2.1 Materials

Empty fruit bunch (EFB) fibres were obtained from Felda Palm Oil Mill, Kulai, Johor, Malaysia. Ethanol (biological grade with a molecular weight of 46.07 g/mol), with a purity of 99.8% (v/v) was purchased from R & M Chemicals. Sodium chloride (NaCl, molecular weight 58.44 g/mol) with a purity of 99.8% assay was obtained from QReC^TM. Heavy mineral oil with a viscosity of 240 mPa.s and density (0.92 g/cm³) was obtained from Merck KGaA, Germany. Unconsolidated sandpack was used as the porous media (Table 2).

2.2 Methods

2.2.2

2.2.2 Formulation of EFBSNP

About 40 g of the thermally treated EFBA was placed in a grinding jar made of tungsten carbide, thereafter, round balls made of tungsten carbide was used to fill the grinding jar up to 150 mL. The grinding jar (holder) was placed eccentrically in a Retsch PM 100 planetary ball mill equipment (Fig. 1a). Dry milling of EFBA was initiated for 20 min (rotating-reversing motion) at 10 min intervals for 2 h. Subsequently, 10 mL of ethanol was added to the EFBA to act as a control agent of the process (Wang et al., 2017; Zhang et al., 2018; Agi et al., 2020b). This time the ball speed (500 rpm) and the grinding jar created an interaction (frictional forces) releasing high energy to the EFBA for 4 h (Fig. 1b). Consequently, the exchange of these forces resulted in comminution. The sample was then washed with ethanol and placed in an ultrasonic bath (Genesis™ XG-500-6, 40 kHz, 500 W) for 1 h. Thereafter, the sample was centrifuged and dried in an oven for 24 h. Fig. 2 shows the schematic illustration of the formulation process.

Close

Fig. 1

Schematic illustration of the working principle of planetary ball mill showing (a) general layout of the disk with the arrows indicating the counter-rotation of the supporting disc and (b) horizontal cross-section of grinding jar with the arrows indicating the counter-rotation of the milling beaker.

Export to PPT

Close

Fig. 2

Schematic illustration of EFBSNP formulation process.

Export to PPT

2.2.13

2.2.13 Oil displacement experiments

The experimental setup for the oil displacement experiments consists of a cylindrical sand pack made with a polyvinyl chloride containing sandstone of size range 125–250 µm. The sand pack was used as a porous medium to investigate the oil displacement ability of EFBSNF under ultrasound (Table 2). The joints of the sand pack were sealed with adhesives to prevent any leakage during the experimental process. Sieves placed at both ends of the cylinders prevented fines migration. Thereafter, the properties of the porous media were determined (Table 2). Then the sand pack was placed in a holder and immersed in the ultrasonic bath to achieve maximum ultrasonic exposure (Fig. 3). The ultrasonic generator (Genesis™ XG-500–6, 500 W) delivered high frequency (40 kHz) waves through the water-bath with the immersion transducer. This frequency was chosen based on the application as high frequency increases the number of free radicals (electron transfer) (Thompson and Doraiswamy, 1999). Subsequently, 2.2 wt% NaCl (brine) was introduced into the sand pack with aid of a syringe pump (NE-1000) until it was completely flooded (saturated). Heavy oil was introduced at 2.0 mL/h until connate water saturation was achieved. Thereafter, the water saturation and oil saturation were determined. The sand pack was then flooded with water until breakthrough occurred. In this study EFBSNF flooding and ultrasound assisted EFBSNF were considered as EOR process. Hence to extract the bypassed oil, 0.5 PV of 0.05–0.2 wt% EFBSNF was injected and then ultrasound wave was used for 15 min to assist EFBSNF flooding. Ultrasound was applied under uncontrolled temperature and the temperature of the system increase to 65 °C. To extract any left-over oil 0.5 PV of chase water was injected into the system. The experiments were repeated twice and the average reported.

Close

Fig. 3

Schematic illustration of ultrasound assisted EFBSNF flooding in a simulated porous media.

