- Email: [email protected]
Effects of sonication radiation on oil recovery by ultrasonic waves stimulated water-flooding
Ultrasonics 53 (2013) 607–614
Contents lists available at SciVerse ScienceDirect
Ultrasonics journal homepage: www.elsevier.com/locate/ultras
Effects of sonication radiation on oil recovery by ultrasonic waves stimulated water-flooding Erfan Mohammadian, Radzuan Junin ⇑, Omeid Rahmani, Ahmad Kamal Idris Department of Petroleum Engineering, Faculty of Petroleum and Renewable Energy Engineering, Universiti Teknologi Malaysia, 81310 UTM, Johor, Malaysia
a r t i c l e
i n f o
Article history: Received 15 April 2012 Received in revised form 10 October 2012 Accepted 11 October 2012 Available online 24 October 2012 Keywords: Ultrasonic waves Enhanced oil recovery Water-flooding Unconsolidated sand pack Emulsification
a b s t r a c t Due to partial understanding of mechanisms involved in application of ultrasonic waves as enhanced oil recovery method, series of straight (normal), and ultrasonic stimulated water-flooding experiments were conducted on a long unconsolidated sand pack using ultrasonic transducers. Kerosene, vaseline, and SAE10 (engine oil) were used as non-wet phase in the system. In addition, a series of fluid flow and temperature rise experiments were conducted using ultrasonic bath in order to enhance the understanding about contributing mechanisms. 3–16% increase in the recovery of water-flooding was observed. Emulsification, viscosity reduction, and cavitation were identified as contributing mechanisms. The findings of this study are expected to increase the insight to involving mechanisms which lead to improving the recovery of oil as a result of application of ultrasound waves. Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction The interest in recovery of oil from oil reservoirs using seismic waves as an enhanced oil recovery method dates back to 1950s when increased oil production was observed because of cultural noise and earthquakes. Due to limited distance ultrasonic waves can travel in the reservoir; most of the field applications were limited to damage removal in near wellbore area. Application of ultrasonic waves on different processes such as gravity drainage, imbibition, and water-flooding has been investigated by several authors [1–7]. Despite the number of publications, patents, and some field trials on the subject, the exact mechanisms of ultrasonic waves are not fully comprehended. In an early work, Albert and Bodine [8] invented a system for increasing the recovery by application of sonic waves. Duhon and Campbell [1] conducted a comprehensive research on long and short cores to find out about the possible utilization of ultrasonic waves in water-flooding. They observed a considerable effect on displacement efficiency resulted from creating a more uniform displacement as a result of sonication. Simkin [9] observed a major growth of oil droplets immediately after the beginning of excitation in a laboratory experiment. He attributed the growth to the sonically induced coalescence. Poesio et al. [10] investigated the influence of acoustic waves on liquid flow through the Berea Sandstone and found out that pressure gradient inside the cores decreases under acoustic energy. This effect was attributed to the ⇑ Corresponding author. Tel.: +60 75535514; fax: +60 75581463. E-mail address: [email protected] (R. Junin). 0041-624X/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ultras.2012.10.006
reduction of fluid viscosity. Amro et al. [4] attributed the increase in the recovery of ultrasonic stimulated water-flooding to the changes of relative permeability of both phases. Guo [11] discussed field application of ultrasonic waves in China. They also conducted experiments to show the effects of ultrasonic waves on viscosity and realized that the viscosity is temporarily reduced due to exposure to ultrasound waves. Najafi [5] analytically and experimentally investigated the effect of ultrasound on gravity drainage and percolation of oil by using fluids of different viscosities. 20 °C rises in the temperature was observed during his experiments in a period of 1000 min. A theory about the generation of ultrasonic waves was developed by Nikolaevsky and Stepanova [12]. He postulated that as a result of non-linear effects associated with seismic and low frequency acoustic waves in porous media saturated with fluid, under conditions of long–short-wave resonance, the non-linear generation of high ultrasonic frequencies by seismic waves is possible. Based on this theory, ultrasonic energy (high frequency waves) could be the main reason for enhancement of oil recovery after artificial or natural seismic activities. Therefore, understanding the effects of high frequency waves on recovery of oil involves a great importance. One way to do so is direct application of high frequency waves to sand pack model and study the results. In this study, a series of straight (normal) and ultrasonic stimulated water-flooding experiments (also known as sonicated waterflooding) were conducted on a long unconsolidated sand pack using ultrasonic transducers. This research concentrates on the involving mechanisms which lead to increase in the recovery of water-flooding stimulated by ultrasonic waves. Therefore, dynamic
608
E. Mohammadian et al. / Ultrasonics 53 (2013) 607–614
experiments were conducted to see the effects of ultrasonic waves on the recovery of oil; furthermore, some supplementary tests (e.g. fluid flow and temperature rise experiments) were carried out to investigate the mechanisms of ultrasonic waves in more details. 1.1. General description of mechanisms and laboratory studies Based on Simkin [13] and Odeh [14], two main mechanisms affect the movement of fluids in the reservoir; including gravitational and capillary forces. The gravitational forces behave on the difference in density between phases saturating the medium [15]. According to Beresnev and Johnson [15], the capillary forces play an important role in liquid infiltration via fine pore channels. During percolation process, liquid films are adsorbed on the pore walls which reduce the percolation via reducing the effective diameter of pore throats [15]. Kuznetsov and Simkin [16] calculated the mean thickness of the surface film of water in the pore throat. The salt concentration inversely affects the average thickness of the surface film which ranges from 5 to 50 microns with concentration NaCl solution 100 to 1 g/l, respectively. Duhon [17] carried out a self-contained laboratory study utilizing the influence of ultrasonic waves on liquids via porous media. In the presence of ultrasound, the oil recovery increased from sandstone as a result of ultrasonic exceeding. Based on Duhon’s studies, the attribute of permeability was reduced while the sound was applied. On the other hand, the amount of recovered oil increased with decreasing oil viscosity [15]. Duhon’s experiment revealed that the percolation rates increased in presence of ultrasound. This change took place at all temperatures; however, the percolation rate greatly increased at temperatures between 100 and 110 °C. 1.2. Wave propagation and mechanical interaction within granular materials Mechanical interaction of particles within granular material plays an important role for wave propagation. Hertz [18] studied the forces and deformations that occur at a contact point of two ideal spheres
Fn ¼
2 E R1=2 U n3=2 3 1 v2
ð1Þ
In which: Fn is contact force; E is Young’s modulus; V is Poisson’s ratio; R⁄ is effective radius; 1/R⁄1/2 = (1/R1 + 1/R2) with R1 and R2 the radii of the contacting particles and Un is normal displacement at the contact point. Hertz’s equation suggests a similitude of curved contacts with a non-linear spring. However, its non-linearity has a number of implications for the mechanical behavior of granular materials [19]. Hertz’s contact model was extended by Mindlin and Deresiewicz [20], Hardy et al. [21], Hutchings [22], Johnson [23], and Kogut and Komvopoulos [24]. The dynamical behavior of granular materials is affected by non-elastic, dissipative forms of granular interactions [24]. According to Michlmayr et al. [19], incomplete restitution of elastic energy during collisions and friction are two forms of energy dissipation that decrease kinetic energy of particles in motion. Levine [25] determined jamming or unjamming of particles within granular matter, i.e. the spontaneous onset or loss of rigidity; based on granular temperature, density, and stress. It suggests that yielding of granular materials under shear stress may be viewed as continuous fluctuations around a jamming–unjamming limit. This limitation is locally associated with increase in granular temperature [19].
