Influence of fluid and operating parameters on the recovery factors and gas oil ratio in high viscous reservoirs under foamy solution gas drive

Fuel 197 (2017) 497–517

Contents lists available at ScienceDirect

Fuel journal homepage: www.elsevier.com/locate/fuel

Full Length Article

Influence of fluid and operating parameters on the recovery factors and gas oil ratio in high viscous reservoirs under foamy solution gas drive Bashir Suleman Abusahmin a, Rama Rao Karri a,⇑, Brij B. Maini b a b

Department of Petroleum & Chemical Engineering, Universiti Teknologi Brunei, Brunei Darussalam Department of Chemical and Petroleum Engineering, Schulich School of Engineering, University of Calgary, Calgary, Alberta T2N1N4, Canada

h i g h l i g h t s
Series of 16 pressure depletion tests carried to evaluate the features of foamy oil.
Foamy oil behavior for four heavy oil-solvent systems were studied systematically.
4 systems: heavy mineral oil with methane, ethane, CO2 and crude oil with methane.
Study investigates the effects of number of parameters on oil recovery trends.
Each high viscous oil system was characterized by evaluating different parameters.

a r t i c l e

i n f o

Article history: Received 26 August 2016 Received in revised form 1 February 2017 Accepted 14 February 2017

Keywords: Foamy oil Gas oil ratio Depletion tests Heavy oil reservoirs Recovery factor Solution gas drive

a b s t r a c t Foamy oil flow behavior is reported in several high viscous reservoirs in the world, wherein reduction of pressure was noticed to be the main factor of such characteristics. It is also believed to be a significant recovery mechanism in numerous high viscous heavy oil reservoirs that have revealed higher recovery factors when compared with the fluid flow using ordinary Darcy equation. This research investigates the effects of number of factors that influence the oil recovery trends, as well as the production rates in high viscous reservoirs under foamy solution gas drive behavior. The factors investigated comprised of refined mineral oil versus crude oil, saturation pressure, oil viscosity, drawdown pressure, flow direction, solution gas, pressure depletion rate and gas oil ratio (GOR). Live oil-gas system is prepared by blending a mixture of dead oil with gases such as CO2, ethane and methane. Each high viscous live oil system was completely characterized by evaluating fluid parameters and operating parameters. The significant outcome of the depletion tests confirms that the decreasing pressure depletion rate result in lower performance. At the similar rate of pressure depletion, higher oil recovery was obtained with methane saturated oil compared to either ethane/CO2 systems, even though it had the lowest solution GOR. At saturation pressure of 500 psi, the solution GOR was 9.1 m3/m3, 28 m3/m3 and 33 m3/m3 with methane, CO2 and ethane gas respectively, whereas solution GOR of methane saturated with crude oil were found to be 11 m3/m3. Both mineral and crude oil systems displayed similar decline in the oil recovery performance with decreasing pressure depletion rate. In high depletion rate tests, the recovery factor was 26.1%, 23.7% and 19.6% with respect to methane, ethane and CO2 respectively, whereas in slow depletion runs, the recovery factor declined from 13.1% with methane to 5.5% with CO2. Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction High viscous reservoirs are a significant source of global energy supply and extraction of oil from these reservoirs have gained more prominence to meet the demand of petroleum and its associated products. Foamy oil terminology is commonly used to

⇑ Corresponding author. E-mail addresses: [email protected], [email protected] (R.R. Karri). http://dx.doi.org/10.1016/j.fuel.2017.02.037 0016-2361/Ó 2017 Elsevier Ltd. All rights reserved.

pronounce a dispersed gas–liquid two-phase fluid, which present in high viscous oil reservoirs, ranges from 10,000 to 100,000 centipoise, mostly in countries like Canada, Oman, Russia, China and Venezuela performed under solution gas drive [1–7]. Recovery factor and gas oil ratio were perceived as prominent features in the high viscous oil reservoirs under foamy solution gas drive when compared with conventional solution gas drive production behavior [1,8–12]. Numerous theoretical and experimental studies [5,10,13–21] have been investigated to understand the

498

B.S. Abusahmin et al. / Fuel 197 (2017) 497–517

mechanisms of the unusual oil recovery factor in high viscous oil reservoirs under solution gas drive like Lloydminster, Canada. The better oil recovery factor can be partly attributed to foamy oil flow. During the depletion process, wherein the pressure of the sand pack falls below the bubble point pressure of high viscous oil, a section of small bubbles are developed and eventually trapped in high viscous heavy oil. Because of the potential energy in high viscous reservoir, the amount of gas bubbles starts to migrate within the viscous oil, and displaces the oil to move in the flow direction. While pressure drops in the process, the gas bubbles grow further (primarily by expansion and diffusion), coalesce and under certain conditions, travel towards the top of the system due to buoyancy forces [22]. As the pressure further decays much lower, the bigger bubbles will merge together and form a continuous gas-phase and then the foamy oil flow culminate [23,24]. Simultaneous flow of gas and oil in porous media is a frequently encountered in oil production from oil-bearing rock formations. The oil in an underground formation is generally found at high pressure with substantial amount of natural gas dissolved in it. To illustrate the behavior of natural gas in foamy oil, gas dissolution tests were conducted at different pressures, and a methodology was established for deriving the gas diffusion coefficient [7,25,26]. Reservoir pressure decreases when oil is extracted from it. This decline in pressure reduces the gas solubility and at a certain characteristic lower pressure (bubble point pressure), the gas starts oozing out of solution. As the pressure declines further, the gas starts to flow with the oil towards the production well. The simultaneous flow of oil and gas through porous media is traditionally defined by expanding Darcy’s law to dual phase flow, by introducing the concept of relative permeability [27]. This description of dual phase flow is established based on the observation that the two phases generally flow separately, but in continuous flow channels and the flow mechanism of each phase is driven by the pressure gradient in that particular phase only. Thus in dual phase flow of gas and oil, the gas flows mostly through a pore system and the oil flows through a parallel but separate system. The distribution of the two phases is controlled by interfacial tensions that work to abate the free energy of the interface. This abatement of the free energy implies that the wetting phase would occupy the smallest available pores, which have larger surface area per unit volume of pore space, and the non-wetting phase would migrate to larger pores. An essential requirement for this description of the two-phase flow to work is that the distribution of the two phases is dependent only on the relative saturation of these two phases that are present and essentially independent of the flow velocity. This makes the relative permeability of each phase a function of only its own saturation [28–30]. The preceding description of the two-phase flow works if the fluid distribution is controlled by the interfacial tension and the contribution of viscous forces towards fluid distribution remains insignificant. It works better for modeling the oil production in conventional oil reservoirs by solution gas drive. However, it does not constantly work for modeling solution gas drive related to high viscous reservoirs. Most of the high viscous reservoirs exhibit unusual production behavior, both in terms of unusually high initial recovery factors and higher than expected well productivity, which cannot be modeled by traditional description of two-phase flow [31]. It is because, the conventional relative permeability based description of two phase flow is not applicable to foamy oil flow in which the gas flows in the form of dispersed bubbles in oil. The conventional model predicts that gas oil ratio in the produced fluid stream will increase rapidly with continued depletion but, in reality, in many heavy oil reservoirs this does not occur. The producing gas oil ratio remains much lower than expected and oil production continues down to low average reservoir pressure. It is as if