Export to PPT

3.1 Thermally treated EFBA

The chemical composition of EFBA before and after thermal treatment was determined by XRF (Table 3). The EFBA before treatment showed 60% SiO₂ and trace-elements such as magnesium oxide (MgO), aluminium oxide (Al₂O₃), phosphorus pentoxide (P₂O₅), potassium oxide (K₂O), calcium oxide (CaO), titanium oxide (TiO₂), and iron (iii) oxide (Fe₂O₃). Thermal treatment was used to improve the SiO₂ content to serve as an alternative source of SiO₂. The XRF results (Table 3) show a significant decrease in the concentration of the trace element with increasing temperature and a corresponding increase in SiO₂ content. This is because when phase transition occurs (when temperature exceed 573 °C), SiO₂ crystal lattice structure changes. Heating will cause the impurities to be mobile and diffuse from the inclusion to the crystal lattice under thermodynamic driving force. However, diffusion during the phase transition is accelerated by thermodynamic forces because SiO₂ has a loose structure, therefore some of the impurities enter the SiO₂ crystal lattice thereby increasing the SiO₂ phase (Jani and Hogland, 2017; Li et al., 2020). This result is consistent with previous study of Kenes et al. (2012), they reported that higher temperature causes thermal decomposition of organic substance in rice husk and hence the SiO₂ content increases.

3.2 Morphology results

The SEM image of untreated EFBA (Fig. 4a) shows spherical, angular and irregular shapes. The treated EFBA showed a decrease in spherical shape with temperature (Fig. 4b), which can be attributed to the collapse of some micro and mesopores at high temperature (Liu et al., 2013). The particle size also increased with temperature (Fig. 4b). At high temperature, crystal growth is enhanced, which promotes the emergence of a more compact structure. Similar result was reported by Chen et al. (2012) when they stated that particle structure becomes more packed when the temperature exceeds 600 °C. The result agrees with previous studies by Fernandes et al. (2017) who found that high temperature decreases the specific surface area values and thus increases the diameter of the particles. TEM was used to confirm the shape and size of EFBSNP (Fig. 5). The size of EFBSNP ranges from 17.78−115.38 nm. The larger particles have a platy and irregular shape (Fig. 5a), while the smaller particles have a spherical shape (Fig. 5b). The irregular shape could be due to cavitation during wet milling. During shearing of ethanol during the wet milling, the wetted particles forms nucleus resulting in nucleation of individual particles coated with a layer of liquid, low pressure produces liquid bubbles (Fig. 2), the subsequent growth and collapse of these bubbles produces cavitation effect which improves the particle size reduction efficiency of the ball mill (Sen, 2017; Agi et al., 2020b). Subsequent ultrasonic treatment of EFBSNP changed the shape of the particles from irregular and angular to spherical (Ahmad et al., 2020). This confirms that the change in size and shape during the thermal treatment of EFBA had no effect on the resultant EFBSNP after wet milling but rather it increased the surface stability (Reynolds et al., 2016).

Close

Fig. 4

SEM images of EFBA showing shapes and sizes (a) untreated having spherical, angular and spherical shapes with small particle sizes and (b) treated at 1100 °C having larger particle sizes.

Export to PPT

Close

Fig. 5

TEM images of EFBSNP showing sizes and shape (a) platy and irregular shapes due to cavitation during wet milling and (b) spherical shape and small size particles due to ultrasonication.

Export to PPT

3.3 Mechanisms of EFBSNP synthesis via ultrasonic-assisted wet milling

During the process of dry milling, a mild compacting force was propagated to produce a mixture of fine micro particles. The alcohol inhibited the aggregation of the samples during wet grinding. Therefore, the alcohol adsorbed the particles by shielding the attractive force while promoting steric stabilization. Nevertheless, the adsorption of alcohol prevented agglomeration through the mechanism of electrostatic stabilization (Loh et al., 2015). Then the mechanical energy of the mill (impact of mill ball) exerts a tension on the particles, which leads to nucleation (first step in the formation of a new phase). A high rotational speed of the ball mill (500 rpm) increases the collision between the particles or between the particle and the wall and increases the energy transfer to the particles leading to coalescence (Fig. 2). In addition, the revolution of the ball mill might have increased the temperature forcing the internal liquid and air out and forming liquid film on the surface. This process accelerates the size increase and growth of the particles (Wang et al., 2017). The final stage in the mechanism of wet milling is the breakage and attrition, which is initiated by the shear force of the ball mill. This occurs when the applied kinetic energy exceeds the energy the granules can resist (Agi et al., 2020b). After wet milling, some EFBSNP was assumed to remain in the aggregates (Fig. 5a). Therefore, ultrasonic treatment was used to dissociate the EFBSNP in the aggregates to form radicals which may have induced the degradation/detexturation and sono-erosion of EFBSNP altering the morphology (Fig. 5b).