The propagation of elastic waves in granular materials is influenced primarily by the grain elastic modulus, porosity, and energy dissipation mechanisms [19]. Based on Somfai et al. [26], the elastic waves propagate relatively uniform through a granular assembly including non-homogenous or a much higher degree of homogeneity network. George et al. [27], Lu and Sabatier [28], and Griffiths et al. [29] have shown that the presence of water, filling pore spaces between grains, affects the properties of wave propagation of the medium. On the other hand, Michlmayr et al. [19] showed that capillary forces provide additional contact stiffness to better wave propagation. However, the motion of contact lines and moderated viscous can increase the propagation of elastic waves. 2. Experimental set-up and procedure 2.1. Equipment Two types of ultrasonic transducers were employed for the experiments. For dynamic experiments, the ultrasonic transducers were specially installed surrounding the column test section. Crest ultrasonic generator with frequency of 40 kHz and power outputs of 100–500 W was used. Fig. 1 illustrates a schematic of the water-flooding (displacement) experiments. A centrifugal pump was utilized for the injection of the fluids. The rate of the injection was fixed on 3 ml/min in all flooding tests. Meanwhile for batch experiments, Branson ultrasonic bath with the frequency of 40 kHz and the power output of 110 W was used. 2.2. Fluid properties Two types of brine were used for the experiments: normal and de-aerated 3% NaCl brine, with the density of 1.05 g/cm3, as displacement fluids. Kerosene, vaseline, and SAE-10 (engine oil) were used as non-wet phase fluid system in the experiments. Table 1 summarizes the properties of fluids which are used. The viscosity of the fluids was measured using Cannon–Fenske Routine Viscometer-100 at 25 and 40 °C. Fluids were chosen to cover a broad range of viscosity and interfacial tension (IFT). SAE-10 was used as a heavier oil compared to vaseline (medium viscosity, high IFT oil) and kerosene (very low viscosity, and intermediate IFT to water). Kerosene and SAE-10 are found to behave as Newtonian fluid. Meanwhile, vaseline exhibits a pronounced non-Newtonian shear–thinning flow behavior which is well described by a power–law equation. 2.3. Porous media The quartz sand grains of nominal 106–225 lm particle size were packed in a stainless steel sample holder of 102 cm in length and 3.8 cm in diameter to represents porous media in all of the dynamic (displacement) experiments. Before used, the grains were carefully washed and dried in the oven. To ensure homogenous packing of sands, the holder was placed on an electrical shaker working at constant rpm of 150. 5 cm3 of sand grains were poured in the holder at each stage and was shaken well. The morphology of grains used ranged from sub-angular to sub-rounded. Rounded grains tend to have greater porosity and thus greater permeability (e.g. 4000 mD in this study) than uniformly graded sediments because there are fewer finer particles to fill pore spaces between larger particles. All of the grains are described so far granular and the particles do not adhere to each other. The schematic of the sand pack holder is shown in Fig. 2. The porosity and permeability of the sand pack were measured as 32 ± 2% and 4000 mD, respectively.
E. Mohammadian et al. / Ultrasonics 53 (2013) 607–614
609
Fig. 1. Laboratory setup of the water-flooding experiments.
Porosity is a ratio of pore volume to bulk volume in a porous media. Total porosity of the sand pack was measured using weighting method. For measuring porosity X, the volume of core holder were calculated by measuring the whole sides (Vb). Then the sand pack without core holder was weighted to determine the volume of grain (Vg). The pore volume (VP) was calculated by subtracting the Vg from Vb. The following equation (Eq. (2)) was used for measuring porosity:
/ ¼ ðV P =V b Þ 100
ð2Þ
Permeability represents the ability of the porous medium to permit the flow of fluids through it. The permeability of sand pack was determined by saturating the core with brine at various flow rates and reading the respective pressure drops. The value of permeability is calculated by applying the Darcy flow equation:
K ¼ ðqlLÞ=ðA DPÞ
ð3Þ
Where: k is rock permeability (Darcy); q is brine flow rate (ml/s); l is brine viscosity (cp); L is length of glass-bead model (cm); A is inner cross section of the glass-bead model (cm) and DP is pressure differential. To compare the reported permeability with typical values of sandstone samples from oil well, the permeability of the sand pack was 4000 mD. However, the permeability of petroleum reservoir rocks may typically range from 0.1 to 1000 or more mD. Meanwhile, the typical porosities of petroleum reservoir rock ranges from 5% to 30%, but most frequently are between 10% and 20%. 2.4. Displacement test Oil saturated sand pack was water-flooded until the residual oil saturation was obtained. Ultrasonic radiation (40 kHz and 250 W) began at this point simultaneous with water flooding. The same
volumes of water (as in water-flooding) were injected and the recovery was calculated for each case. The graphs of recovery versus time were plotted for straight and ultrasonic stimulated waterflooding. 2.5. Fluid flow and temperature rise experiments Two types of supplementary experiments were conducted, namely: one phase flow experiment and temperature rise experiments. One phase flow experiments were conducted using the same setup of displacement tests. In this experiment, the core was saturated with brine (normal and de-aerated) then was exposed to ultrasonic waves; the pressure changes for the system were recorded. Temperature rise experiments were conducted using both ultrasonic column and ultrasonic bath. Using ultrasonic column, the saturated sample (with oil and brine) was exposed to ultrasonic waves of different power outputs and temperature rises for the system was measured via a thermometer installed in the sand pack. Meanwhile, for the temperature rises experiments conducted in an ultrasonic bath, the sand was packed and saturated with kerosene, vaseline, and SAE-10. The temperature rises of the system were recorded regularly. 3. Results and discussion 3.1. Thermal conductivity of fluids The effective thermal conductivity (ETC) of porous media is one of the prominent properties for thermal design and numerical simulations [30]. The study of the ETC in fluid saturated porous media can be used in petroleum science and engineering. Temperature, pressure, and grain morphology are the most important factors affected the ETC [30].
610
E. Mohammadian et al. / Ultrasonics 53 (2013) 607–614
Table 1 Testing fluid properties used in the displacement experiments.
a b c
Name of fluid
Densitya @ 27 °C (g/cm3)
Viscosityb @ 25 °C (cp)
Viscosityb @ 40 °C (cp)
IFTc with brine @ 27 °C (dyne/cm)
IFTc with de-aerated brine @ 27 °C (dyne/cm)
Thermal conductivity 20 °C (Watt/m °C)
Kerosense Vaseline SAE-10 Normal brine De-aerated brine Water 3.5% saline Quartz grains Stainless steel
0.770 ± 0.0001 0.856 ± 0.0001 0.923 ± 0.0001 1.03 1.04 1 2.65 –
0.99 ± 0.01 22.0 240 0.92 0.93 – – –
0.9 ± 0.01 12 65 0.65 0.65 – – –
31 38 18 – – – – –
33 35 20 – – – – –
0.145 0.140 0.146 – – 0.058 [31]–0.6 [32] 6.5–11.3 [33] 14–14.3 [34]
Hydrometer method (ASTM D1298). Cannon–Fenske Routine viscometer (kinematic viscosity, ASTM D445). Du Nouy type tensiometer (Kruss Tensiometer Platinum–Iridium ring, ASTM D1331).
According to Abdulagatova [30], the morphology complexity of the rock structures is the main difficulty to predict the ETC in the porous media; however,the morphology of the studied samples ranges from sub-angular to sub-rounded in shape. The ETC values vary with the temperature (0–250 °C) whereas up to 16% at low pressure (0.1 MPa) and up to 10% at high pressure (400 MPa). Based on the ETC values, the thermal expansion coefficients can be calculated [30]. Considering the ambient conditions, the values of thermal coefficient are negligible. The thermal coefficient at the pressure range of <100 MPa sharply decreases while at the high pressure very slightly decreases. At low pressure, the change of temperature is very slight for the thermal coefficient. However, at the high pressure, the temperature changing of the ETC is large. As shown in Table 1, the thermal conductivity for kerosene and vaseline is 0.145 and 0.140 (Watt/m 20 °C), respectively. However, this value for stainless steel and quartz grains is 6.15–11.3 and 14–14.3 (Watt/m 20 °C), respectively. Although in this research we did not correctly measure the thermal conductivity, we extracted the values of the material, or range of the values from previous studies [31–34].
3.2. Temperature rises experiments Temperature rises during stimulation with ultrasonic were reported by some authors [1,35,36]. However, the methods they used were sometimes too simplistic to show the real changes of temperature in the system; therefore, the magnitude of the effect is not sufficiently discussed. The temperature rises of the system (normal brine–saturated sand pack) were 4, 12, and 16 °C for the respective power outputs of 100, 250, and 400 W (Fig. 3). In the second part of the experiment, temperature rises for vaseline, kerosene, and SAE-10 were measured in the ultrasonic bath. Due to the same conductivity coefficient (i.e. 0.145, 0.146, and 0.140 for kerosene, SAE-10, and vaseline, respectively) in three of the cases the temperature rise was almost the same as expectation. Fig. 4 presents the temperature rise for different fluids. The temperature rises can affect fluid properties i.e. viscosity and IFT. IFT between oil and brine is a function of temperature. At elevated temperature, IFT is lower as it can be inferred from Firoozabadi and Ramey’s equation (Eq. (4)). Firoozabadi and Ramey [37] correlated surface tension using the difference between pure water and hydrocarbon density, temperature, and critical temperature. The correlation involved measured data from methane, propane, and benzene–water system. According to Firoozabadi and Ramey’s equation, the interfacial tension (r) between water–hydrocarbon (gas or oil) phases can be calculated.
r1=4 ¼
a1 Dqb1 T 0:3125 r
Dq ¼ qx qhc
Fig. 2. The schematic of the sand pack holder.
