead to the reduction in viscosity caused by a re-organization of the crude oil viscoelastic network [17,73,74]. This mechanism is explained considering that nanoparticle-asphaltene interactions are much stronger than asphaltene-asphaltene interactions [34,57,75,76], for this reason, asphaltene molecules are more prone to migrate from the bulk phase to the surface of the nanoparticles, generating a re-organization of the oil microstructure [74,77]. Consequently, the asphaltene aggregation system in the bulk phase will be affected as individual molecules and the smaller aggregates will be rapidly adsorbed by the CeNi0.89Pd1.1 nanoparticles [65]. On the other hand, the second mechanism associated with the enhanced oil recovery by the addition of nanoparticles is associated with the wettability alteration of the porous medium. It has been reported that nanoparticles allows the decoration of the porous media with CeNi0.89Pd1.1 nanoparticles which alter the system wettability to a strong water-wet, which favors the oil flow in the system [17].
After, nitrogen injection was carried out in order to obtain the base gas permeability estimated as 771.9 mD. However, at this point, an increase in oil recovery from 72.3% to 75.3% was obtained. This incremental is associated with the steam condensation in the porous media, which worked as a divergent agent for the injected nitrogen, allowing the contact of areas that were not previously invaded by the steam. Finally, the injection of 1 PV of foaming nanofluid was made. The gas permeability at this point was established at 302.2 mD, decreasing approximately 60%. The in situ formation of the foam was carried out successfully and is corroborated by the change in conductivity from 71.15 to 26 cm 3 ·min −1 ·psi −1 for the base system and after the foam injection, respectively. Again, continuous steam injection was made, obtaining an additional recovery of 6.5%.
In the case of the pressure profile in panel b from Figure 3, initially the pressure drops due to the oil displacement by the steam injected into the porous medium. Then, the injection of the nanofluid does not generate a significant change in the system pressure. Besides, the steam injection after nanofluid injection, promotes an additional sudden drop in pressure due to the improvement in oil quality with less restriction to flow.
For nitrogen injection, an increase in pressure is observed because the condensed liquids could cause a divergent effect. When the pressure values stabilize, the foaming agent is injected, and the pressure increases which corroborates the improved foam formation. Finally, during steam injection, a pressure incremental is maintained and then, decreases due to crude oil displacement from the non-invaded area.
Finally, Table 3 shows the oil and water saturation states in each stage, as well as the ultimate oil recovery and incremental recovery. Here it is important to clarify that saturation states were determined regarding the system prior to the first oil recovery by steam injection in order not to overestimate the volumes recovered and the oil recovery factor.

Temperature Profile during Steam Injection Tests
The temperature profile through 6 nodes located at the beginning, inside, and at the exit of the porous media, for the steam injection scenarios with and without nanocatalysts is shown in Figure 4. The temperature in each of the nodes was measured using integrated thermocouples along the steam generator output line and the entire sample-holder in order to corroborate the improved thermal conductivity. The drop-in temperature in general is due primarily to the tortuosity of the medium [92,93], to the energy transfer from the steam to the fluids (water and oil) [94], and the composition of the rock (99% silica) can be considered as heat sinks [95]. On the other hand, the change in temperature for the system after the nanofluid injection was lower than in the absence thereof. In the absence of nanocatalyst the temperature drops 127 • C, while after nanofluid injection it is 95.5 • C. Besides, in each node, the temperature tested is always lower for the first scenario compared with the second. The chemical structure that nanoparticles acquire with the metal impregnation improves the conductivity of porous media and hence the heat of transfer is higher than without nanoparticles. Also, the system due to nanoscale metal oxides allows the adsorption of heat very quickly, improving the heat capacity of the injected steam when the molecules interact within the porous medium [96][97][98].