there is something present to severely reduce the flow of gas and thereby divert the drive energy towards continued oil production. It was concluded that, this uncharacteristic production behavior is closely related with the observed foamy oil nature of the produced oil. This type of inconsistent production was first described by Smith [32], who noted that the cold production of heavy oils from several Canadian reservoirs did not fit the conventional solution gas drive models. He described two interesting features of such production: (1) the oil was produced in the form of thick foam that was remarkably stable and (2) substantial volume of sand was produced with the heavy oil. Both of these appear to play a role in the anomalous production behavior. Sand production leads to dilation of sand near the production well and to formation of wormholes that extend the reach of the production well and increase the inflow rate [33]. The foaminess of oil alters the distribution of gas and oil in the pore space and delays the formation of continuous gas phase during oil production [22]. The difference between the solution gas drives in conventional oil and high viscous reservoirs is the relative magnitude of viscous forces. Because of the low mobility of heavy oils, it is necessary to apply very high drawdown pressure in production wells. This large difference in the average reservoir pressure and the well pressure increases the gradient (pressure) and makes the viscous forces comparable to capillary forces. The local capillary number can become high enough to activate an isolated gas bubble which leads to a dispersed gas flow. The gas exsolution study specifies that the pore rebound response pressure depends on liquid and gas properties and sand matrix deformation properties [34–36]. The process has some similarities to in situ generation of emulsions during simultaneous flow of oil and water at high rates and in situ formation of aqueous foams in the presence of surfactants. However, there are also major differences in their characteristics. In foamy oil, the size of dispersed bubbles is much larger than emulsion droplets and the volume fraction of dispersed phase is much smaller than aqueous foams. The mobilization of dispersed gas bubbles dramatically changes the nature of two phase flow. Because the gas is no longer continuous, its flow is no longer confined to a separate pore network. Moreover, the flow of gas, at least on the macroscopic scale, is now driven primarily by pressure gradient in the oil. The gas bubbles have to displace the oil in order to progress in the direction of flow and their movement is now greatly affected by the oil viscosity. The larger the oil viscosity, lower would be the mobility of gas bubbles [37]. This reduced gas mobility greatly improves the oil recovery performance by preventing, or at least greatly reducing, the rapid dissipation of reservoir energy with production of large volume of gas. Hence, the gas oil ratio now remains relatively small. Most of the previous laboratory experimental tests are mainly focused on heavy oil methane system [4,10,15,38–43]. From these studies it was established that the oil recovery factor rises with the increase in pressure depletion rate. This conclusion can be attributed due the fact that propane has higher solubility in high viscous oil. In one of the recent study [6] where pressure depletion tests on foamy oil flow were conducted in a one-meter-long sand pack system to investigate the effect of pressure depletion rates on different heavy oil–solvent systems in porous media. Pure methane, pure propane, and a mixture of methane and propane were used as solvents in these tests. In their study, they have concluded that as the pressure decline rate of the high viscous oil methane and propane systems increases, the foamy oil pressure decreases. To the best knowledge of all the authors, there has no published work reported in open literature on the foaminess behavior of saturated mineral oil with gases like methane, ethane, carbon dioxide and saturated crude oil – methane system on recovery factors and production rates. To investigate the effect of the above aspects and influence of parameters such as gas-oil-ratio, saturation pressure, and the type

B.S. Abusahmin et al. / Fuel 197 (2017) 497–517

of oil used on oil recovery and production rates on flow of foamy oil in high viscous reservoirs, a sequence of experimental runs were conducted systematically in this research study on a two meter long porous media (sand pack) using saturated mineral oil with different solvents. Various tests were conducted to investigate the production behavior under different saturation pressure and depletion rates, which are presented in detail in this research paper. This paper proceeds with a description of experimental setup, methodology, preparation of necessary components, procedure to carry depletion experimental runs, description of different live oils and observations derived from the depletion tests. The results are presented and analyzed systematically to derive the effects of above mentioned relevant parameters that are mainly related to oil recovery trends, as well as the production rates in high viscous reservoirs under foamy solution gas drive behavior.

2. Experimental setup To investigate the influence of drawdown pressure on mineral and high viscous oil solution gas drive recovery, the experimental setup is systematically designed as shown in schematic diagram presented in Fig. 1. The key component of this experimental setup is a two meter long porous media (sand pack) as the reservoir along with seven pressure transducers. On either sides of the sand pack, there are two fluid distribution caps at the inlet and the outlet with two stem manifold valves. These ends act as injection and production ports. The oil gas mixture was very well mixed to reach to the desired saturation pressure which is then injected from one end of the sand-pack using a transfer vessel to the sand-pack. A back pressure regulator is fixed on the other end of sand pack. Oil and gas collectors are assembled to measure the recovery of oil and gas through the system. To data log the significant process parameters in the whole process; data logging software deLogger pro with data tracker 505 hardware along with the mass flow controller was used and the recorded data were analyzed. The instruments and sensors used in the experimental setup for data collection were thermocouples, pressure gauges and a digital weighing balance. The thermocouples are served to monitor the temperature in the sand pack, whereas the pressure gauges were located along the sand-pack and their functions were to record the pressure at each point or locator until the depletion test is done. The digital weighing balance was used to monitor and record the whole amount of oil produced. The sand pack holder is an annular in shape and is made of steel (SAE 316). It is a two meter long and 2.500 ID cylindrical housing that has a lead pipe of (100 ID and ¼00 thickness) inside the cylindrical barrel creating an annular space between them for the purpose

499

of packing the desired sand. The sand was dry packed in this space while the holder was being slightly vibrated. The sand (inside the lead pipe) was compressed by pressurizing with nitrogen gas and water which were used as the overburden pressure medium. The pressure was kept at 3447.4 kPa (500 psi) above the system pressure. The outer surface of the sand-pack was well insulated to reduce the effect of external thermal fluctuations in the room temperature, however all the pressure depletion decline tests were conducted at room temperature and the pressure at the back pressure regulator was decreased step-by-step using a nitrogen gas bottle that is connected to the back pressure regulator and a mass flow controller. 3. Methodology The following components and necessary modules are prepared before conducting the experimental depletion tests. 3.1. Sand pack (porous media) preparation Porous media, (sand pack reservoir) is the crucial component in this whole experimental set-up. Initially, the sand pack was evacuated using a vacuum pump for about one day, to achieve a good vacuum. Thereafter de-ionized water was introduced into the porous media and the volume of water taken up by the porous media (sand-pack) was measured. This procedure is used to measure the porosity and pore volume. The water used in filling the pressure transducers was deionized water that was boiled to remove dissolved gases to minimize the formation of gas bubbles in the pressure transmitting lines and resulting artifacts in the pressure measurements. 3.2. Overburden pressure Once the sand pack was fully packed, the system need to be maintained at high overburden pressure. The lead pipe is the inner annular cylinder containing sand. The expandable lead pipe was pressurized within inside, to compress the sand against the inner surface of the cylindrical housing. By doing so, any air pockets within the lead pipe will be eliminated. Water and nitrogen gas were used in the sand-pack to get the desired overburden pressure, which was 1000 psi in all tests carried in this research study. A small pressurized nitrogen tank was connected to the lead pipe to maintain the overburden pressure. The pressure in the lead pipe causes the sleeve to expand radially and compress the sand pack, hence this porous media is ready to conduct depletion experimental tests.