ð4Þ ð5Þ
Where: qx is the density of water (g/cm3); qhc is the density of hydrocarbon (g/cm3); a1, b1 is the constant (dyne/cm or mN/m) and Tr is the critical temperature (°C). Table 2 shows IFT reduction of different fluids versus time calculated by Firoozabadi and Ramey’s equation. Although the IFT was decreased in all the cases, the reduction is not high enough to have significant effects on the capillary number; therefore, it is not able to contribute in reduction of residual oil saturation. It should be noted that mentioned discussion is only related to IFT reduction as a consequence of temperature rises. However, the direct effect of ultrasonic waves on reduction of IFT is excluded in this section. Hence, it can be concluded that, if the waves have any effect on reduction of IFT, it cannot be related to temperature rises of the system. Viscosity of fluids is a strong function of temperature. Temperature rises as much as 24 °C will largely affect viscosity of fluids. The viscosity of water and oil will be reduced as a result of temperature rise. Since the viscosity of fluid at different temperatures was needed it was calculated via proper formulas i.e. Meehan and
E. Mohammadian et al. / Ultrasonics 53 (2013) 607–614
611
Fig. 3. Temperature of water versus time at various power outputs. Fig. 5. Reduction in viscosity of water.
Fig. 6. Reduction in viscosity of kerosene.
Fig. 4. Temperature rises for SAE-10, vaseline, and kerosene.
Table 2 IFT reduction for different fluids versus time. T (°C)
Kerosene (dyne/cm)
Vaseline (dyne/cm)
SAE-10 (dyne/cm)
23 27 31 35 39 43 47 51 55
31.0 30.5 30.0 29.5 29.0 28.6 28.1 27.7 27.3
38.0 37.4 36.8 36.2 35.6 35.0 34.5 33.9 33.4
18.0 17.7 17.4 17.1 16.8 16.6 16.3 16.1 15.8
Glaso’s relations. The measured values of viscosity (using Cannon–Fenske viscometer as shown in Table 1) have 5% variance from calculated values using Glaso’s formula. As the viscosity of water is reduced by temperature rises, it was calculated using Meehan’s [38] correlation. The reduction in viscosity is not the same for various fluids. Figs. 5–8 illustrate viscosity reduction for each case together with a third degree polynomial trend line drawn for each fluid to facilitate the comparison between slopes of the graphs. 3.3. One phase flow experiments Pressure reached a peak initially then decreased to lower values according to the intensity of waves (Fig. 9). One possible explanation is that cavitation is responsible for erratic behavior of pres-
Fig. 7. Reduction in viscosity of vaseline.
sure. As mentioned by Duhon and Campbell [1], during the implosion of bubbles, high pressure surges are created within the system. However, after all bubbles implode, the pressure drops. Therefore, cavitation could be accounted as one of the contributing mechanisms in application of ultrasonic waves. On the other hand, the decrease in the pressure drop could be attributed to increase in absolute permeability only if the viscosity of water remains the same. This is only possible if the temperature increase within the system could be considered negligible. In none of the experiments during the sonication, temperature remains constant. Since all the dynamic experiments were conducted on unconsolidated sand pack and only certain size of sand grains were
612
E. Mohammadian et al. / Ultrasonics 53 (2013) 607–614
Fig. 10. Pressure gradient versus time for various intensities of waves (equivalent to power outputs of 100, 200, 300, and 400 W) for one phase flow using deaerated brine. Fig. 8. Reduction in viscosity of SAE-10 (engine oil).
started to decline and finally stabilized to a value at the end of each experiment. Fig. 10 shows pressure response versus time for various power outputs. Considering the results from temperature effect experiment and one phase flow experiment, it is implied that, reduction in viscosity of water could be the main reason behind pressure drop observed in one phase flow experiment. This result is in agreement with that of Poesio et al. [10]; they concluded that the reduction in pressure is due to decrease in the viscosity of fluids. However, they did not conduct any other experiment to confirm their theory. 3.4. Displacement experiments
Fig. 9. Pressure gradient versus time for various intensities of waves (equivalent to power outputs of 100, 200, 300, and 400 W) for one phase flow using normal brine.
used without any cementing materials, it can be assumed that the changes in absolute permeability did not occur. The reduction in pressure is caused by the reduction in viscosity of water. As an example, during sonication using 2.4 W/cm2 pressure increased to 2.3 psi from initial value of 2.1 psi; at the end of the experiment pressure was constant at 1.8 psi which is 0.1 psi less than its initial value. It could be attributed to the reduction in the viscosity of water
DP ¼
Q lL KA
ð6Þ
Terms Q, L, A, K, and l are considered constant. As there is only one phase flowing in the system, the term l stands for viscosity of water (lw). The increase in temperature leads to viscosity reduction; since DP is proportional to lw, any reduction in the viscosity decreases the pressure drops by the same magnitude. Due to existence of cavitation as a result of using aerated water, quantification of the effect becomes difficult. But as through the whole one phase flow experiment the pressure reaches to a value lower than its initial value, it could be concluded that the reduction of pressure is due to reduction in viscosity of water [39]. This result is in agreement with that of Poesio et al. [10]. The second series of one phase flow experiments were conducted using de-aerated brine to remove effects of cavitation. The same procedure was performed and the pressure responses were recorded. Once the ultrasonic radiation started, the pressure
Recovery for water-flooding was 67%; sonication improved the recovery by 16%. Assuming relative permeability of non-wet and wetting phase at their end points, the mobility ratio for kerosene and water may be calculated at ambient temperature (27 °C) as follows:
M¼
krw lo 0:3 0:98 ¼ 0:44 ¼ kro lw 0:7 0:95
ð7Þ
The above mobility ratio is only valid for the ultrasonic stimulated water-flooding if viscosity remains the same throughout the experiment. In this research the point of interest is the viscosity ratio fluids and not the relative permeability of fluids. During ultrasonic stimulation, temperature rises were observed for the system. 10 °C increase in the temperature reduces viscosity of water to 0.79 cp (17% decreases) and viscosity of kerosene to 0.79 cp (20% decrease). Therefore, the viscosity ratio of the system becomes 1. Having IFT of 32 dyne/cm for brine and kerosene, the capillary number for the water is 1.58 107. Capillary numbers in orders of 105–104 are needed to decrease residual oil saturations. Another interesting result was formation of emulsion. After three hours the two phases were completely separated and the interface between the phases could be easily seen. Fig. 11 shows total recovery of kerosene as a result of straight and ultrasonic stimulated water-flooding. In the case of vaseline, the viscosity ratio at 27 °C is 23.9. The recovery for normal water-flooding is 56% ultrasonic stimulated water-flooding added 11% to the recovery. Fig. 12 shows total recovery of vaseline. Viscosity reduction is known as one of the significant mechanisms in case of vaseline as viscosity ratio reduces to 13.9 from initial value of 23.9 as the result of 10 °C temperature rise. Therefore, it can be concluded that ultrasonic stimulation increases the recovery via reducing mobility ratio and improving sweep efficiency. In the case of vaseline also, formation of emulsion
E. Mohammadian et al. / Ultrasonics 53 (2013) 607–614
613
4. Conclusions Series of displacement and supplementary experiments were conducted in this study and the following conclusions were made:
Fig. 11. Recovery of kerosene as result of straight and sonicated water-flooding.
Fig. 12. Recovery of vaseline as result of straight and sonicated water-flooding.
The recovery of water-flooding increases as a result of sonication for all the cases (from 3% to 16%). The recovery of ultrasonic assisted water-flooding was higher for less viscous fluid being kerosene. Where normal brine was used, the recovery was higher in comparison with the cases where de-aerated brine was used. This could be explained through the existence of cavitation in the system when using normal brine instead of de-aerated brine. Severe temperature rises were observed in the experiments. This leads to reduction in viscosity of fluids as well as reduction in the interfacial tension. The reduction in viscosity of fluid is detected as one of the contributing mechanisms of production. On the other hand, the temperature rises are not high enough to reduce the IFT to a large extent, in other word IFT reduction from temperature rises are so small that cannot contribute in improving the recovery. The one phase flow experiments were conducted using brine. The initial increase in the pressure in one phase flow experiments (using normal brine) can be attributed to cavitation. Conducting one phase flow experiments (using normal and de-aerated brine) proved that the reduction in pressure is caused by reduction in viscosity of water as results of ultrasonic stimulation. Formation of emulsion was observed during displacement experiments for vaseline and kerosene. The generated emulsion was unstable and the phases were completely separated after 3 h. It could be concluded that cavitation, viscosity reduction and emulsification are contributing mechanisms in application of ultrasonic waves to water-flooding.