Temperature Profile during Steam Injection Tests
The temperature profile through 6 nodes located at the beginning, inside, and at the exit of the porous media, for the steam injection scenarios with and without nanocatalysts is shown in Figure 4. The temperature in each of the nodes was measured using integrated thermocouples along the steam generator output line and the entire sample-holder in order to corroborate the improved thermal conductivity. The drop-in temperature in general is due primarily to the tortuosity of the medium [92,93], to the energy transfer from the steam to the fluids (water and oil) [94], and the composition of the rock (99% silica) can be considered as heat sinks [95]. On the other hand, the change in temperature for the system after the nanofluid injection was lower than in the absence thereof. In the absence of nanocatalyst the temperature drops 127 °C, while after nanofluid injection it is 95.5 °C. Besides, in each node, the temperature tested is always lower for the first scenario compared with the second. The chemical structure that nanoparticles acquire with the metal impregnation improves the conductivity of porous media and hence the heat of transfer is higher than without nanoparticles. Also, the system due to nanoscale metal oxides allows the adsorption of heat very quickly, improving the heat capacity of the injected steam when the molecules interact within the porous medium [96][97][98]. The crude oil upgrading is estimated based on changes in API gravity values. Figure 5 shows the values obtained from the calculation of the API gravity for the oil produced after the steam

API Gravity, Residue Conversion (R%) and SARA Content
The crude oil upgrading is estimated based on changes in API gravity values. Figure 5 shows the values obtained from the calculation of the API gravity for the oil produced after the steam injection, without nanoparticles, steam injection after soaking with nanofluid for in situ upgrading, N 2 injection, and steam injection after foaming nanofluid. The API increases by 6.7%, 46.5%, 50%, and 51.7% for each stage, respectively The change in API gravity is related to the decomposition of the asphaltenes molecules adsorbed on the nanoparticles' surface that leads to the formation of components of lower molecular weight by the development of gasification reactions between H 2 O (g) molecules, heavy hydrocarbons, and functional groups on the surface of the NPs [35,65,66,99]. The presence of only steam achieves a slight change in API gravity possibly because at that temperature, there may be a rearrangement of the molecules within the fluid achieving a small change in the fraction of the heavy oil fractions [100]. Meanwhile, the nanofluid injection promotes the catalytic oil upgrading and the hydrogen production generated from water and water-gas shift reaction [101]. Hence, free alkyl chain molecules convert into lighter molecules, and the hydrogenation of the cracked radicals is performed to avoid addition reactions [102]. N2 injection, and steam injection after foaming nanofluid. The API increases by 6.7%, 46.5%, 50%, and 51.7% for each stage, respectively The change in API gravity is related to the decomposition of the asphaltenes molecules adsorbed on the nanoparticles' surface that leads to the formation of components of lower molecular weight by the development of gasification reactions between H2O(g) molecules, heavy hydrocarbons, and functional groups on the surface of the NPs [35,65,66,99]. The presence of only steam achieves a slight change in API gravity possibly because at that temperature, there may be a rearrangement of the molecules within the fluid achieving a small change in the fraction of the heavy oil fractions [100]. Meanwhile, the nanofluid injection promotes the catalytic oil upgrading and the hydrogen production generated from water and water-gas shift reaction [101]. Hence, free alkyl chain molecules convert into lighter molecules, and the hydrogenation of the cracked radicals is performed to avoid addition reactions [102]. The similarity in the API gravity values obtained for the effluents collected during nitrogen injection and after foam formation with the steam injection assisted by CeNi0.89Pd1.1 nanocatalysts is possibly related to retention phenomena and gravitational segregation of nanoparticles, caused by steam condensation. In this order, nanoparticles achieved contact with non-invaded zones and generated an upgrading in crude oil quality that later was mobilized by nitrogen current, and steam current after the foam formation. These results agree with the reported by our research group [17] and Hamedi et al. [103,104].