Fig. 1. Schematic diagram of whole experimental setup to investigate the effect of drawdown pressure on recovery factor of solution gas drive.

500

B.S. Abusahmin et al. / Fuel 197 (2017) 497–517

3.3. Properties of sand pack The physical properties of the sand pack are systematically determined which are listed in the following Table 1, before the experiment runs are carried out. The method and determination of these properties are not presented here, as they are the not within the scope of this research article. 3.4. Preparation and injection of live oil Blending the desired gas with dead oil using a mixing cell, a live oil is produced. Initially, the mixer was filled with dead oil, about six liters (three quarter by volume), and methane/ethane/ carbon dioxide gasses were used at different saturation pressures at 500, 300, and 500 psi respectively. The selected gas was introduced into the mixer at a pressure marginally higher than the desired bubble point (BP) pressure of the live oil. The bubble point pressures of CH4, C2H6 and CO2 are 400, 260 and 400 psi respectively. The mixture was thoroughly mixed to dissolve the gas into the oil. Once the system pressure is reduced below the BP, more gas was added to the mixer to reach the desired pressure. Final pressure was the desired saturation pressure, which varied with the gas used. This process was continued until there was no further decrease in the pressure due to gas dissolution. This mixture is retained for 24 h to achieve equilibrium steady state before it is used for experimental runs. Dead oil was introduced into the porous medium (sand-pack) to displace the de-ionized water at constant rate and this process was continued until the traces of water disappeared in the oil collected at sand-pack outlet. Once the porous media was completely saturated with the dead oil, it is then filled by injection of live oil while maintaining a back pressure higher than the saturation pressure of the live oil. Approximately, about 2½ pore volume of live oil were introduced to displace the dead oil. The displaced oil was produced from the production port through a back pressure regulator, and its pressure was set at 25 psi above the saturation pressure (500 psi) to eliminate chances of free gas formation in the porous media. After completion of two pore volumes of live oil injection, the gas oil ratio of the produced solution was measured using a 10 ml pycnometer. To determine the solution gas oil ratio of the produced oil, a sample (10 ml) of live oil was collected from sample collection port. If this gas oil ratio was found to be the same as the solution gas oil ratio of the injected live oil, then the displacement of dead oil was considered complete. Otherwise the live oil injection was continued until this condition was achieved. 3.5. Fluid properties determination A refined mineral oil and crude oil were used as dead oil in the depletion tests carried in this research study. The crude oil used in this study was from Grimbeek field in Argentina, which is Newtonian in nature at room temperature. Methane, carbon dioxide, and ethane gases were used as the gas medium. Live oil was prepared by mixing the desired oil with respective gas and the corresponding viscosity of the live oil was determined at different saturation

Table 1 Sand pack (porous media) property. Length of sand pack (cm) Pore volume (cc) Absolute permeability Porosity (%) Sand grain size (Mesh) Water saturation (%) Compressibility (psi1) Overburden pressure (psi)

200 1620 24 34 30–50 8 7.49  106 1000

pressures. The profiles of variation of dead oil viscosity at various temperatures are shown in the Fig. 2. The viscosities of the refined mineral oil and the crude oil were measured using a cylindrical rotational viscometer, Thermo Haake model RotoVisco-1 and RheoWin data manager software. The cylindrical viscometer which has 3.4 cm internal diameter and 6 m height was first calibrated with a standard fluid of known viscosity. After calibrating the cylindrical viscometer, the viscosity of both oils was measured at room temperature. The live oil viscosity was measured using a viscometer at high pressure and room temperature. The viscosities were measured at different saturation pressures using the sandpack as a viscometer. The live oil was made to flow through the sand-pack at a fixed flow rate while the pressure transducers along the sand-pack were monitored using Delogger Pro software and the differential pressure was determined. The effective permeability to dead oil was measured to be 24 Darcy. The live oil viscosity was determined from the measured pressure drop at a known flow rate assuming that the effective viscosity to live oil was same as that for the dead oil. The pressure drop during flow of live oil was determined near the completion of the live oil injection and compared with the pressure drop recorded during the flow of dead oil. The live oil viscosity was calculated by multiplying the dead oil viscosity with the ratio of pressure drops during live oil flow and dead oil flow at the same flow rate. Surface tension of the mineral oil and the crude oil were measured using the drop shape analysis method with Angstrom Tensio-meter. The averaged values of several measurements for mineral oil and crude oil were considered in this study. Surface tension of live oil was measured using the drop volume technique, which relies on the measurement of the volume of drops formed at the tip of a capillary tube. A Jergusen cell with a capillary tube injector at the top was used for this purpose. Live oil was injected into the cell at a very low flow rate using a positive displacement pump and the average time for formation and detachment of individual drops was determined. The drop volume was then calculated from the known flow rate and the time elapsed per drop. The average drop volume was used for surface tension calculation. The fluid properties of live oil produced by blending of respective gases with mineral oil and crude oil are presented in Table 2. The viscosities of different dead oil and live oil are of wide range, hence depicts the features of different high viscous oils to study the solution gas drive recoveries. 3.6. Procedure to conduct depletion experimental runs To prepare and regenerate the porous media for new depletion tests (decline tests), drainage and imbibition processes were applied, wherein two-pore volumes of dead oil which is free of gas were introduced into the porous media. The BPR is then set at a higher pressure than the saturation pressure of live oil. The same regenerated sand-pack was used for all the further depletion tests after regeneration. The most significant issue was to get rid of any gas that was trapped in the porous media from the previous tests. Later an imbibition process is applied, where about 2½ pore volume of the live oil was introduced into the porous media to confirm that the porous media is completely saturated with live oil i.e. dead oil plus a desired type of gas prior to the commencement of depletion test. After that a soak period is applied for a day for pressure to stabilize in the system. After confirming that all the pressure readings are steady, then the depletion runs are started. The back pressure regulator was adjusted and decreased step-by-step by releasing nitrogen gas from the dome. The adjustment of mass flow controller provides a constant decline in the pressure setting. During the process of the experimental runs, all the prominent parameters such as pressure, room temperature, total gas and oil produced were monitored for further analysis. Oil produced from the porous media was then collected in a container that was placed

501

B.S. Abusahmin et al. / Fuel 197 (2017) 497–517

Fig. 2. Profiles of refined mineral oil and crude oil viscosities at different saturation temperatures in sand-pack.