References
Fig. 13. Recovery of SAE-10 by straight and sonicated water-flooding.
was observed. Assuming the IFT value of 38 dyne/cm for vaseline and brine, the capillary number for water-flooding is in orders of 107. It can be concluded that increase in the oil recovery cannot be attributed to reduction of IFT from temperature rise. SAE-10 is a viscous fluid (243 cp at 23 °C). The amount of oil produced by water-flooding was 12%. The sonication added merely 3% to the recovery (Fig. 13). 10 °C increase in the temperature (around 23 °C increase in the temperature was observed by temperature experiment), is changing viscosity of fluids to a large extent. Viscosity ratio reduces to 97.1 from initial value of 255.0. Therefore, one of the effective mechanisms in improving the recovery is viscosity reduction. One may expect higher recoveries considering huge reduction in viscosity ratio. But an unexpected result was deposition of paraffin as results of sonication. It was observed during emulsification experiment which was a part of the author’s previous work [40]. In this case also the IFT reduction from temperature increase is not large enough to reduce capillary number effectively and reduce the residual oil saturation.
[1] R.D. Duhon, J. Campbell, The effect of ultrasonic energy on the flow of fluids in porous media, in: SPE 1316, The 2nd Annual Eastern Regional Meeting of SPE/ AIM, Charlston, USA, 1965. [2] K. Naderi, T. Babadagli, Effect of ultrasonic intensity and frequency on oil/ heavy–oil recovery from different wettability rocks, SPE International Thermal Operations and Heavy Oil Symposium, Calgary, Canada, 2008. [3] T. Hamida, T. Babadagli, Fluid–fluid interaction during miscible and immiscible displacement under ultrasonic waves, European Physical Journal 60 (2007) 447–462. [4] M. Amro, M. Al-Mobarky, Al-Homadhi, Improved oil recovery by application of ultrasound waves to water-flooding, in: SPE 105370, SPE Middle East Oil & Gas Show, Bahrain, 2007. [5] I. Najafi, A mathematical analysis of the mechanism of ultrasonic induced fluid percolation in porous media, in: SPE 141126, SPE Annual Technical Conference And Exhibition, Florance, Italy, 2010. [6] T. Hamida, Effect of Ultrasonic Waves on Immiscible and Miscible Displacement in Porous Media, Master thesis, University of Alberta, Canada, 2006. [7] T. Hamida, T. Babadagli, Effects of ultrasonic waves on the interfacial forces between oil and water, Ultrasonics Sonochemistry 15 (2008) 274–278. [8] G. Albert, J.V. Bodine, No. Paten 2,667,962, US Patent, California, 1948. [9] E. Simkin, Advanced vibroseismic technique for water-flooded reservoir stimulation, mechanims and field tests results, in: The 6th European IOR Symposium, Stavenger, Norway, 1991. [10] P. Poesio, G. Ooms, A. Schraven, F.V.D. Bas, Theoretical and experimental investigation of acoustic streaming in a porous material, Physical Review E (66) (2002). [11] X. Guo, High Frequency Vibration Recovery Enhancement Technology in the Heavy oIl Fields in China, in: SPE 89656, SPE International Thermal Operations and Heavy Oil Symposium and Western Regional Meeting, California, USA, 2009. [12] V.N. Nikolaevsky, G.S. Stepanova, Non-linear seismics and the acoustic action on the oil recovery from oil pool, Acoustical Physics 51 (2005) 131–139. [13] E.M. Simkin, Oil will return in three months, Energy (Energiya) 3 (1985) 44–47.
614
E. Mohammadian et al. / Ultrasonics 53 (2013) 607–614
[14] A.S. Odeh, Mathematical modeling of the behavior of hydrocarbon reservoirs– the present and the future, in: J. Bear, M.Y. Corapcioglu (Eds.), Advances in Transport Phenomena in Porous Media, Martinus Nijhoff Publ, 1987, pp. 821– 848. [15] I.A. Beresnevand, P.A. Johnson, Elastic–wave stimulation of oil production: a review of methods and results, Geophysics 59 (6) (1994) 1000–1017. [16] O.L. Kuznetsov, E.M. Simkin, Transformation and Interaction of Geophysical Fields in Lithosphere, Nedra Press, 1990. 106–112. [17] R.D. Duhon, An Investigation of the Effect of Ultrasonic Energy on the Flow of Fluids in Porous Media, Ph.D thesis, Univ. of Oklahoma, 1964. [18] H. Hertz, Über die Berührung fester elastischer Körper (On the contact of elastic solids), Journal für die Reine und Angewandte Mathematik 92 (1882) 156–171. For English translation see miscellaneous papers by H. Hertz, (Ed.), Jones and Schott, London, 1896.. [19] G. Michlmayr, D. Cohen, D. Or, Sources and characteristics of acoustic emissions from mechanically stressed geologic granular media – A review, Earth–Science Reviews 112 (2012) 97–114. [20] R.D. Mindlin, H. Deresiewicz, Elastic spheres in contact under varying oblique forces, Journal of Applied Mechanics 75 (1953) 327–344. [21] C. Hardy, C.N. Baronet, G.V. Tordion, The elasto-plastic indentation of a halfspace by a rigid sphere, International Journal for Numerical Methods in Engineering 3 (1971) 451–462. [22] I.M. Hutchings, Energy absorbed by elastic waves during plastic impact, Journal of Physics D: Applied Physics (12) (1979). [23] K.L. Johnson, Contact Mechanics, Cambridge University Press, Cambridge, 1985. [24] L. Kogut, K. Komvopoulos, Analysis of the spherical indentation cycle for elastic perfectly plastic solids, Journal of Materials Research 19 (2004) 3641–3653. [25] D. Levine, Jamming and the statistics of granular materials, in: A. Liu, S. Nagel (Eds.), Jamming and Rheology, Taylor & Francis, London and New York, 2001, pp. 9–19. [26] E. Somfai, J.N. Roux, J.H. Snoeijer, M.V. Hecke, W.V. Saarloos, Elastic wave propagation in confined granular systems, Physical Review E (72) (2005).
[27] L.A. George, M.M. Dewoolkar, D. Znidarcic, Simultaneous laboratory measurement of acoustic and hydraulic properties of unsaturated soils, Vadose Zone Journal 8 (2009) 633–642. [28] Z. Lu, J.M. Sabatier, Effects of soil water potential and moisture content on sound speed, Soil Science Society of America Journal 73 (2009) 1614–1625. [29] S. Griffiths, A. Rescaglio, F. Melo, Ultrasound propagation in wet and airless non-consolidated granular materials, Ultrasonics 50 (2010) 139–144. [30] Z. Abdulagatova, I.M. Abdulagatov, V.N. Emirov, Effect of temperature and pressure on the thermal conductivity of sandstone, International Journal of Rock Mechanics & Mining Sciences 46 (2009) 1055–1071. [31] R.M. Digullio, A.S. Teja, Thermal conductivity of aqueous salt solutions at high temperatures and high concentrations, American Chemical Society, 1992. 1081–1085. [32] H.P. Young, D. Hugh, University Physics, seventh ed., Addison Wesley, 1992. [33] K.M. Smits, T. Sakaki, A. Limsuwat, T.H. Illangasekare, Determination of the thermal conductivity of sands under varying Moisture drainage/wetting and porosity conditions–applications in near–surface soil moisture distribution analysis, Hydrology Days (2009). [34] J.G. Hust, P. Giarratano, Thermal conductivity and electrical resistivity, Standard Reference Materials (1975). [35] W.I. Chen, Influence of Ultrasonic Energy Upon the Rate of Flow of Liquids Through Porous Media, PhD Thesis, West Virginia University, 1969. [36] H.V. Fairbanks, W.I. Chen, Ultrasonic acceleration of liquid flow through porous media, Chem. Eng. Symp. vol. 67, 1971 pp. 108 . [37] A. Firoozabadi, H.J. Ramey, Surface tension of water hydrocarbon systems at reservoir conditions, Journal of Canada Petrol Technology 27 (3) (1988) 41–48. [38] D.N. Meehan, A correlation for water compressibility, Petroleum Engineer (1980) 125–126. [39] E. Mohammadian, Ultrasonic Assisted Water-Flooding, M.Sc. thesis, Universiti Teknologi Malaysia, 2010. [40] A. Danesh, PVT and Phase Behaviour of Petroleum Fluids, Herriot–Watt University, Edinburgh, Scotland, 1998.