3.4 Size distribution and surface charge of EFBSNP

The size distribution of EFBSNP reduced to nanosize by wet milling and ultrasonic treatment compared to the micron size of the untreated (supplementary information Fig. S1-2). EFBA (Figure S1) showed a broad size distribution consistent with the different sizes and shapes of the SEM result (Fig. 4). In contrast, EFBSNP (Figure S2) exhibited a narrow size distribution, which could be attributed to the effect of ultrasonic homogenization that controlled the particle size, resulting in a narrow size distribution (Gokce et al., 2014). Zeta potential was used to determine the stability of EFBSNP in solution. As a reference, zeta potential values within ±10 mV, ±10–20 mV, ±20–30 mV and ±30 mV suggest very unstable, reasonably stable, moderately stable, and highly stable nanofluids, respectively (Chakraborty and Panigrahi, 2020). Fig. S3-4 show the zeta potential and surface charge of EFBA and EFBSNP. The surface chemistry imparts stability to the colloids, which greatly influence the flow behaviour of nanosuspension. Therefore, the zeta potential can affect the stability of solutions. This is because the electrostatic repulsion between particles is greater at high zeta potential. Figure S4 shows that the zeta potential of EFBSNP can be affected by thermal treatment. This is because above 600 °C, the ratio of inaccessible silanol group to associated silanol increases, which may have increased the surface charge of EFBSNP (Tettey and Lee, 2013). Consequently, EFBSNP zeta potential at different concentration lies between −36.7 and 38 mV which is higher compared to EFBA at different concentration (−26.1 to 28.8 mV) (Figures S3 and 4). This implies that pH has a direct effect on zeta potential since the electrostatic forces were retained by the pH during wet grinding. In the preparation of nanoparticles during wet milling, pH is a vital parameter (Sakthivel et al., 2007). Kaya and Yukselen (2005) recommended a pH of 7.2 for quartz materials. Therefore, the pH of 7.3 for the alcohol used might have limited deposition of EFBSNP and maintained the electrostatic and steric stabilization of the crystals in EFBSNP to produce a very stable nanoparticle through wet milling. Nevertheless, the constant pH of the medium implies that the ultrasound had no significant effect on the zeta potential (Jafari et al., 2014). Also, the particle size distribution, zeta potential of EFBA and EFBSNP at different concentration showed significant similarity this implies that the change in concentration had no significant impact on the size distribution and zeta potential.

3.5 Crystalline structure of EFBSNP

The XRD results of EFBA show the presence of amorphous and crystalline phases (Fig. 6a). Two crystal phases of SiO₂ (quartz and tridymite) were identified. Quartz was identified as the major crystalline phase with at sharp peak of 2θ at 26.57° and at 20.8°, 50.08° and 59.8°. These peaks correspond to the reflection from 011, 110, 112 and 121. Also, tridymite crystalline phase of 2θ at 21.8°, 28.3°, 31.1° and 36.4° corresponds to 101, 111, 102 and 200 crystal planes of alpha-SiO₂. These results matched with the standard given in the international centre of diffraction data (ICDD) 01-077-1060. This implies that the crystal plane of the SiO₂ is hexagonal with a lattice parameter of a-axis (4.9 Å), b-axis (4.9 Å) and c-axis (5.4 Å). This indicates that the EFBA has both quartz and tridymite peaks. Similar result was reported by previous studies of Fernandes et al. (2017); Borouni et al. (2018) and Imoisili et al. (2020). Subsequently, the thermal treatment affected the crystalline phase of the EFBSNP (Fig. 6a). This is because high temperature assists the formation of the crystalline phase due to the energy added to the system (Fernandes et al., 2017). For example, crystallization of quartz occurred between 573 and 1055 °C, while tridymite crystallized at 867 °C (Fernandes et al., 2017; Borouni et al., 2018). Consequently, the crystallization rate increased from 10% (EFBA) to 46.78% (EFBSNP). This could be due to a diffusion-induced mechanism during the wet milling that may have influenced the kinetic transformation to a more crystalline phase (Agi et al., 2020c) Also, the ultrasonic treatment could have decomposed the molecular chains of EFBSNP by shearing and exposed the hydroxyl group. This enhanced the interaction between the EFBSNP and the alcohol molecules, leading to the destruction of the molecules in solution and recrystallization of the EFBSNP (Lin et al., 2020). Nevertheless, a residual amorphous phase remains after the temperature treatment (Drisko et al., 2015).

Close

Fig. 6

(a) XRD results of EFBA and EFBSNP showing crystalline and amorphous phases (b) comparison of the FTIR spectra of EFBA and EFBSNP and (c) DSC thermograms of EFBA and EFBSNP showing different endothermic peaks.