On the other hand, Figure 6 shows the percentage in the mass fraction of saturates, aromatics, resins, and asphaltenes content of the effluents obtained in each stage. For the injection of steam without nanoparticles, no appreciable or significant change in the asphaltene content is observed. It is due to the thermal decomposition of these compounds occurring at much higher temperatures, around 450 °C [34,35,66]. The presence of CeNi0.89Pd1.1 nanoparticles generates a decrease from 28.7% to 12.9%, 12.9%, and 14.5% of the asphaltenes content in the system for the respective stage after the injection of nanofluid for in situ upgrading, nitrogen injection, and divergent agent injection, respectively, demonstrating the high catalytic activity of the catalyst. Thanks to the beneficial effect between the metal oxides on the surface of the nanoparticles [83], the development of the redox cycle by the support [65], the anions oxygen vacancies provided by Ce 3+ ions for the adsorption of water vapor molecules [105], and the catalysis of reactions that favor hydrogen production [106], a great improvement in the quality of the heavy crude oil is obtained, reducing the presence of heavy oil compounds, increasing the saturated components, and reducing in turn the presence of aromatic The similarity in the API gravity values obtained for the effluents collected during nitrogen injection and after foam formation with the steam injection assisted by CeNi0.89Pd1.1 nanocatalysts is possibly related to retention phenomena and gravitational segregation of nanoparticles, caused by steam condensation. In this order, nanoparticles achieved contact with non-invaded zones and generated an upgrading in crude oil quality that later was mobilized by nitrogen current, and steam current after the foam formation. These results agree with the reported by our research group [17] and Hamedi et al. [103,104].
On the other hand, Figure 6 shows the percentage in the mass fraction of saturates, aromatics, resins, and asphaltenes content of the effluents obtained in each stage. For the injection of steam without nanoparticles, no appreciable or significant change in the asphaltene content is observed. It is due to the thermal decomposition of these compounds occurring at much higher temperatures, around 450 • C [34,35,66]. The presence of CeNi0.89Pd1.1 nanoparticles generates a decrease from 28.7% to 12.9%, 12.9%, and 14.5% of the asphaltenes content in the system for the respective stage after the injection of nanofluid for in situ upgrading, nitrogen injection, and divergent agent injection, respectively, demonstrating the high catalytic activity of the catalyst. Thanks to the beneficial effect between the metal oxides on the surface of the nanoparticles [83], the development of the redox cycle by the support [65], the anions oxygen vacancies provided by Ce 3+ ions for the adsorption of water vapor molecules [105], and the catalysis of reactions that favor hydrogen production [106], a great improvement in the quality of the heavy crude oil is obtained, reducing the presence of heavy oil compounds, increasing the saturated components, and reducing in turn the presence of aromatic content. Finally, the residue conversion (%R) follows the order: oil after ISU ≈ oil after N2 injection > oil after foam injection > oil after steam injection in the absence of nanoparticles. In each system, the obtained %R was 30.5 ± 0.2, 30.3 ± 0.2, 14.3 ± 0.2, and 3.6 ± 0.4, respectively. These results corroborate the formation of hydrocarbons of lower molecular weight and the suppression of the addition reactions, in agreement with the increase in the content of saturated components, and the decrease of the asphaltenes compounds.

Rheological Behavior
The rheological behavior of the oil obtained after the different steps of the displacement tests is showed in Figure 7, where it is observed that to the extent that heat enters the porous media at residual oil conditions, the viscosity values fall exponentially with the increase in shear rate, indicating the pseudoplastic nature of the virgin extra-heavy oil before and after steam injection [107]. However, an initial decrease in viscosity values is observed after the injection of steam because the viscous forces are surpassed by the kinetic energy of the system [108]. On the other hand, the presence of CeNi0.89Pd1.1 nanocatalysts generates a considerable reduction in viscosity related to the viscoelastic network disruption, besides the asphaltenes aggregates sizes, are reduced, making the asphaltene-asphaltene interactions limited by the increase in nanoparticle-asphaltene interactions [109,110]. This behavior is similar to the reported by Cardona et al. [36] and Shokrlu et al. [111].