Contents lists available at SciVerse ScienceDirect
Ultrasonics journal homepage: www.elsevier.com/locate/ultras
Effects of sonication radiation on oil recovery by ultrasonic waves stimulated water-flooding Erfan Mohammadian, Radzuan Junin ⇑, Omeid Rahmani, Ahmad Kamal Idris Department of Petroleum Engineering, Faculty of Petroleum and Renewable Energy Engineering, Universiti Teknologi Malaysia, 81310 UTM, Johor, Malaysia
a r t i c l e
i n f o
Article history: Received 15 April 2012 Received in revised form 10 October 2012 Accepted 11 October 2012 Available online 24 October 2012 Keywords: Ultrasonic waves Enhanced oil recovery Water-flooding Unconsolidated sand pack Emulsification
a b s t r a c t Due to partial understanding of mechanisms involved in application of ultrasonic waves as enhanced oil recovery method, series of straight (normal), and ultrasonic stimulated water-flooding experiments were conducted on a long unconsolidated sand pack using ultrasonic transducers. Kerosene, vaseline, and SAE10 (engine oil) were used as non-wet phase in the system. In addition, a series of fluid flow and temperature rise experiments were conducted using ultrasonic bath in order to enhance the understanding about contributing mechanisms. 3–16% increase in the recovery of water-flooding was observed. Emulsification, viscosity reduction, and cavitation were identified as contributing mechanisms. The findings of this study are expected to increase the insight to involving mechanisms which lead to improving the recovery of oil as a result of application of ultrasound waves. Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction The interest in recovery of oil from oil reservoirs using seismic waves as an enhanced oil recovery method dates back to 1950s when increased oil production was observed because of cultural noise and earthquakes. Due to limited distance ultrasonic waves can travel in the reservoir; most of the field applications were limited to damage removal in near wellbore area. Application of ultrasonic waves on different processes such as gravity drainage, imbibition, and water-flooding has been investigated by several authors [1–7]. Despite the number of publications, patents, and some field trials on the subject, the exact mechanisms of ultrasonic waves are not fully comprehended. In an early work, Albert and Bodine [8] invented a system for increasing the recovery by application of sonic waves. Duhon and Campbell [1] conducted a comprehensive research on long and short cores to find out about the possible utilization of ultrasonic waves in water-flooding. They observed a considerable effect on displacement efficiency resulted from creating a more uniform displacement as a result of sonication. Simkin [9] observed a major growth of oil droplets immediately after the beginning of excitation in a laboratory experiment. He attributed the growth to the sonically induced coalescence. Poesio et al. [10] investigated the influence of acoustic waves on liquid flow through the Berea Sandstone and found out that pressure gradient inside the cores decreases under acoustic energy. This effect was attributed to the ⇑ Corresponding author. Tel.: +60 75535514; fax: +60 75581463. E-mail address: [email protected] (R. Junin). 0041-624X/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ultras.2012.10.006
reduction of fluid viscosity. Amro et al. [4] attributed the increase in the recovery of ultrasonic stimulated water-flooding to the changes of relative permeability of both phases. Guo [11] discussed field application of ultrasonic waves in China. They also conducted experiments to show the effects of ultrasonic waves on viscosity and realized that the viscosity is temporarily reduced due to exposure to ultrasound waves. Najafi [5] analytically and experimentally investigated the effect of ultrasound on gravity drainage and percolation of oil by using fluids of different viscosities. 20 °C rises in the temperature was observed during his experiments in a period of 1000 min. A theory about the generation of ultrasonic waves was developed by Nikolaevsky and Stepanova [12]. He postulated that as a result of non-linear effects associated with seismic and low frequency acoustic waves in porous media saturated with fluid, under conditions of long–short-wave resonance, the non-linear generation of high ultrasonic frequencies by seismic waves is possible. Based on this theory, ultrasonic energy (high frequency waves) could be the main reason for enhancement of oil recovery after artificial or natural seismic activities. Therefore, understanding the effects of high frequency waves on recovery of oil involves a great importance. One way to do so is direct application of high frequency waves to sand pack model and study the results. In this study, a series of straight (normal) and ultrasonic stimulated water-flooding experiments (also known as sonicated waterflooding) were conducted on a long unconsolidated sand pack using ultrasonic transducers. This research concentrates on the involving mechanisms which lead to increase in the recovery of water-flooding stimulated by ultrasonic waves. Therefore, dynamic
608
E. Mohammadian et al. / Ultrasonics 53 (2013) 607–614
experiments were conducted to see the effects of ultrasonic waves on the recovery of oil; furthermore, some supplementary tests (e.g. fluid flow and temperature rise experiments) were carried out to investigate the mechanisms of ultrasonic waves in more details. 1.1. General description of mechanisms and laboratory studies Based on Simkin [13] and Odeh [14], two main mechanisms affect the movement of fluids in the reservoir; including gravitational and capillary forces. The gravitational forces behave on the difference in density between phases saturating the medium [15]. According to Beresnev and Johnson [15], the capillary forces play an important role in liquid infiltration via fine pore channels. During percolation process, liquid films are adsorbed on the pore walls which reduce the percolation via reducing the effective diameter of pore throats [15]. Kuznetsov and Simkin [16] calculated the mean thickness of the surface film of water in the pore throat. The salt concentration inversely affects the average thickness of the surface film which ranges from 5 to 50 microns with concentration NaCl solution 100 to 1 g/l, respectively. Duhon [17] carried out a self-contained laboratory study utilizing the influence of ultrasonic waves on liquids via porous media. In the presence of ultrasound, the oil recovery increased from sandstone as a result of ultrasonic exceeding. Based on Duhon’s studies, the attribute of permeability was reduced while the sound was applied. On the other hand, the amount of recovered oil increased with decreasing oil viscosity [15]. Duhon’s experiment revealed that the percolation rates increased in presence of ultrasound. This change took place at all temperatures; however, the percolation rate greatly increased at temperatures between 100 and 110 °C. 1.2. Wave propagation and mechanical interaction within granular materials Mechanical interaction of particles within granular material plays an important role for wave propagation. Hertz [18] studied the forces and deformations that occur at a contact point of two ideal spheres
Fn ¼
2 E R1=2 U n3=2 3 1 v2
ð1Þ
In which: Fn is contact force; E is Young’s modulus; V is Poisson’s ratio; R⁄ is effective radius; 1/R⁄1/2 = (1/R1 + 1/R2) with R1 and R2 the radii of the contacting particles and Un is normal displacement at the contact point. Hertz’s equation suggests a similitude of curved contacts with a non-linear spring. However, its non-linearity has a number of implications for the mechanical behavior of granular materials [19]. Hertz’s contact model was extended by Mindlin and Deresiewicz [20], Hardy et al. [21], Hutchings [22], Johnson [23], and Kogut and Komvopoulos [24]. The dynamical behavior of granular materials is affected by non-elastic, dissipative forms of granular interactions [24]. According to Michlmayr et al. [19], incomplete restitution of elastic energy during collisions and friction are two forms of energy dissipation that decrease kinetic energy of particles in motion. Levine [25] determined jamming or unjamming of particles within granular matter, i.e. the spontaneous onset or loss of rigidity; based on granular temperature, density, and stress. It suggests that yielding of granular materials under shear stress may be viewed as continuous fluctuations around a jamming–unjamming limit. This limitation is locally associated with increase in granular temperature [19].