Table 2 Fluid properties of live oil produced by blending of respective gases with mineral oil and crude oil. Parameter

Refined mineral oil

Crude oil

Sat. mineral oil with CH4

Sat. mineral oil with C2H6

Sat. mineral oil with CO2

Sat. crude oil with CH4

Saturation Pressure (psi) Density (kg/m3) Viscosity @ 23 °C (mPa.s) Compressibility (psi1)  106 Surface Tension (dyne/cm) Solution GOR (Std. m3/m3)

N/A 896 1876 6.8

N/A 936 2608 6.9

500 891 1080 4.52

300 862 649 4.40

500 879 1031 4.35

500 928 1300 4.37

31 N/A

29 N/A

28 9.1

26 N/A

29.56 28

20 11

on a digital balance, whereas the gas released from the porous media sand pack was pumped into the gas cylinder. The depletion tests were considered complete whenever the mass flow controller indicated a zero sign. All the above mentioned procedures for carrying out further experimental runs were need to be repeated. 4. Results and experimental outcomes from depletion tests that are conducted using various gases blended with saturated mineral/crude oil 4.1. Depletion experimental runs Initially four depletion tests were carried out using mineral oil saturated with methane gas. The live oil saturation pressure in these tests was maintained at 500 psi and the pressure depletion rates were arbitrarily set at 0.406 psi/min, 0.247 psi/min, 0.086 psi/min, and 0.021 psi/min respectively for four different experimental runs. The depletion rate is steadily set by varying the rate of gas withdrawal from the back pressure regulator system. An additional depletion test using methane saturated mineral oil at 300 psi saturation pressure was conducted at depletion rate of 0.264 psi/min to estimate the consequence of saturation pressure on performance rate. To study the effect of gas properties on the gas oil recovery, additional three depletion tests were carried out using mineral oil saturated with ethane at depletion rates of 0.398 psi/min, 0.109 psi/min, and 0.025 psi/min respectively. An addition of four more depletion tests were performed using carbon dioxide saturated with mineral oil at saturation pressure of 500 psi, and these runs are carried out at different depletion rates of 0.434 psi/min, 0.220 psi/min, 0.067 psi/min, and 0.035 psi/min

respectively. In order to assess the effect of change of dead oil on gas oil recovery, further four more depletion tests were conducted where crude oil is blended with methane and these tests were carried at depletion rates of 0.350 psi/min, 0.230 psi/min, 0.048 psi/ min, and 0.023 psi/min respectively. Overall, sixteen depletion experiments were conducted using different gases blended with mineral oil/crude oil in this research study and summary of these tests along with solution gas oil recovery is tabulated in the Table 3. All these experiments were done at least for three times for repeatability and found that the variance is less than 5% among each repeated experiment. 4.2. Analysis of depletion tests conducted using mineral oil and methane gas As indicated in Table 3, altogether five pressure depletion tests were done using refined mineral oil saturated with methane gas at 300 psi and 500 psi at different depletion rates. All of these tests were investigated at room temperature (23 ± 0.5 °C). Table 4 presents a summary of the experimental conditions and the important results of these tests in this system. Details of the experimental observations for each depletion test are presented below. 4.2.1. Depletion tests done with mineral oil saturated over methane for a period of one day The pressure at the producing port of the porous media was reduced from the starting pressure of 3447.4 kPa (500 psi) down to 101.353 kPa (14.7 psi) in 24 h at a depletion rate set at 0.406 psi/min. The variation of both the pressure at the production port of the porous media and the average pressure in the porous

502

B.S. Abusahmin et al. / Fuel 197 (2017) 497–517

Table 3 Summary of the depletion tests carried with three different gases (CH4, C2H6, and CO2) and two saturated oils (mineral and crude oil). Oil

Run #

Saturation pressure, psi

Depletion Type

Pressure depletion rate psi/min

Methane saturated mineral oil

1 2 3 4 5

500 psi 500 psi 500 psi 500 psi 300 psi

Very Fast Fast Medium Slow Fast

0.406 0.247 0.086 0.021 0.264

Carbon dioxide saturated mineral oil

6 7 8 9

500 psi 500 psi 500 psi 500 psi

Very Fast Fast Medium Slow

0.434 0.220 0.067 0.035

Ethane saturated mineral oil

10 11 12

300 psi 300 psi 300 psi

Fast Intermediate Medium

0.398 0.109 0.025

Methane saturated Crude Oil

13 14 15 16

500 psi 500 psi 500 psi 500 psi

Very Fast Fast Medium Slow

0.350 0.230 0.048 0.023

Table 4 Summary of depletion tests conducted with mineral oil-methane system with different pressure depletion rates. Run #

Saturation Pressure (psi)

Solution GOR

Pressure Depletion Rate (psi/min)

Recovery Factor (%)

1 2 3 4 5

500 500 500 500 300

9.1 9.1 9.1 9.1 6.1

0.406 0.247 0.086 0.021 0.264

26.1 23.9 13.7 13.1 23.8

media are monitored and plots of these profiles are shown in Fig. 3. The recovery factor of oil against the experimental run time is shown in secondary axis in the same figure. At the start of the decline test, the pressure gauge at the production port looks similar to the average pressure. However, a substantial difference between the average pressure and the pressure at the production port prolonged until the test is completed. There is almost no oil production during the initial first six hours of the decline test whereas the pressure dropped to 2551.06 kPa (370 psi). There is a brief period of time, where the average pressure in the porous media sand-pack remains constant, meanwhile the oil recovery continues rapidly. Then the average pressure starts to decline almost the same rate as the pressure at the production port. After

that, the oil production starts and hence the average pressure starts to deviate from the pressure at the production port. The final recovery factor for this test was recorded to be closely to 26.1%. The cumulative production of mineral oil – methane system is presented in Fig. 4(a). It is obvious that production gas rate is quiet dawdling, and at the same time oil production rate is fast. As the time progress, the oil production rate starts to dawdle and the gas production rate becomes higher and eventually reaches steady state from around 650 min (10 h) until the depletion test is completed. The profiles of cumulative gas and oil production are plotted against average pressure in the sand-pack as shown in Fig. 4(b). It is observed that the pressure declines to about 2551.06 kPa (370 psi) before the gas production starts. There is substantial oil

Fig. 3. Profiles of pressure and percentage recovery factor in 24 h depletion test using mineral oil – methane system.

503

B.S. Abusahmin et al. / Fuel 197 (2017) 497–517

Oil produced, ml

480

(a)

15000

400

12000

320

9000

240

6000

160

3000

80

0

Cumulave Oil Producon, mL

Cumulave Gas Producon, std. cc

cumulave gas, std.cc

18000

0 0

4

8

12

16

20

24

Time, hr Cum Gas (Std. cc)

400

Oil Produced (mL)

18000

(b)

16000

350

14000

300

12000

250

10000

200

8000

150

6000

100

4000

50

2000

0 500.0

Cumulative Gas (Std. cc)

Cumulative Oil Production (mL)

450

0 400.0

300.0

200.0

100.0

0.0

Average Pressure in Sand-Pack (psi) Fig. 4. Profiles of cumulative gas and oil production against (a) experimental run time and (b) average pressure in 24 h depletion test with saturated mineral oil – methane system.

production near this pressure gradient with little or no decrease in average pressure. During this period, growth of gas bubbles arrests the pressure declination and provides the force to drive the oil out. In the course of the latter part of experimental run, it was observed that the oil production curve cumulates with decreasing slope whereas, the gas production curve has higher increasing trend. This trend can be attributed to the fact that the continued pressure depletion produces less oil and more gas per unit waning in the average pressure. This phenomenon is observed more clearly as shown in Fig. 5(a), which presents the oil and gas production rates plotted against average pressure in the porous media. After reaching a peak around 2551.06 kPa (370 psi) in the porous media, the oil rate steadily declines while the gas rate increases gradually. Only near the end of this experimental run, when the average pressure in the porous media has become low, the gas production rate instigates to drop. This is primarily due to the rapidly increasing volume factor of the gas as the average pressure in the porous media approaches atmospheric pressure. Profiles of oil production rate and gas oil ratio in 24 h depletion test are shown in Fig. 5(b). From these profiles, it is observed that the gas oil ratio near the end of the decline test is about ten times higher than the solution gas oil ratio. The oil production rate declines as the producing gas oil ratio continues to grow, as the average pressure of the porous media declines.