Export to PPT

3.6 Chemical compound and structure

The functional group and chemical compound of the EFBA and EFBSNP were determined by FTIR (Fig. 6b). The EFBSNP demonstrated some structural differences compared to the EFBA. Considerable number of adsorption bands disappeared while some decreased. The band at 650–1368 cm⁻¹ is a standard feature of SiO₂. The band at 776 cm⁻¹ was ascribed to (-Si-O-Si) symmetric whereas the band at 1032.31 cm⁻¹ to Si-OH stretching vibrational bond, which are the main peaks of SiO₂. The peak at 1742.32 cm⁻¹ was attributed to the H-O-H of adsorbed water, which is because of –OH adsorption band (Jafari et al., 2014). Small band occurring at 3749 cm⁻¹ was assigned to sequence in silanol group (Si-O-H) and hydrogen bonding of water to SiO₂. With increasing thermal treatment, the peak at 3749 cm⁻¹ fades, the disappearance of the peak at higher temperature indicates the evolution of CO₂ at higher temperature (Fernandes et al., 2017). The gradual decrease in peak height at 1032.31 cm⁻¹ with increasing temperature is due to the removal of physiosorbed water (Tettey and Lee, 2013). Initially, the amount of inaccessible silanol group increased but with increase in temperature, the isolated silanol group undergoes a condensation reaction to form siloxane bonds and water vapour as a by-product (Tettey and Lee, 2013). Consequently, the silanol peak reduced at temperatures above 700 °C. After thermal treatment up to 1200 °C, the silanol group may be present to some extent (Tettey and Lee, 2013). Nevertheless, –OH intensity at 1742.32 cm⁻¹ shows an increase (EFBSNP) compared to EFBA. This could be attributed to the ultrasonic treatment as more free –OH groups are released with the decrease in size (Gu et al., 2020). The non-appearance of any other adsorption band indicates that it is pure SiO₂. Consequently, the position of the characteristic peaks is consistent which means that the synthesis method did not change the chemical composition.

3.7 Thermal properties results

The improvement of thermal properties of nanoparticles can prove uniform dispersion and their ability to interact with other materials (Borouni et al., 2018). Fig. 6c shows the thermal stability of EFBA and EFBSNP. The thermograms show distinct endothermic peaks. The first endothermic peak between 86.96 and 104.77 °C (Fig. 6c) was attributed to the evaporation of absorbed alcohol/water (Wang et al. 2010; Hossain et al., 2019). The second endothermic peak between 239.09 and 293.46 °C corresponds to the rupture of hydrogen bond, which was represented as Si-OH (Wang et al. 2010; Hossain et al., 2019). Consequently, the endothermic peaks of EFBSNP decrease compared to EFBA (Table 3), which could be due to the introduction of a more reactive surface as a result of the synthesis method (Singh et al., 2007). High temperature during thermal treatment supports the formation of crystalline phase due to the energy supplied to the system. Besides, during the wet milling, the ball mill strongly influences the kinetic transformation of amorphous phase to crystalline phase in a reaction and diffusion induced mechanisms (Borouni et al., 2018). Also, ultrasonic treatment might have decomposed the molecular chains of the EFBSNP by shearing, exposing the hydroxyl group, which enhanced the interaction between the EFBSNP and alcohol molecules leading to destruction of the molecules in solution and recrystallization of the EFBSNP. This suggests that EFBSNP acted as an additional nucleation site and agent by lowering the surface free energy barriers towards nucleation thus inducing crystallization. Similar result was reported by previous study of Han et al. (2018) they stated that grafting of hydroxyl group on SiO₂ nanoparticles can improve its compatibility as a nucleating agent and improves its stability. This is consistent with the shift of the first endothermic peak of EFBSNP to higher temperatures compared to EFBA, implying that the ultrasonic treatment accelerated crystallization. Therefore, the higher relative crystallinity of EFBSNP compared to EFBA (Table 4) confirmed that EFBSNP acted as a nucleation site (Amigo et al., 2019). Subsequently, the glass transition temperature (Tg) of EFBSNP increased compared to EFBA. The increase in Tg indicates that a strong physical bond was formed between EFBSNP, hindering their movement at high temperatures (Hong et al., 2007). This implies that the mobility of EFBSNP subsequently decreased and became ordered (Soleimani and Mohammadi, 2018).