The propagation of elastic waves in granular materials is influenced primarily by the grain elastic modulus, porosity, and energy dissipation mechanisms [19]. Based on Somfai et al. [26], the elastic waves propagate relatively uniform through a granular assembly including non-homogenous or a much higher degree of homogeneity network. George et al. [27], Lu and Sabatier [28], and Griffiths et al. [29] have shown that the presence of water, filling pore spaces between grains, affects the properties of wave propagation of the medium. On the other hand, Michlmayr et al. [19] showed that capillary forces provide additional contact stiffness to better wave propagation. However, the motion of contact lines and moderated viscous can increase the propagation of elastic waves. 2. Experimental set-up and procedure 2.1. Equipment Two types of ultrasonic transducers were employed for the experiments. For dynamic experiments, the ultrasonic transducers were specially installed surrounding the column test section. Crest ultrasonic generator with frequency of 40 kHz and power outputs of 100–500 W was used. Fig. 1 illustrates a schematic of the water-flooding (displacement) experiments. A centrifugal pump was utilized for the injection of the fluids. The rate of the injection was fixed on 3 ml/min in all flooding tests. Meanwhile for batch experiments, Branson ultrasonic bath with the frequency of 40 kHz and the power output of 110 W was used. 2.2. Fluid properties Two types of brine were used for the experiments: normal and de-aerated 3% NaCl brine, with the density of 1.05 g/cm3, as displacement fluids. Kerosene, vaseline, and SAE-10 (engine oil) were used as non-wet phase fluid system in the experiments. Table 1 summarizes the properties of fluids which are used. The viscosity of the fluids was measured using Cannon–Fenske Routine Viscometer-100 at 25 and 40 °C. Fluids were chosen to cover a broad range of viscosity and interfacial tension (IFT). SAE-10 was used as a heavier oil compared to vaseline (medium viscosity, high IFT oil) and kerosene (very low viscosity, and intermediate IFT to water). Kerosene and SAE-10 are found to behave as Newtonian fluid. Meanwhile, vaseline exhibits a pronounced non-Newtonian shear–thinning flow behavior which is well described by a power–law equation. 2.3. Porous media The quartz sand grains of nominal 106–225 lm particle size were packed in a stainless steel sample holder of 102 cm in length and 3.8 cm in diameter to represents porous media in all of the dynamic (displacement) experiments. Before used, the grains were carefully washed and dried in the oven. To ensure homogenous packing of sands, the holder was placed on an electrical shaker working at constant rpm of 150. 5 cm3 of sand grains were poured in the holder at each stage and was shaken well. The morphology of grains used ranged from sub-angular to sub-rounded. Rounded grains tend to have greater porosity and thus greater permeability (e.g. 4000 mD in this study) than uniformly graded sediments because there are fewer finer particles to fill pore spaces between larger particles. All of the grains are described so far granular and the particles do not adhere to each other. The schematic of the sand pack holder is shown in Fig. 2. The porosity and permeability of the sand pack were measured as 32 ± 2% and 4000 mD, respectively.
E. Mohammadian et al. / Ultrasonics 53 (2013) 607–614
609
Fig. 1. Laboratory setup of the water-flooding experiments.
Porosity is a ratio of pore volume to bulk volume in a porous media. Total porosity of the sand pack was measured using weighting method. For measuring porosity X, the volume of core holder were calculated by measuring the whole sides (Vb). Then the sand pack without core holder was weighted to determine the volume of grain (Vg). The pore volume (VP) was calculated by subtracting the Vg from Vb. The following equation (Eq. (2)) was used for measuring porosity:
/ ¼ ðV P =V b Þ 100
ð2Þ
Permeability represents the ability of the porous medium to permit the flow of fluids through it. The permeability of sand pack was determined by saturating the core with brine at various flow rates and reading the respective pressure drops. The value of permeability is calculated by applying the Darcy flow equation:
K ¼ ðqlLÞ=ðA DPÞ
ð3Þ
Where: k is rock permeability (Darcy); q is brine flow rate (ml/s); l is brine viscosity (cp); L is length of glass-bead model (cm); A is inner cross section of the glass-bead model (cm) and DP is pressure differential. To compare the reported permeability with typical values of sandstone samples from oil well, the permeability of the sand pack was 4000 mD. However, the permeability of petroleum reservoir rocks may typically range from 0.1 to 1000 or more mD. Meanwhile, the typical porosities of petroleum reservoir rock ranges from 5% to 30%, but most frequently are between 10% and 20%. 2.4. Displacement test Oil saturated sand pack was water-flooded until the residual oil saturation was obtained. Ultrasonic radiation (40 kHz and 250 W) began at this point simultaneous with water flooding. The same
volumes of water (as in water-flooding) were injected and the recovery was calculated for each case. The graphs of recovery versus time were plotted for straight and ultrasonic stimulated waterflooding. 2.5. Fluid flow and temperature rise experiments Two types of supplementary experiments were conducted, namely: one phase flow experiment and temperature rise experiments. One phase flow experiments were conducted using the same setup of displacement tests. In this experiment, the core was saturated with brine (normal and de-aerated) then was exposed to ultrasonic waves; the pressure changes for the system were recorded. Temperature rise experiments were conducted using both ultrasonic column and ultrasonic bath. Using ultrasonic column, the saturated sample (with oil and brine) was exposed to ultrasonic waves of different power outputs and temperature rises for the system was measured via a thermometer installed in the sand pack. Meanwhile, for the temperature rises experiments conducted in an ultrasonic bath, the sand was packed and saturated with kerosene, vaseline, and SAE-10. The temperature rises of the system were recorded regularly. 3. Results and discussion 3.1. Thermal conductivity of fluids The effective thermal conductivity (ETC) of porous media is one of the prominent properties for thermal design and numerical simulations [30]. The study of the ETC in fluid saturated porous media can be used in petroleum science and engineering. Temperature, pressure, and grain morphology are the most important factors affected the ETC [30].
610
E. Mohammadian et al. / Ultrasonics 53 (2013) 607–614
Table 1 Testing fluid properties used in the displacement experiments.
a b c
Name of fluid
Densitya @ 27 °C (g/cm3)
Viscosityb @ 25 °C (cp)
Viscosityb @ 40 °C (cp)
IFTc with brine @ 27 °C (dyne/cm)
IFTc with de-aerated brine @ 27 °C (dyne/cm)
Thermal conductivity 20 °C (Watt/m °C)
Kerosense Vaseline SAE-10 Normal brine De-aerated brine Water 3.5% saline Quartz grains Stainless steel
0.770 ± 0.0001 0.856 ± 0.0001 0.923 ± 0.0001 1.03 1.04 1 2.65 –
0.99 ± 0.01 22.0 240 0.92 0.93 – – –
0.9 ± 0.01 12 65 0.65 0.65 – – –
31 38 18 – – – – –
33 35 20 – – – – –
0.145 0.140 0.146 – – 0.058 [31]–0.6 [32] 6.5–11.3 [33] 14–14.3 [34]
Hydrometer method (ASTM D1298). Cannon–Fenske Routine viscometer (kinematic viscosity, ASTM D445). Du Nouy type tensiometer (Kruss Tensiometer Platinum–Iridium ring, ASTM D1331).
According to Abdulagatova [30], the morphology complexity of the rock structures is the main difficulty to predict the ETC in the porous media; however,the morphology of the studied samples ranges from sub-angular to sub-rounded in shape. The ETC values vary with the temperature (0–250 °C) whereas up to 16% at low pressure (0.1 MPa) and up to 10% at high pressure (400 MPa). Based on the ETC values, the thermal expansion coefficients can be calculated [30]. Considering the ambient conditions, the values of thermal coefficient are negligible. The thermal coefficient at the pressure range of <100 MPa sharply decreases while at the high pressure very slightly decreases. At low pressure, the change of temperature is very slight for the thermal coefficient. However, at the high pressure, the temperature changing of the ETC is large. As shown in Table 1, the thermal conductivity for kerosene and vaseline is 0.145 and 0.140 (Watt/m 20 °C), respectively. However, this value for stainless steel and quartz grains is 6.15–11.3 and 14–14.3 (Watt/m 20 °C), respectively. Although in this research we did not correctly measure the thermal conductivity, we extracted the values of the material, or range of the values from previous studies [31–34].
3.2. Temperature rises experiments Temperature rises during stimulation with ultrasonic were reported by some authors [1,35,36]. However, the methods they used were sometimes too simplistic to show the real changes of temperature in the system; therefore, the magnitude of the effect is not sufficiently discussed. The temperature rises of the system (normal brine–saturated sand pack) were 4, 12, and 16 °C for the respective power outputs of 100, 250, and 400 W (Fig. 3). In the second part of the experiment, temperature rises for vaseline, kerosene, and SAE-10 were measured in the ultrasonic bath. Due to the same conductivity coefficient (i.e. 0.145, 0.146, and 0.140 for kerosene, SAE-10, and vaseline, respectively) in three of the cases the temperature rise was almost the same as expectation. Fig. 4 presents the temperature rise for different fluids. The temperature rises can affect fluid properties i.e. viscosity and IFT. IFT between oil and brine is a function of temperature. At elevated temperature, IFT is lower as it can be inferred from Firoozabadi and Ramey’s equation (Eq. (4)). Firoozabadi and Ramey [37] correlated surface tension using the difference between pure water and hydrocarbon density, temperature, and critical temperature. The correlation involved measured data from methane, propane, and benzene–water system. According to Firoozabadi and Ramey’s equation, the interfacial tension (r) between water–hydrocarbon (gas or oil) phases can be calculated.
r1=4 ¼
a1 Dqb1 T 0:3125 r
Dq ¼ qx qhc
Fig. 2. The schematic of the sand pack holder.