4.2.2. Depletion tests conducted using mineral oil saturated with methane at a faster depletion rate which leads the experimental run over a period of one and half day The primary objective of this decline test is to analyze whether the performance of the system is significantly affected by slightly reducing the rate of pressure decline. Here the pressure was slowly decreased from initial system pressure at 3447.4 kPa (500 psi) to near atmospheric pressure at a depletion rate set at 0.247 psi/ min over a period of 1.5 days instead of 1 day (0.406 psi/min) as observed in earlier experimental run. The percentage oil recovery and variation of pressure for this experimental run is shown in Fig. 6 (a). A comparison of these trends with profiles presented in Fig. 3, shows that the behavior is very similar except for a small reduction in the percentage recovery factor and a slightly lower difference between the average pressure of the porous media and the pressure at the production port. In other words, the reduction in rate of pressure at the production port has resulted in slightly reduced drawdown pressure during the productive phase of the run and this appears to have reduced the ultimate oil recovery. The profiles of cumulative gas production and oil production against the average pressure in the porous media are shown in Fig. 6 (b). The behavioral pattern is qualitatively very similar to the trend observed in Fig. 4 (a). Therefore, it can be concluded that

B.S. Abusahmin et al. / Fuel 197 (2017) 497–517

Produced GOR

Oil Rate (ml/min)

(a) 1.20

20

0.90

15

0.60

10

0.30

5

0.00 500.0

400.0

300.0

200.0

100.0

0 0.0

Oil Rate (ml/min)

1.50

Oil Production Rate (ml/min)

Oil Production Rate (ml/min)

25

Gas Production Rate (Std. cc/min)

Gas Rate (Std cc/min)

1.50

150

(b) 1.20

120

0.90

90

0.60

60

0.30

30

0.00 500.0

400.0

300.0

200.0

100.0

Producing GOR (Std. cc/mL)

504

0 0.0

Average Sand-Pack Pressure (psi)

Average Sand-Pack Pressure (psi)

Fig. 5. Profiles of (a) oil and gas production rates (b) gas oil ratio and oil production rate in 24 h depletion test with saturated mineral oil – methane system.

(a)

Cumulative Gas

18000

(b)

400

16000

350

14000

300

12000

250

10000

200

8000

150

6000

100

4000

50

2000

0 500.0

Cumulative Gas (Std. cc)

Cumulative Oil Production (mL)

450

Cumulative Oil

0 400.0

300.0

200.0

100.0

0.0

Average Pressure in Sand-Pack (psi) Fig. 6. Profiles of (a) oil recovery factor and pressure variation (b) Cumulative gas and oil production against average porous media pressure for a depletion rate of 1.5 days in mineral oil – methane system.

B.S. Abusahmin et al. / Fuel 197 (2017) 497–517

at this level of reduction in pressure depletion rate, the nature of the oil recovery mechanisms involved has changed. 4.2.3. Depletion tests conducted for the same system at a medium depletion rate To further assess the influence of pressure decline on the percentage oil recovery, the rate of pressure decline was further reduced by a larger margin in this experimental run to examine the trend in the performance. At this medium depletion rate, it took approximately 4 days (precisely 4.2 days) for the pressure at the production end to reduce down to atmospheric pressure. The objective of this experimental run was to confirm that the performance is considerably affected by reducing the pressure depletion rate. Here the pressure was lowered to atmospheric pressure from the system pressure around 500 psi at a depletion rate set at 0.086 psi/min which takes the residence of the fluid in the sand pack over a period of 4 days. The oil recovery and pressure behavior for this scenario is shown in Fig. 7 (a). A comparison of this trend with behavior depicted in Figs. 3 and 6 (a) presents that the behavior of the average pressure of the porous media and the pressure at the production port is quite different. In other words, the ultimate percentage oil recovery is in relation to the rate at which pressure dropped at the production port. As a result it can be concluded that from this experimental run, the nature of the oil recovery mechanisms involved has indeed changed, as the final recovery factor has reduced to 16.8%. The profiles of gas and oil

505

production rates in four days decline test with mineral oil methane system is presented in Fig. 7 (b). The gas rate remained steady during the experimental run whereas the oil rate fluctuates until near the end of the run. 4.2.4. Depletion runs conducted for the same system at a slow depletion rate leading to last the experimental run for over 16 days The experimental procedure and hence production behavior was purposely kept different in this test compared to earlier experimental runs (fast runs). The quite obvious change was in the pressure drawdown that is generated as a result of the continued reduction in the production port pressure. As observed in Fig. 8, the average pressure is very close to pressure at production port all throughout the experimental run. It implies that the draw down pressure remained very low all through the run. In this experimental run, the pressure decline rate was reduced to last the experimental run for over 16 days (depletion rate at 0.021 psi/min). The oil production in this test run was considerably dissimilar from the earlier tests. The obvious change was in the drawdown pressure that is generated as a result of the continued reduction in the production end pressure. As shown in Fig. 8, the average pressure is very close until the rate of gas production becomes high. The gas rate remained steady in the rest of the test. Similarly, there was no significant period of stabilized pressure with continued oil production, as observed in fast depletion rate tests carried in earlier tests.

Fig. 7. Profiles of (a) pressure and recovery factor (b) gas and oil production rates for depletion rate of four days using mineral oil-methane system.

506

B.S. Abusahmin et al. / Fuel 197 (2017) 497–517

Fig. 8. Profiles of pressure and percentage oil recovery factor in mineral oil – methane system in 16 days depletion.

4.2.5. Fast depletion tests starting from lower saturation pressure According to conventional understanding, the solution-gas drive performance improves significantly as the saturation pressure increases (keeping other parameters remained the same) because more drive energy becomes available for oil production. However, the effect of saturation pressure on the performance

Gas Rate (Std cc/min)

16000

(a)

14000 12000

180

10000 8000

120

6000 4000

60

2000 0 500.0

400.0

300.0

200.0

100.0

Average Sand-Pack Pressure (psi)

0 0.0

Oil Production Rate (ml/min)

Oil Produced (mL)

Cumulative Gas (Std. cc)

Cumulative Oil Production (mL)

Cum Gas (Std. cc)