3.8 Surface chemical composition of EFBA and EFBSNP

XPS analysis was carried out to confirm the existence of silicon (Si) and oxygen (O) elements in EFBA and EFBSNP. Fig. 7 shows the survey and high-resolution spectra of Si (2p) and O (1s) indicating the presence of Si and O component of SiO₂ in EFBA and EFBSNP. It was observed that the surface composition of EFBSNP was similar to the bulk (EFBA) composition (Fig. 7a-b). Nevertheless, the major difference was observed in the peak occurring around 99 eV for EFBA (Fig. 7c) corresponding to the Si 2p orbit assigned to bulk Si and at 100 eV for EFBSNP (Fig. 7d) which was ascribed to SiO₂. The chemical shift to higher energies was due to the thick SiO₂ layer (Hou et al., 2017). This is because after wet milling, high concentration of SiO₂ was observed on the surface. Also, due to size reduction, the surface of Si was exposed and likely to form a SiO₂ layer as result of the reaction with ethanol during wet milling (Hou et al., 2017). This concurs with previous study of Krishnarao and Godkhindi (1992) they reported that the change in spectra could be because of the bonding of SiO₂ to organic material present in the EFBA. Fig. 7e and f shows the O 1s spectra for EFBA and EFBSNP. The peaks at 538 and 537 eV correspond to Si⁴⁺ core level in EFBA and EFBSNP, respectively. The shift in the O 1s peak to lower energies might be due to the reaction of Si with ethanol during wet milling resulting in the formation of Si-O-CH₂CH₃ on the surface of the EFBSNP (Hou et al., 2017).

Close

Fig. 7

XPS spectra for EFBA and EFBSNP: (a) survey of EFBA (b) survey of EFBSNP (c) high-resolution spectrum of peak Si 2p for EFBA (d) high-resolution spectrum of Si 2p peak for EFBSNP (e) high-resolution spectrum of O 1s peak for EFBA and (f) high-resolution spectrum of O 1s peak for EFBSNP.

Export to PPT

3.9 Ultrasonic absorption performance of EFBSNP

The ability of EFBSNP to absorb ultrasonic waves was investigated. Figs. 8 and 9 show the performance at different concentration of EFBSNF in different fluids. The temperature of both medium (water and oil) increased with ultrasonication time (Figs. 8 and 9). This is because high energy output of ultrasonic energy increases the frictional limit, which led to an increase in the temperature of the samples. However, the addition of EFBSNP increased the temperature compared to water and oil alone. This is because energy is transferred to the medium (oil and water), during the ultrasonic process, resulting in cavitation within the medium. Cavitation energies help to disintegrate agglomerated particles through the reaction force of the bubble implosion. The shear effect created by the implosion increases the heat transfer and thus the temperature increases (Adio et al., 2016). In addition, the ultrasonic treatment increased the Brownian motion of the EFBSNP, which increased the probability of contact between the particles. The collision of the EFBSNP converted the potential energy of the particles into kinetic energy generating heat that increased the temperature of the system. Likewise, the high temperature increase is caused by attenuation of ultrasound energy however, the addition of EFBSNP changed the coefficient of the medium thereby increasing the rate of heating. This agrees with prior work of Sviridov et al. (2019) they reported that mesoporous silicon-based nanoparticles enhanced cavitation effect of ultrasound which led to additional heating of the surrounding and initiation of sonochemical reaction. Nevertheless, the higher temperature in deionized water compared to heavy oil is due to cavitation. Cavitation forms more easily in lighter fluids than in viscous fluids (Hamidi et al., 2014; Agi et al., 2019a). This is because the formation of cavitation in a liquid requires that the negative pressure in the rarefaction region be above the natural cohesive force acting in the liquid (Hamidi et al., 2014). Therefore, cavitation is less intense in viscous fluids where the forces are pronounced (Agi et al., 2019a). Furthermore, the higher temperature in EFBSNF could be due to the high surface charge of EFBSNP (Fig. 5d), which were immediately available for thermal interactions, leading to microconvection and increased heat transfer (Das et al., 2006). Therefore, it can be concluded that the temperature increase is dependent on ultrasonic cavitation and the increase in Brownian motion due to mechanical vibration. Absorption performance of EFBSNP increased with concentration, this is because of the high number of EFBSNP active site and sufficient surface available for absorption. However, above 0. 1 wt% the absorption rate diminished, this might be attributed to gradual saturated of the active site at higher concentration and lack of access to the active site. Similar result was reported by previous study by Montes et al. (2018) they observed a slight increase of only 2% at higher concentration. This signifies that the optimum concentration has been exceeded. Also, the effect of particle size on absorption of ultrasound in different fluids was investigated (Figs. 8 and 9). The EFBSNF showed higher absorption capacity compared to the EFBA in micron size. This could be attributed to the increase in surface area as the particle size decreased. Consequently, more particles were in solution for absorption of ultrasound. It might also be because of the increase in solubility with decreasing particles size (Sandri et al., 2013).