ð4Þ ð5Þ
Where: qx is the density of water (g/cm3); qhc is the density of hydrocarbon (g/cm3); a1, b1 is the constant (dyne/cm or mN/m) and Tr is the critical temperature (°C). Table 2 shows IFT reduction of different fluids versus time calculated by Firoozabadi and Ramey’s equation. Although the IFT was decreased in all the cases, the reduction is not high enough to have significant effects on the capillary number; therefore, it is not able to contribute in reduction of residual oil saturation. It should be noted that mentioned discussion is only related to IFT reduction as a consequence of temperature rises. However, the direct effect of ultrasonic waves on reduction of IFT is excluded in this section. Hence, it can be concluded that, if the waves have any effect on reduction of IFT, it cannot be related to temperature rises of the system. Viscosity of fluids is a strong function of temperature. Temperature rises as much as 24 °C will largely affect viscosity of fluids. The viscosity of water and oil will be reduced as a result of temperature rise. Since the viscosity of fluid at different temperatures was needed it was calculated via proper formulas i.e. Meehan and
E. Mohammadian et al. / Ultrasonics 53 (2013) 607–614
611
Fig. 3. Temperature of water versus time at various power outputs. Fig. 5. Reduction in viscosity of water.
Fig. 6. Reduction in viscosity of kerosene.
Fig. 4. Temperature rises for SAE-10, vaseline, and kerosene.
Table 2 IFT reduction for different fluids versus time. T (°C)
Kerosene (dyne/cm)
Vaseline (dyne/cm)
SAE-10 (dyne/cm)
23 27 31 35 39 43 47 51 55
31.0 30.5 30.0 29.5 29.0 28.6 28.1 27.7 27.3
38.0 37.4 36.8 36.2 35.6 35.0 34.5 33.9 33.4
18.0 17.7 17.4 17.1 16.8 16.6 16.3 16.1 15.8
Glaso’s relations. The measured values of viscosity (using Cannon–Fenske viscometer as shown in Table 1) have 5% variance from calculated values using Glaso’s formula. As the viscosity of water is reduced by temperature rises, it was calculated using Meehan’s [38] correlation. The reduction in viscosity is not the same for various fluids. Figs. 5–8 illustrate viscosity reduction for each case together with a third degree polynomial trend line drawn for each fluid to facilitate the comparison between slopes of the graphs. 3.3. One phase flow experiments Pressure reached a peak initially then decreased to lower values according to the intensity of waves (Fig. 9). One possible explanation is that cavitation is responsible for erratic behavior of pres-
Fig. 7. Reduction in viscosity of vaseline.
sure. As mentioned by Duhon and Campbell [1], during the implosion of bubbles, high pressure surges are created within the system. However, after all bubbles implode, the pressure drops. Therefore, cavitation could be accounted as one of the contributing mechanisms in application of ultrasonic waves. On the other hand, the decrease in the pressure drop could be attributed to increase in absolute permeability only if the viscosity of water remains the same. This is only possible if the temperature increase within the system could be considered negligible. In none of the experiments during the sonication, temperature remains constant. Since all the dynamic experiments were conducted on unconsolidated sand pack and only certain size of sand grains were
612
E. Mohammadian et al. / Ultrasonics 53 (2013) 607–614
Fig. 10. Pressure gradient versus time for various intensities of waves (equivalent to power outputs of 100, 200, 300, and 400 W) for one phase flow using deaerated brine. Fig. 8. Reduction in viscosity of SAE-10 (engine oil).
started to decline and finally stabilized to a value at the end of each experiment. Fig. 10 shows pressure response versus time for various power outputs. Considering the results from temperature effect experiment and one phase flow experiment, it is implied that, reduction in viscosity of water could be the main reason behind pressure drop observed in one phase flow experiment. This result is in agreement with that of Poesio et al. [10]; they concluded that the reduction in pressure is due to decrease in the viscosity of fluids. However, they did not conduct any other experiment to confirm their theory. 3.4. Displacement experiments
Fig. 9. Pressure gradient versus time for various intensities of waves (equivalent to power outputs of 100, 200, 300, and 400 W) for one phase flow using normal brine.
used without any cementing materials, it can be assumed that the changes in absolute permeability did not occur. The reduction in pressure is caused by the reduction in viscosity of water. As an example, during sonication using 2.4 W/cm2 pressure increased to 2.3 psi from initial value of 2.1 psi; at the end of the experiment pressure was constant at 1.8 psi which is 0.1 psi less than its initial value. It could be attributed to the reduction in the viscosity of water
DP ¼
Q lL KA
ð6Þ
Terms Q, L, A, K, and l are considered constant. As there is only one phase flowing in the system, the term l stands for viscosity of water (lw). The increase in temperature leads to viscosity reduction; since DP is proportional to lw, any reduction in the viscosity decreases the pressure drops by the same magnitude. Due to existence of cavitation as a result of using aerated water, quantification of the effect becomes difficult. But as through the whole one phase flow experiment the pressure reaches to a value lower than its initial value, it could be concluded that the reduction of pressure is due to reduction in viscosity of water [39]. This result is in agreement with that of Poesio et al. [10]. The second series of one phase flow experiments were conducted using de-aerated brine to remove effects of cavitation. The same procedure was performed and the pressure responses were recorded. Once the ultrasonic radiation started, the pressure
Recovery for water-flooding was 67%; sonication improved the recovery by 16%. Assuming relative permeability of non-wet and wetting phase at their end points, the mobility ratio for kerosene and water may be calculated at ambient temperature (27 °C) as follows:
M¼
krw lo 0:3 0:98 ¼ 0:44 ¼ kro lw 0:7 0:95
ð7Þ
The above mobility ratio is only valid for the ultrasonic stimulated water-flooding if viscosity remains the same throughout the experiment. In this research the point of interest is the viscosity ratio fluids and not the relative permeability of fluids. During ultrasonic stimulation, temperature rises were observed for the system. 10 °C increase in the temperature reduces viscosity of water to 0.79 cp (17% decreases) and viscosity of kerosene to 0.79 cp (20% decrease). Therefore, the viscosity ratio of the system becomes 1. Having IFT of 32 dyne/cm for brine and kerosene, the capillary number for the water is 1.58 107. Capillary numbers in orders of 105–104 are needed to decrease residual oil saturations. Another interesting result was formation of emulsion. After three hours the two phases were completely separated and the interface between the phases could be easily seen. Fig. 11 shows total recovery of kerosene as a result of straight and ultrasonic stimulated water-flooding. In the case of vaseline, the viscosity ratio at 27 °C is 23.9. The recovery for normal water-flooding is 56% ultrasonic stimulated water-flooding added 11% to the recovery. Fig. 12 shows total recovery of vaseline. Viscosity reduction is known as one of the significant mechanisms in case of vaseline as viscosity ratio reduces to 13.9 from initial value of 23.9 as the result of 10 °C temperature rise. Therefore, it can be concluded that ultrasonic stimulation increases the recovery via reducing mobility ratio and improving sweep efficiency. In the case of vaseline also, formation of emulsion
E. Mohammadian et al. / Ultrasonics 53 (2013) 607–614
613
4. Conclusions Series of displacement and supplementary experiments were conducted in this study and the following conclusions were made:
Fig. 11. Recovery of kerosene as result of straight and sonicated water-flooding.
Fig. 12. Recovery of vaseline as result of straight and sonicated water-flooding.
The recovery of water-flooding increases as a result of sonication for all the cases (from 3% to 16%). The recovery of ultrasonic assisted water-flooding was higher for less viscous fluid being kerosene. Where normal brine was used, the recovery was higher in comparison with the cases where de-aerated brine was used. This could be explained through the existence of cavitation in the system when using normal brine instead of de-aerated brine. Severe temperature rises were observed in the experiments. This leads to reduction in viscosity of fluids as well as reduction in the interfacial tension. The reduction in viscosity of fluid is detected as one of the contributing mechanisms of production. On the other hand, the temperature rises are not high enough to reduce the IFT to a large extent, in other word IFT reduction from temperature rises are so small that cannot contribute in improving the recovery. The one phase flow experiments were conducted using brine. The initial increase in the pressure in one phase flow experiments (using normal brine) can be attributed to cavitation. Conducting one phase flow experiments (using normal and de-aerated brine) proved that the reduction in pressure is caused by reduction in viscosity of water as results of ultrasonic stimulation. Formation of emulsion was observed during displacement experiments for vaseline and kerosene. The generated emulsion was unstable and the phases were completely separated after 3 h. It could be concluded that cavitation, viscosity reduction and emulsification are contributing mechanisms in application of ultrasonic waves to water-flooding.