240

under foamy oil flow conditions has not been explored. This run was similar to 1.5 days depletion test discussed earlier, except that the live oil was prepared at 300 psi saturation pressure. The pressure depletion rate used in this run was 0.264 psi/min which was only marginally higher than 0.247 psi/min used in the 1.5 days depletion test. The objective of this test was to evaluate the effect of saturation pressure on the performance of foamy oil solution. The pressure response and the oil recovery factor for this run are presented in Fig. 10. Interistingly, the recovery factor obtained in this run was very similar to that in the 1.5 days depletion that started from 500 psi saturation pressure. There is again a delay in the start of oil production and a high drawdown pressure develops and persists throughout the remainder of the run. The profiles showing the comparison of oil production rates and gas production rates at different saturation pressures are shown in Fig. 11 (a) and (b). The oil production starts and ends earlier in the run at lower saturation pressure, but the rates are similar, whereas the experimental run at 500 psi saturation pressure produces gas at a higher rate. It appears that the main contribution of higher saturation pressure is to increase the total gas production with only a small improvement in the oil production. This is also evident from the plot (producing GOR versus cumulative oil production) shown

0.030

Oil Rate (ml/min)

3.0

(b)

0.025

2.5

0.020

2.0

0.015

1.5

0.010

1.0

0.005

0.5

0.000 500.0

400.0

300.0

200.0

100.0

0.0 0.0

Gas Production Rate (Std. cc/min)

Another significant difference becomes apparent by observing the average sand-pack pressure versus cumulative production curves as seen in Fig. 9 (a). The behavioral pattern is qualitatively very similar to the trend observed in Figs. 4 (a) and 6 (b), except the rate of oil produced marginally declines at 220 psi which eventually leads to crossover of gas production rate at 100 psi. The actual oil and gas production rates in this experimental run are shown in Fig. 9(b). After reaching a peak around 410 psi (average pressure) in the porous media, the oil rate fluctuates and declines, while the gas rate increases gradually. Near the end of this experimental run, when the average pressure in the porous media tends low, the oil production rate instigates to fluctuate more. This is primarily due to the rapidly increasing volume factor of the gas as the average pressure in the porous media approaches atmospheric pressure.

Average Sand-Pack Pressure (psi)

Fig. 9. Profiles of (a) cumulative production of oil and gas (b) oil and gas production rates against average sand-pack pressure in 16 days depletion in mineral oil-methane system.

B.S. Abusahmin et al. / Fuel 197 (2017) 497–517

507

Fig. 10. Pressure and oil recovery factor in fast depletion starting from 300 psi saturation pressure.

pressure depletion rates as shown in Table 3. Owing to the fact that CO2 is more soluble in oil compared to methane and thus permits evaluation of the effect of increased solution GOR without increasing the saturation pressure. The main objective was to assess the influence of increased GOR over the foamy solution gas drive performance. A summary of depletion tests conducted with mineral oil-carbon dioxide system and major outcomes are presented in Table 5 for the four experimental depletions tests. An analogous comparison with results obtained with methane saturated mineral oil (shown in Table 4), shows that the percentage recovery obtained with CO2 are lower compared to those observed with methane at similar pressure depletion rates. The reasons for this deviation and other salient features of these three depletions with CO2 are discussed below in-detail.

Fig. 11. Comparison of (a) oil production rates and (b) gas production rates in runs starting from 300 psi saturation pressure and 500 psi saturation pressure.

in Fig. 12 that at higher saturation pressure has higher GOR during most of oil production. 4.3. Depletion tests using mineral oil–carbon dioxide system To further examine the effect of gas properties on the performance of gas oil recovery, four depletion tests were conducted with carbon dioxide (CO2) saturated mineral oil at different

4.3.1. Depletion test conducted with mineral oil CO2 system at higher pressure depletion rate In view of the higher solution GOR with CO2 and with fast pressure depletion rate, it is expected that this may lead to better oil recovery as observed in the one day depletion with methane. The pressure depletion rate in this run was maintained at 0.434 psi/ min, which was indeed slightly faster than that is used in the one-day depletion with methane. Surprisingly the recovery factor as shown in Fig. 13 obtained with CO2 was lower than that observed with methane which has a value of 26.1%. This may be the fact that drawdown pressure develops as a result of continued pressure reduction at the production end. To evaluate its performance, a similar comparison presented in Fig. 3 shows that the difference between the production end pressure and the average sand pack pressure was slightly higher in the experimental run conducted with methane. With methane gas as solvent, it touched a maximum value of 270 psi pressure soon after the start of bubble nucleation and thus remained at higher value during the rest of the experimental run. The difference between the production end pressure and the average sand-pack pressure is what drives the oil towards the production port. In this test with CO2, the drawdown pressure started low and reduced much faster. Due to the fact that higher solution GOR, more gas is oozed out from solution immediately after bubble nucleation begins. Consequently, it starts to flow as a continuous phase. The cause of considerably lower drawdown with CO2 appears to be due to the fact that collapse of foamy flow at higher volume. The gas production rate becomes high soon after the oil production starts and fluctuates wildly during the run as observed in Fig. 14(a). The producing GOR turn out to be higher than oil

508

B.S. Abusahmin et al. / Fuel 197 (2017) 497–517

GOR @500 psi

GOR @300 psi

Producing GOR (Std. cc/mL)

300

240

180

120

60

0 0

50

100

150

200

250

300

350

400

Cumulative Oil Produced (mL) Fig. 12. Producing GOR versus cumulative volume of oil produced in runs starting from 300 psi saturation pressure and 500 psi saturation pressure.

Table 5 Summary of depletion tests conducted with mineral oil CO2 system. Run #

Saturation Pressure (psi)

Solution GOR

Pressure Depletion Rate (psi/min)

Recovery Factor (%)

6 7 8 9

500 500 500 500

28 28 28 28

0.434 0.220 0.067 0.035

19.6 12.4 12.3 5.5

Fig. 13. Pressure history and oil recovery factor in one-day depletion in mineral oil – CO2 system.

production rate in the early stages and fluctuated widely as sandpack pressure further reduces as shown in Fig. 14(b). These fluctuations may be related to formation and collapse of dispersed gas flow. This again reflects instability in the foamy oil flow structure, which strongly suggests that the foamy oil flow becomes more difficult to maintain when too much gas is released from solution. 4.3.2. Depletion test conducted for same system at pressure depletion rate of 0.22 psi/min In this experimental run with CO2 saturated oil, the pressure depletion rate at the production port was reduced to 0.22 psi/ min, which required two days to reduce the pressure from its initial saturation pressure of 500 psi to near atmospheric pressure.

This change in depletion rate resulted in a dramatic reduction in the recovery factor, which dropped down to 12.4% as shown in Fig. 15. This recovery factor is marginally lower than the recovery factor obtained with methane in 16-days depletion test. It is clear evident that the maximum pressure drawdown that progresses is very small and for most of the run the average pressure virtually overlaps the production port pressure. This lack of high drawdown pressure appears to be the main reason for the low recovery factor. The oil and gas production rates presented in Fig. 16 (a) show that oil rate declines rapidly while the gas rate becomes high early in the depletion and remains high until near the end of the run. The cumulative gas increases linearly with pressure decrease during most of the run, while there is very little production of oil below

509

B.S. Abusahmin et al. / Fuel 197 (2017) 497–517

(a)

Producing GOR

Oil Rate 500

(b) 0.80

400

0.60

300

0.40

200

0.20

100

0.00 500.0

Producing GOR (Std. cc/mL)

Oil Production Rate (ml/min)

1.00

0 400.0

300.0

200.0

100.0

0.0

Average Sand-Pack Pressure (psi) Fig. 14. (a) Oil and gas production rates (b) Oil production rate and producing GOR at very fast depletion rate with mineral oil CO2 system.