Close

Fig. 8

Performance ability of EFBA (0.2 wt%) and EFBSNF to adsorb ultrasonic wave with time in deionized water.

Export to PPT

Close

Fig. 9

Performance ability of EFBA (0.2 wt%) and EFBSNF to adsorb ultrasonic wave with time in heavy oil.

Export to PPT

3.10 Effectiveness of EFBSNP in heavy oil viscosity reduction

The effects of different concentration of EFBSNF and ultrasound on the viscocity of heavy oil is shown in Fig. 10. The viscosity of the samples decreased with increasing ultrasonic exposure time. This could be due to the fact that as the ultrasonic exposure time increases, a higher number of bubbles are involved in the cavitation process (Montes et al., 2018). Cavitation triggers a chemical reaction that breaks down high molecular weight compunds associated with the viscosity of heavy oil. Therefore, the cavitation effect generates high temperature, high pressure and micro jets that change the internal structure of the oil sample (Montes et al., 2018; Olaya-Escobar et al., 2020). Consequently, several macromolecules in the oil samples were broken-down into smaller fragments and carbon-carbon (C-C) bond of the heavy oil breaks to two free radicals compunds (R* and R’*) with lower molecular weight (Eq. (1)), reducing the viscosity. Also, both hydrogen and hydroxyl radicals are produced during the homolytic dissassociation of water molecules (Eq. (2)). Consequently, hydroxyl radical reacts with alkane (abstraction reaction) to form a new alkyl radical (Eq. (3)) (Lin and Yen, 1993).

Close

Fig. 10

Influence of EFBSNP in reducing the viscosity of heavy oil in the presence of ultrasound.

Export to PPT

Therefore, hydrogen radical may come from water or dissolved hydrogen by ultrasound (Eq. (4)). However, when ultrasound alone was applied, the effect of droplet coalescence was minimal and fewer hydrogen free radicals were produced during the cracking reaction (Montes et al., 2018). Hence, when few hydrogen free radical are produced, the large molecule free radical will experience acumulation to bigger structure impeding viscosity reduction and heavy oil modification (Wan et al., 2019). However, with the addition of EFBSNF, functional hydrogen radical were generated through cavitational impact to end the movement of the macromolecules free radicals. In this way, alkyl radical was terminated by reaction with hydrogen radical (Equation (5)). In similitude, the hydrogen radical reacted with hydroxly or hydrogen radical to form water or hydrogen molecule (Eqs. (6) and (7)). The viscosity of the heavy oil decreased further with increase in concentration of EFBSNP, which might be attributed to the availability of the dissociated hydrogen to freely interact with the macromolecular free radical thereby, preventing growth of the heavy oil, inhibit condensation reaction, promote hydrogenation reaction, improve the quality of the heavy oil and further reduce viscosity.

3.11 Effect of temperature rise on interfacial tension (IFT)

The temperature rise can affect the fluid properties such as IFT and viscosity. The IFT between EFBSNF (0. 1 wt%) and heavy oil was calculated using Firoozabadi and Ramey’s Equation (8) (Firoozabadi and Ramey, 1988; Agi et al., 2018).

Close

Fig. 11

Ability of EFBSNF in reducing IFT of heavy oil as a function of temperature.

Export to PPT

3.11.1

3.11.1 Possible mechanisms for EFBSNF assisted ultrasound in IFT reduction

Fig. 12 depicts the schematic illustration of the possible mechanisms that aided IFT reduction during the EFBSNF assisted ultrasound process. Fig. 12 shows that cavitation helps in the separation of phases exposed to ultrasound. Since the EFBSNP tends to drift towards the O/W interface, the high temperature reduces the rigidity at the interface, making it easier for the droplet to coalesce upon collision. Ultrasound facilitated the adsorption of EFBSNP droplets at the interface forming an interfacial film. Ultrasonication might have stimulated the interfacial film making it more hydrophobic enabling the EFBSNF to adsorb faster at the O/W interface to form a layer (Fig. 12b) which increased the potential between the EFBSNP and imparted a repulsive force between them. Compression and expansion bring stress to O/W interface during ultrasound application, overcoming the linked forces that hold large droplets together and breaking them into smaller ones (Agi et al., 2020c). Hence, O/W boundary was replaced by EFBSNF consequently, a hydrophilic/water and hydrophobic/oil interfaces was formed. The repulsive force was induced by the hydroxyl group of EFBSNP which bonded together with water while the hydrophobic component reacts with oil causing inter-molecular O/W integrated force to reduce thus, decreasing IFT (Sukmarani and Ledyastuti, 2019; Agi et al., 2020a; 2021). In similitude, Sukmarani and Ledyastuti (2019) stated that hydroxyl group can easily bind to water thereby decreasing IFT. Consequently, rapid adsorption of EFBSNP at hydrophobic interfacial film was enabled, which induced steric and electrostatic interactions that synergistically reduced the IFT (Agi et al., 2020b). As a result, the entropy of the whole suspension increases due to the increase in the freedom of EFBSNP, leading to a decrease in Gibbs free energy (Eq. (12)). The decline in Gibbs free energy reduces the free energy at the surface thus, decreasing the IFT (Eq. (13)) (Agi et al., 2020a).