The typical oil behavior is non-Newtonian, where the viscosity changes with the shear rate and becomes more noticeable at the lower values; notwithstanding, the nanocatalyst injection modifies the rheological properties making that the EHO behaves more as a Newtonian fluid and reducing its viscosity in all the shear rate range [61]. In this sense, the effluent obtained after the CeNi0.89Pd1.1 base-nanofluid injection, presents the best performance in terms of viscosity reduction, and rheological behavior; due to when the shear rate increases, the behavior of the fluid analyzed begins to be like a Newtonian liquid, since the internal structure of the fluid does not change drastically, independently of the shear rate. Thus, independently of the shear rate, the internal structure of the fluid recovered after the CeNi0.89Pd1.1 base-nanofluid injection does not change drastically. Finally, Cross model parameters are shown in Table S2 of   Finally, the residue conversion (%R) follows the order: oil after ISU ≈ oil after N 2 injection > oil after foam injection > oil after steam injection in the absence of nanoparticles. In each system, the obtained %R was 30.5 ± 0.2, 30.3 ± 0.2, 14.3 ± 0.2, and 3.6 ± 0.4, respectively. These results corroborate the formation of hydrocarbons of lower molecular weight and the suppression of the addition reactions, in agreement with the increase in the content of saturated components, and the decrease of the asphaltenes compounds.

Rheological Behavior
The rheological behavior of the oil obtained after the different steps of the displacement tests is showed in Figure 7, where it is observed that to the extent that heat enters the porous media at residual oil conditions, the viscosity values fall exponentially with the increase in shear rate, indicating the pseudoplastic nature of the virgin extra-heavy oil before and after steam injection [107]. However, an initial decrease in viscosity values is observed after the injection of steam because the viscous forces are surpassed by the kinetic energy of the system [108]. On the other hand, the presence of CeNi0.89Pd1.1 nanocatalysts generates a considerable reduction in viscosity related to the viscoelastic network disruption, besides the asphaltenes aggregates sizes, are reduced, making the asphaltene-asphaltene interactions limited by the increase in nanoparticle-asphaltene interactions [109,110]. This behavior is similar to the reported by Cardona et al. [36] and Shokrlu et al. [111]. With the values from Table S2, it is observed that for all fluids there is a RMSE% < 1%, which means that the model adjusts to the rheological behavior exhibited by the crude oil studied in the different stages. For the extreme viscosity parameters at zero and infinity, the values decrease in the order virgin EHO < steam injection < foam injection < nitrogen injection < steam injection assisted by CeNi0.89Pd1.1 base-nanofluid, corroborating the change in properties of the fluid through its stimulation in steam injection and nanotechnology application. Figure 8 shows the percentage of viscosity reduction (%VRD) for a shear rate value of 10 s −1 , such that the change for each of the recovered fluids is compared to untreated extra-heavy oil. This is the first measurement, the steam injection generates a %VRD of approximately 7.2%, due to the latent heat transferred by the steam towards the formation fluids, increasing the crude oil mobility, without generating a significant increase in the quality of the same. On the other hand, the The typical oil behavior is non-Newtonian, where the viscosity changes with the shear rate and becomes more noticeable at the lower values; notwithstanding, the nanocatalyst injection modifies the rheological properties making that the EHO behaves more as a Newtonian fluid and reducing its viscosity in all the shear rate range [61]. In this sense, the effluent obtained after the CeNi0.89Pd1.1 base-nanofluid injection, presents the best performance in terms of viscosity reduction, and rheological behavior; due to when the shear rate increases, the behavior of the fluid analyzed begins to be like a Newtonian liquid, since the internal structure of the fluid does not change drastically, independently of the shear rate. Thus, independently of the shear rate, the internal structure of the fluid recovered after the CeNi0.89Pd1.1 base-nanofluid injection does not change drastically. Finally, Cross model parameters are shown in Table S2 of the Supplementary Material Document, along with the %RMSE values. With the values from Table S2, it is observed that for all fluids there is a RMSE% < 1%, which means that the model adjusts to the rheological behavior exhibited by the crude oil studied in the different stages. For the extreme viscosity parameters at zero and infinity, the values decrease in the order virgin EHO < steam injection < foam injection < nitrogen injection < steam injection assisted by CeNi0.89Pd1.1 base-nanofluid, corroborating the change in properties of the fluid through its stimulation in steam injection and nanotechnology application. Figure 8 shows the percentage of viscosity reduction (%VRD) for a shear rate value of 10 s −1 , such that the change for each of the recovered fluids is compared to untreated extra-heavy oil. different stages. For the extreme viscosity parameters at zero and infinity, the values decrease in the order virgin EHO < steam injection < foam injection < nitrogen injection < steam injection assisted by CeNi0.89Pd1.1 base-nanofluid, corroborating the change in properties of the fluid through its stimulation in steam injection and nanotechnology application. Figure 8 shows the percentage of viscosity reduction (%VRD) for a shear rate value of 10 s −1 , such that the change for each of the recovered fluids is compared to untreated extra-heavy oil. This is the first measurement, the steam injection generates a %VRD of approximately 7.2%, due to the latent heat transferred by the steam towards the formation fluids, increasing the crude oil mobility, without generating a significant increase in the quality of the same. On the other hand, the This is the first measurement, the steam injection generates a %VRD of approximately 7.2%, due to the latent heat transferred by the steam towards the formation fluids, increasing the crude oil mobility, without generating a significant increase in the quality of the same. On the other hand, the catalytic effect of the nanoparticles, their capacity to mediate water-gas shift reactions for the hydrogen production taking advantage of the formation of CO [103], and considering the lower activation energy required to break the C-S bond [112], added to the high electronegativity of S and that the H + ions attack mainly sulfur atoms, with the effect of high pressure and temperature, reactions of aquathermolysis are developed [71], breaking C-S and C = S bonds, interrupting the crude oil viscoelastic network and considerably decreasing crude oil viscosity by approximately 79.39%, 77.60%, and 31.25%, for the three scenarios evaluated, respectively.
Considering the previous results, among the advantages offering by the use of CeNi0.89Pd1.1 nanocatalysts to assist the steam injection processes are: (1) reduction of heat consumption, the catalyst in contact with oil molecules accelerate the cracking and decomposition of heavy hydrocarbons, promoting more heat and gaseous products that further improve the oil displacement; (2) reduction of water consuming, because with its application, it is possible to achieve the same recovery values with much less steam injected. This behavior is consistent with several reported studies [17], even leading to a decrease in the water cut [113]; and (3) reduction of toxic chemicals and capital costs of lifting and transportation operations. The main techniques that exist for the OH and EHO transportation require the addition of solvents (light crude oil, naphtha, diesel, among others) or gasses (mainly, CO 2 ), as they contribute to reducing oil viscosity [114][115][116][117]. High consumption of solvents increases transportation costs and generates different environmental risks due to the production of polluting gasses. It has been shown that with the addition of 0.2% by mass of CeNi0.89Pd1.1 nanocatalyst, a reduction of approximately 80% of viscosity is achieved. Finally, (4) Reduction of greenhouse gases, CO 2 leakage prevention mechanism through nanoparticle application have been reported. All advantages mentioned make evident that nanoparticles applied in EOR processes have the ability to improve oil production being environmentally friendly, less expensive, and more efficient compared to traditional treatments [118].

Conclusions
Thanks to the catalytic advantages that CeO 2 -based materials present, the nanocatalyst employed in this study showed a high performance in terms of oil recovery factor and quality improvement of extra-heavy oil. In this way, comparing the scenario in the absence of these nanomaterials, there is a significant increase in all the properties of crude oil. Starting with an increase in oil recoverable, followed by improvement in properties such as a decrease in asphaltene content, increase in API values, and viscosity reduction. Increases in oil recovery from 63.5% to 72.3%, 75.3%, and 80.8% are obtained for the injection of nanofluid, nitrogen, and foaming agent. Besides the quality of the EHO was improved, reducing its asphaltene content in 15.8%, 15.8%, and 14.2% unites for the same respective stages. Finally, the implementation of CeNi0.89Pd1.1 nanocatalysts shows improvements of the steam injection processes, to the extent that it was not necessary to use a hydrogen donor agent to promote its high performance.