Fig. 15. Recovery factor and pressure history of 2-day depletion in mineral oil-CO2 system.

510

B.S. Abusahmin et al. / Fuel 197 (2017) 497–517

250 psi average pressure as shown in Fig. 16 (b). It suggests that the gas mobility becomes very high in the early stage of depletion run and reaches steady state in the later stage.

4.3.3. Depletion test conducted at medium pressure depletion rate for the same system In this run with CO2 saturated oil, the pressure depletion rate at the production port was further decreased to 0.067 psi/min, which takes approximately 5½ days to reduce the pressure from 3447.4 kPa (500 psi) to nearly atmospheric pressure. This change in depletion rate from 0.220 psi/min to 0.067 psi/min resulted in lower recovery factor of 12.38% as shown in Fig. 17 (a). This recovery factor is marginally lower than the recovery factor obtained with methane in 4-days depletion test, which was 16.8%. It is also observed that the highest pressure drawdown that develops is very small and for the whole run the average pressure almost overlaps in tandem. This lack of high drawdown pressure appears to be the main reason for the low recovery factor. The cumulative volumes of produced gas increases more or less linearly with pressure decrease along the test as shown in Fig. 17 (b), and it suggests that the gas mobility becomes high in the early stages of depletion test. Whereas the cumulative oil production for this test was around

185.5 ml that is significantly lower than production of oil (250 ml) for 4-day methane saturated mineral oil depletion test.

4.3.4. Depletion test conducted with mineral oil – CO2 system at slow pressure depletion rate This experimental run in this system was calculatedly kept at slow depletion of 0.035 psi/min, hence the run lasted for twelve days. This depletion rate (0.035 psi/min) is slightly higher than the corresponding depletion rate (0.021 psi/min) used in mineral oil-methane system, which lasted for 16 days. Slow decline test showed a rapid decrease in performance. It is observed that the recovery factor declined dramatically at this slow rate and as expected, resulted in lower recovery factor of around 5% as shown in Fig. 18(a). Furthermore, in this run, the difference between the pressure of the porous media and the pressure at the production port is not much as depicted in Fig. 18(a). In terms of oil production, it was observed that in the early stages of the run, the oil rate is very high and then dropped suddenly, contrary the gas production rate steadily increases as shown in Fig. 18(b). There was a short period of time wherein oil production rate reaches high near the end of run which was resulted from the release of the pressure from the gas collection vessel. The major outcome of this run

(a)

Cumulative Gas

200

45000.0

(b)

40000.0

175

35000.0

150

30000.0

125

25000.0

100

20000.0

75

15000.0

50

10000.0

25

5000.0

0 500.0

Cumulative Gas (Std. cc)

Cumulative Oil Production (mL)

225

Cumulative Oil

0.0 400.0

300.0

200.0

100.0

0.0

Average Pressure in Sand-Pack (psi) Fig. 16. (a) Oil and gas production rates, (b) Cumulative volumes of oil and gas produced versus average sand-pack pressure in two day depletion in mineral oil – CO2 system.

511

B.S. Abusahmin et al. / Fuel 197 (2017) 497–517

(a)

Cumulative Gas

Cumulative Oil

200

Cumualtive OilProduction (mL

180

(b)

160

40000.0

140 120

30000.0

100 80

20000.0

60 40

10000.0

20 0

Cumualtive Gas Production (std.cc/min

50000.0

0.0 500

400

300

200

100

0

Average Sand-Pack Pressure (psi) Fig. 17. (a) Recovery factor and pressure variation (b) Cumulative volumes of oil and gas produced versus average pressure of the porous media for 5½ days depletion using mineral oil saturated carbon dioxide system.

indicates that at high dissolved mineral oil saturated with carbon dioxide leads to a low effective solution gas drive. 4.4. Depletion tests using mineral oil–ethane system To study further the influence of gas properties on the overall performance of oil and gas production, few additional depletion tests were conducted using ethane saturated mineral oil system and the results are summarized in Table 6. In terms of the solubility, ethane is capable of being dissolved with mineral oil at higher stage than carbon dioxide. To keep the solution GOR at reasonable level, ethane saturated mineral oil was saturated at 2068.43 kPa (300 psi). GOR was calculated to be 33 m3/m3 which was slightly higher than the CO2 system (28 m3/m3). These experimental runs in this system at three pressure depletion rate shown in Table 6 was also conducted very similar to earlier mineral oil – methane and mineral oil – CO2 systems. Hence, all the results (shown in other runs) are not presented here, but only the major observations are reported and discussed. Surprisingly high oil recovery factor

was obtained around 23.7% (as shown in Fig. 19) which is a value higher than the fast depletion with CO2 which had resulted in 19.6%. The recovery factor in this depletion test was found to be around 9% lower than the corresponding recovery factor observed in fast depletion with methane, which is 26.1%. The difference between the pressure at the production end and the pressure of the porous media is very similar to comparable runs carried in other similar systems. The second depletion test carried at 0.109 psi/min, had a dramatic effect on solution gas drive performance. The recovery factor (8.5%) obtained in medium depletion run was around one third of the previous test conducted (which had a recovery factor of 23.7%) at fast depletion rate. These results revealed that, in spite of relatively slow pressure decline rate at the production port, foamy oil flow never developed in this test. It again confirms that high solution GOR is not favorable to development of foamy oil flow. The third run conducted using this system was at 0.025 psi/min pressure decline rate. At this slow decline rate, the recovery performance (6.8%) is further deteriorated along with drop in pressure depletion rate.

512

B.S. Abusahmin et al. / Fuel 197 (2017) 497–517

(a)

Gas Rate

Oil Rate 15

Oil Production Rate (ml/min)

(b) 0.04

12

0.03

9

0.02

6

0.01

3

0.00 500.0

Gas Production Rate (Std. cc/min)

0.05

0 400.0

300.0

200.0

100.0

0.0

Average Sand-Pack Pressure (psi) Fig. 18. (a) Recovery factor and pressure variation (b) gas and oil production rates of slow pressure depletion rate in saturated mineral oil – CO2 system.

Table 6 Summary of depletion tests conducted with mineral oil-ethane system. Run #

Saturation Pressure (psi)

Solution GOR

Pressure Depletion Rate (psi/min)

Recovery Factor (%)

10 11 12

300 300 300

33 33 33

0.398 0.109 0.025

23.7 8.5 6.8

4.5. Depletion tests using crude oil–methane system In order to evaluate the effect of change of dead oil on gas oil recovery performance, further four more depletion tests were conducted where crude oil is blended with methane and these tests were carried at depletion rates of 0.350 psi/min, 0.230 psi/min, 0.048 psi/min, and 0.023 psi/min respectively. All these runs started with the porous media saturated completely with live oil at irreducible water saturation and the experimental setup and procedures were the same as those conducted with mineral oil. These runs were also conducted at room temperature of 23 °C. In the very fast depletion test at 0.350 psi/min, the pressure at production port was slowly reduced from initial saturation pressure of 3447.4 kPa (500 psi) to the atmospheric pressure over a

period of approximately 24 h. The pressure profiles in porous media (sand-pack) which were monitored at different locations along the length are shown in Fig. 20, where PT1 represents the production port pressure and PT7 is shut-in end pressure. During the initial stages of the depletion test, the pressure at all locations decreases linearly in tandem along with the pressure at the production end. The pressure gradient remains too small and there is very little oil production during this period. This behavior is consistent with liquid phase expansion under decreasing pressure. As the pressure declines, the dissolved gas supersaturation continues to increase and eventually leads to nucleation of gas bubbles. This occurs at elapsed time of about 6 h, when the pressure has declined to 350 psi. The pressure behavior changes dramatically after the nucleation of gas bubbles. The pressure at the production port

513

B.S. Abusahmin et al. / Fuel 197 (2017) 497–517

Fig. 19. Pressure variation and recovery factor using saturated mineral oil-ethane system at a fast depletion rate.