(12)

$$\mathrm{\Delta }G=\mathrm{\Delta }H-T\mathrm{\Delta }S$$

(13)

$$\mathrm{\Delta }G=\gamma \partial A(constantT\&P)$$ whereas $\mathrm{\Delta }G$ is difference in Gibbs free energy (Kj mol⁻¹), $\mathrm{\Delta }S$ is difference in entropy (J.K⁻¹), $\mathrm{\Delta }H$ is difference in enthalpy (KJ), P is pressure (Pa), T is temperature (K) and $\gamma$ is free energy at O/W interface (N/m). A decrease in IFT reduces the capillary pressure generated by the trapped oil droplets in the pores, effectively mobilizing the oil.

Close

Fig. 12

Schematic illustration of IFT reduction between O/W by EFBSNF (a) IFT of O/W due to unbalanced forces (b) oil and EFBSNF with ultrasound (c) adsorption of EFBSNP at O/W interface (d) decrease in is free energy at O/W interface and (e) decrease in IFT.

Export to PPT

3.12 Nanofluid oil displacement without ultrasound

Six different sand packs were used for the initial series of tests (Table 2) to investigate the efficiency of EFBSNF in heavy oil recovery at different concentration (Fig. 13). About 24.5% original oil in place (OOIP) was recovered by waterflooding. This indicates the inefficiency of waterflooding in recovering heavy oil, which can be attributed to viscous fingering and early water breakthrough. Therefore, EFBSNF flooding was investigated to recover the bypassed oil. As a result, the oil recovery increased by 20.5%, 26.33% and 28.5% with 0.05 wt%, 0.1 wt% and 0.2 wt% EFBSNF, respectively (Fig. 13). This increase can be attributed to the improvement in the flow properties of the porous medium as the EFBSNP plugged the channels created by the waterflooding (Cheraghian and Tardasti, 2012; Cheraghian et al., 2020). This resulted in uniform sweep efficiency and hence increased oil recovery (Cheraghian, 2016; Nezhad and Cheraghian, 2016). In addition, flooding with EFBSNF caused a change in the viscous force distribution that had occurred during the initial waterflooding (viscosity difference). These changes resulted in a more stable displacement front due to the reduction in viscosity. Similar result was reported by previous work of Cheraghian et al. (2017) they stated that SiO₂ nanoparticles can modify the flow of displacing fluid due to the change in viscosity. The results are consistent with the efficiency of viscosity reduction by EFBSNF (Fig. 10). Oil recovery increased with concentration of EFBSNF due the increase in viscosity of the nanofluids which mobilize the bypassed oil. However, at higher concentration, the increase in oil recovery was minimal which might be due to pore blocking.

Close

Fig. 13

Oil recovery efficiency of EFBSNF at different concentrations.