References
Fig. 13. Recovery of SAE-10 by straight and sonicated water-flooding.
was observed. Assuming the IFT value of 38 dyne/cm for vaseline and brine, the capillary number for water-flooding is in orders of 107. It can be concluded that increase in the oil recovery cannot be attributed to reduction of IFT from temperature rise. SAE-10 is a viscous fluid (243 cp at 23 °C). The amount of oil produced by water-flooding was 12%. The sonication added merely 3% to the recovery (Fig. 13). 10 °C increase in the temperature (around 23 °C increase in the temperature was observed by temperature experiment), is changing viscosity of fluids to a large extent. Viscosity ratio reduces to 97.1 from initial value of 255.0. Therefore, one of the effective mechanisms in improving the recovery is viscosity reduction. One may expect higher recoveries considering huge reduction in viscosity ratio. But an unexpected result was deposition of paraffin as results of sonication. It was observed during emulsification experiment which was a part of the author’s previous work [40]. In this case also the IFT reduction from temperature increase is not large enough to reduce capillary number effectively and reduce the residual oil saturation.
[1] R.D. Duhon, J. Campbell, The effect of ultrasonic energy on the flow of fluids in porous media, in: SPE 1316, The 2nd Annual Eastern Regional Meeting of SPE/ AIM, Charlston, USA, 1965. [2] K. Naderi, T. Babadagli, Effect of ultrasonic intensity and frequency on oil/ heavy–oil recovery from different wettability rocks, SPE International Thermal Operations and Heavy Oil Symposium, Calgary, Canada, 2008. [3] T. Hamida, T. Babadagli, Fluid–fluid interaction during miscible and immiscible displacement under ultrasonic waves, European Physical Journal 60 (2007) 447–462. [4] M. Amro, M. Al-Mobarky, Al-Homadhi, Improved oil recovery by application of ultrasound waves to water-flooding, in: SPE 105370, SPE Middle East Oil & Gas Show, Bahrain, 2007. [5] I. Najafi, A mathematical analysis of the mechanism of ultrasonic induced fluid percolation in porous media, in: SPE 141126, SPE Annual Technical Conference And Exhibition, Florance, Italy, 2010. [6] T. Hamida, Effect of Ultrasonic Waves on Immiscible and Miscible Displacement in Porous Media, Master thesis, University of Alberta, Canada, 2006. [7] T. Hamida, T. Babadagli, Effects of ultrasonic waves on the interfacial forces between oil and water, Ultrasonics Sonochemistry 15 (2008) 274–278. [8] G. Albert, J.V. Bodine, No. Paten 2,667,962, US Patent, California, 1948. [9] E. Simkin, Advanced vibroseismic technique for water-flooded reservoir stimulation, mechanims and field tests results, in: The 6th European IOR Symposium, Stavenger, Norway, 1991. [10] P. Poesio, G. Ooms, A. Schraven, F.V.D. Bas, Theoretical and experimental investigation of acoustic streaming in a porous material, Physical Review E (66) (2002). [11] X. Guo, High Frequency Vibration Recovery Enhancement Technology in the Heavy oIl Fields in China, in: SPE 89656, SPE International Thermal Operations and Heavy Oil Symposium and Western Regional Meeting, California, USA, 2009. [12] V.N. Nikolaevsky, G.S. Stepanova, Non-linear seismics and the acoustic action on the oil recovery from oil pool, Acoustical Physics 51 (2005) 131–139. [13] E.M. Simkin, Oil will return in three months, Energy (Energiya) 3 (1985) 44–47.
614
E. Mohammadian et al. / Ultrasonics 53 (2013) 607–614
[14] A.S. Odeh, Mathematical modeling of the behavior of hydrocarbon reservoirs– the present and the future, in: J. Bear, M.Y. Corapcioglu (Eds.), Advances in Transport Phenomena in Porous Media, Martinus Nijhoff Publ, 1987, pp. 821– 848. [15] I.A. Beresnevand, P.A. Johnson, Elastic–wave stimulation of oil production: a review of methods and results, Geophysics 59 (6) (1994) 1000–1017. [16] O.L. Kuznetsov, E.M. Simkin, Transformation and Interaction of Geophysical Fields in Lithosphere, Nedra Press, 1990. 106–112. [17] R.D. Duhon, An Investigation of the Effect of Ultrasonic Energy on the Flow of Fluids in Porous Media, Ph.D thesis, Univ. of Oklahoma, 1964. [18] H. Hertz, Über die Berührung fester elastischer Körper (On the contact of elastic solids), Journal für die Reine und Angewandte Mathematik 92 (1882) 156–171. For English translation see miscellaneous papers by H. Hertz, (Ed.), Jones and Schott, London, 1896.. [19] G. Michlmayr, D. Cohen, D. Or, Sources and characteristics of acoustic emissions from mechanically stressed geologic granular media – A review, Earth–Science Reviews 112 (2012) 97–114. [20] R.D. Mindlin, H. Deresiewicz, Elastic spheres in contact under varying oblique forces, Journal of Applied Mechanics 75 (1953) 327–344. [21] C. Hardy, C.N. Baronet, G.V. Tordion, The elasto-plastic indentation of a halfspace by a rigid sphere, International Journal for Numerical Methods in Engineering 3 (1971) 451–462. [22] I.M. Hutchings, Energy absorbed by elastic waves during plastic impact, Journal of Physics D: Applied Physics (12) (1979). [23] K.L. Johnson, Contact Mechanics, Cambridge University Press, Cambridge, 1985. [24] L. Kogut, K. Komvopoulos, Analysis of the spherical indentation cycle for elastic perfectly plastic solids, Journal of Materials Research 19 (2004) 3641–3653. [25] D. Levine, Jamming and the statistics of granular materials, in: A. Liu, S. Nagel (Eds.), Jamming and Rheology, Taylor & Francis, London and New York, 2001, pp. 9–19. [26] E. Somfai, J.N. Roux, J.H. Snoeijer, M.V. Hecke, W.V. Saarloos, Elastic wave propagation in confined granular systems, Physical Review E (72) (2005).
[27] L.A. George, M.M. Dewoolkar, D. Znidarcic, Simultaneous laboratory measurement of acoustic and hydraulic properties of unsaturated soils, Vadose Zone Journal 8 (2009) 633–642. [28] Z. Lu, J.M. Sabatier, Effects of soil water potential and moisture content on sound speed, Soil Science Society of America Journal 73 (2009) 1614–1625. [29] S. Griffiths, A. Rescaglio, F. Melo, Ultrasound propagation in wet and airless non-consolidated granular materials, Ultrasonics 50 (2010) 139–144. [30] Z. Abdulagatova, I.M. Abdulagatov, V.N. Emirov, Effect of temperature and pressure on the thermal conductivity of sandstone, International Journal of Rock Mechanics & Mining Sciences 46 (2009) 1055–1071. [31] R.M. Digullio, A.S. Teja, Thermal conductivity of aqueous salt solutions at high temperatures and high concentrations, American Chemical Society, 1992. 1081–1085. [32] H.P. Young, D. Hugh, University Physics, seventh ed., Addison Wesley, 1992. [33] K.M. Smits, T. Sakaki, A. Limsuwat, T.H. Illangasekare, Determination of the thermal conductivity of sands under varying Moisture drainage/wetting and porosity conditions–applications in near–surface soil moisture distribution analysis, Hydrology Days (2009). [34] J.G. Hust, P. Giarratano, Thermal conductivity and electrical resistivity, Standard Reference Materials (1975). [35] W.I. Chen, Influence of Ultrasonic Energy Upon the Rate of Flow of Liquids Through Porous Media, PhD Thesis, West Virginia University, 1969. [36] H.V. Fairbanks, W.I. Chen, Ultrasonic acceleration of liquid flow through porous media, Chem. Eng. Symp. vol. 67, 1971 pp. 108 . [37] A. Firoozabadi, H.J. Ramey, Surface tension of water hydrocarbon systems at reservoir conditions, Journal of Canada Petrol Technology 27 (3) (1988) 41–48. [38] D.N. Meehan, A correlation for water compressibility, Petroleum Engineer (1980) 125–126. [39] E. Mohammadian, Ultrasonic Assisted Water-Flooding, M.Sc. thesis, Universiti Teknologi Malaysia, 2010. [40] A. Danesh, PVT and Phase Behaviour of Petroleum Fluids, Herriot–Watt University, Edinburgh, Scotland, 1998.