Export to PPT

3.13 Ultrasound assisted nanofluid flooding

The optimum concentration of 0.1 wt% EFBSNF was used for the ultrasound experiment. Ultrasound-assisted EFBSNF flooding was used as a tertiary oil recovery method and the results obtained were compared with the first set of experiments without ultrasound. Fig. 14 shows that ultrasound assisted EFBSNF flooding recovered 44.3% OOIP, increasing oil recovery by 18% compared to oil recovery without ultrasound. This is due to the fact that ultrasound caused rapid movement of trapped oil, which was initially bypassed by waterflooding and EFBSNF flooding (Agi et al., 2019c). This movement caused the high molecules of the heavy oil to break up, thereby increasing their mobility (Agi et al.,2018). Similarly, it could be due to the attractive forces between the oscillating oil droplets (Bjerknes forces), leading to coalescence of smaller oil droplets into a single stream with higher mobility thereby, flowing out of the porous medium (Agi et al. 2018). In addition, ultrasound interacts with the intermolecular forces that set it in motion (Brownian forces), and the ultrasonic vibration of EFBSNF causes stresses that led to an increase in pressure and temperature. This is consistent with the pressure drop profile (Fig. 15), as ultrasound assisted EFBSNF flooding showed a higher pressure drop compared to EFBSNF flooding without ultrasound. This high pressure drop indicates a more stable displacement and lower viscosity pushing the oil ahead of the EFBSNF (Hamidi et al., 2017). Nevertheless, the pressure dropped back to lower values after reaching a peak (Fig. 15). This is due to the cavitation effect, where a high-pressure surge is generated in the system during the implosion of the bubbles. However, after the bubble implosion the pressure drops as shown in Fig. 15 (Mohammadian et al., 2013). Furthermore, the temperature rise might have increased the surface activity and hydrophobicity at the interface, reducing the IFT, thereby, resulting in increased oil recovery. As a result, the aqueous phase flows (peristaltic motion) rather than adhering to the porous media (Agi et al., 2019b).

Close

Fig. 14

Oil recovery efficiency of 0.1 wt% EFBSNF with and without ultrasound compared to water flooding.

Export to PPT

Close

Fig. 15

Pressure drop profile of EFBSNF flooding with and without ultrasound compared to water flooding.

Export to PPT

3.13.1

3.13.1 Possible mechanisms for EFBSNF assisted ultrasound oil recovery

The possible heavy oil recovery mechanisms were proposed based on heavy oil displacement experimental results. Fig. 16 depicts the schematic illustration of the possible oil recovery mechanisms during EFBSNF assisted flooding. For effective heavy oil recovery, deformation is required for trapped oil to move through the pore throat. The deformation was aided by the reduction of the O/W IFT. The degree of residual oil mobilization in porous media is determined by the mobility ratio (Eq. (14)). Fig. 16a shows that flooding with water could not mobilize the trapped oil and the mobility ratio (M) is greater than 1. This means that the mobility of water is higher than that of the heavy oil (M > 1), an adverse event known as viscous fingering (Fig. 16c). This can be attributed to the high viscosity of the heavy oil hence a mobility ratio of less than 1 cannot be achieved resulting in early breakthrough and poor recovery during water flooding (Fig. 16c). This poor recovery might be due to precipitation and deposition of the heavy oil molecules (Fig. 16c). The precipitation and deposition of heavy oil molecules might be due to the changes in crude oil pressure during the waterflooding (Yakasai et al., 2021). However, Fig. 16b shows that the EFBSNF-assisted ultrasound had a piston like dispalcement shape indicating that the M $\le$ 1. This confirms the ability of EFBSNF flooding assisted ultrasound to simultaneously increase the viscosity of the injection fluid and also minimise the heavy oil viscosity. However, the reduction is higher with ultrasound due to the stronger effect of aquathermolysis catalysed in the presence of EFBSNF (Fig. 16d), resulting in uniform, efficient and increased oil recovery (Farooqui et al., 2015). During the EFBSNF flooding, the EFBSNF acted as inhibitor to prevent the heavy oil from precipitation by adsorbing the molecules. The breakdown of these heavy oil molecules resulted in the reduction in viscosity of the heavy oil thereby preventing aggregation, precipitation and deposition of the heavy oil molecules on the rock surface. In combination with ultrasound the reaction of the EFBSNF increased by creating more active surface for contact with the heavy oil molecules, which is consistent with viscosity results (Fig. 10).

(14)

$$\mathit{Mobility}ratio\left(M\right)=\frac{K_{\mathit{rd}}}{K_{\mathit{ro}}}\times \frac{{\mu }_o}{{\mu }_d}$$ whereas $K_{\mathit{rd}}$ is relative permeability of displacing fluid, $K_{\mathit{ro}}$ is relative permeability of displaced fluid, ${\mu }_o$ and ${\mu }_d$ are viscosity of displaced fluid and displacing fluid, respectively.

Close

Fig. 16

Schematic illustration of heavy oil recovery process using EFBSNF and Ultrasound (a) viscous fingering during water flooding (b) piston like font during EFBSNF flooding assisted with ultrasound (c) precipitation and deposition of heavy oil molecules during water flooding resulting in ununiform sweep efficiency (d) EFBSNP and ultrasound breakdown large molecules of heavy oil and reduced viscosity of heavy oil and (e) heavy oil molecules deposition.

Export to PPT

3.14 Economic feasibility

The energy and cost of the process was estimated using Eqs. (15) and (16).