PT2

PT3

PT4

PT5

PT6

PT7

Cumulative Gas Productionl

500

10000

400

8000

300

6000

200

4000

100

2000

0

Cumulative gas, Std. cc

Pressure, psig

PT1

0 0

2

4

6

8

10

12

14

16

18

20

22

24

Time, hr Fig. 20. Pressure and cumulative gas production in crude oil-ethane system for the fast depletion test.

(PT1) continues to decline linearly but the decline rate is sluggish at all other locations. There is actually a slight rebound in pressure at all intermediate locations, it is being most pronounced at the shut-in end (PT7). This minor rebound may occurred due to fast release of gas from solution and leads to development of a considerable pressure gradient in the porous media. Beyond this point, the flow of oil and gas out of the porous media is driven by this pressure gradient. The oil production starts soon after bubble nucleation and the oil rate remain substantial all through the decline period. The overall cumulative oil production observed in this test was 388 ml, which signifies a recovery factor of around 26%. In the second fast depletion test at the rate of 0.230 psi/min, the product port pressure was decreased from the initial saturation pressure (about 500 psi) to the atmospheric pressure which eventually takes around 1.6 days. This depletion rate (0.230 psi/min) was slower than the earlier test (0.350 psi/min) but is still considered good enough to initiate and maintain foamy oil flow. The pressure gradients developed within the sand-pack in this test are smaller than the previous one but still quite substantial. The overall pressure gradient reaches a peak soon after start of gas release and then declines very slowly during the depletion. The cumulative oil grows relatively steadily with time up to 200 ml of oil production, which matches up to the time at which the gradient pressure starts to decrease. The cumulative oil produced in this test was found to be 376 ml, which corresponds to a recovery

factor of 25.3%. Thus the decrease in pressure decline rate from 0.350 psi/min to 0.23 psi/min had very insignificant impact on the final recovery factor. In the medium rate depletion test at 0.048 psi/min, the pressure at production port was again reduced from the initial saturation pressure of 500 psi to the atmospheric pressure which eventually takes 6.6 days. The bubble nucleation, which can be identified from start of oil production and a minor hump in pressure which occurred at 410 psi as shown in Fig. 21 was found to be due to gas evolution. A trivial pressure difference develops between the shut-in and the production end but they eventually shrunken. There is no significant high pressure gradient is observed along the porous media for rest of the depletion. In this run, the amount of oil produced per unit decline in the average pressure is high initially but declines slowly during most of the depletion period. The amount of gas produced per unit decline in pressure remains nearly constant. The absolute value of the cumulative oil production in was found to be 323 ml, which corresponds to a recovery factor of 22%. In the slow pressure depletion test conducted at depletion rate at 0.023 psi/min, the pressure again decreased steadily to atmospheric pressure over a time horizon of 17 days. There was no significant oil production for the first four days of depletion, while the sand-pack pressure decline to about 420 psi. After that it was noticed that the gas nucleation started and oil production began. It was observed a minor hump in pressure gradient between the

B.S. Abusahmin et al. / Fuel 197 (2017) 497–517

Pressure, psig

Production End Pressure

Mid-Point Pressure

"Far End Pressure"

Cumulative Oil

700

350

600

300

500

250

400

200

300

150

200

100

100

50

0

Cumulative Oil Produced, mL

514

0 0

1

2

3

4

5

6

7

Time, days Fig. 21. Pressure and cumulative oil produced for an intermediate depletion test.

5.1. Effect of oil viscosity Oil viscosity influences the process in many ways and determines the level of pressure gradients that will exist in conjunction with the displacement of oil by growing gas bubbles. It also affects the diffusivity of gas in the oil. Furthermore, it also controls the mobility ratio between oil and gas, which strongly affects the producing gas oil ratio after the gas phase becomes continuous. Tang and Firoozabadi [39] varied oil viscosity by changing the tests temperature and found that in slow depletion tests increasing viscosity of oil actually improved the recovery factor. The depletion tests carried in this research study with different oil viscosities also confirmed the above behavior. The rate of pressure depletion needed to introduce foamy oil flow was faster in lower viscous systems. 5.2. Effect of saturation pressure Fig. 22. Comparative profiles of oil production at various depletion rates.

two ends of the porous media immediately after bubble nucleation, whereas it eventually fades during the rest of depletion run. The total cumulative oil production found in this test was around 240 ml (16% recovery factor). The effect of depletion rates on oil recovery in all four experimental runs carried in this system are presented in Fig. 22. It is deceptive that the oil production starts at higher pressure in slower depletion tests, or in other words, gas nucleation occurs at lower supersaturation in slower tests. It is not unexpected since the nucleation is a stochastic process that requires both residence time and supersaturation. Since higher residence time results in slower depletions, hence nucleation can occur at lower supersaturation. The oil recovery declines with decreasing rate of pressure depletion. It is also confirmed from these tests that higher the pressure depletion rate eventually leads to higher oil recovery.

5. Discussion The impact of various process parameters has been systematically investigated in laboratory-scale solution gas drive experiments conducted in this research study. The effect of each process parameter, their characteristics and major observations derived in the exhaustive list of depletions tests are presented here.

In conventional solution gas drive, the performance of the system is expected to improve with increasing saturation pressure. Higher saturation pressure provides higher drive energy and increases the life of the process. In solution gas drive with foaminess, the effect of saturation pressure appears to be less pronounced. In experimental tests done with the same sand-fluid system using similar depletion rate but different saturation pressures, it was found that there was only a small reduction in the recovery factor when the saturation pressure was reduced from 500 psi to 300 psi. 5.3. Pressure decline analysis at the porous media The pressure decline, which is the distinction between the pressure at the production port and the overall pressure of the porous media that varied along all the decline tests at different depletion rates using various type of gases as solvent. The most important mechanism or parameter in all these tests conducted are the pressure decline and therefore it is so beneficial to observe carefully the amount of oil and gas produced known as the cumulative oil and gas factor. From a mechanistic point of view, the effect of pressure decline would be reduced by the length of drainage path engrossed. In the depletion tests presented in this paper, the length of the porous media is a two meter long, and therefore the average length of the drainage path is found to be approximately one meter. Practical point of view, the drainage area radius varies from a range of 50 to 500 m. Fig. 23 shows a profile of oil recovery factor

B.S. Abusahmin et al. / Fuel 197 (2017) 497–517

Methane

CO2

by using different solution gases. Our side by side experimental runs done with methane, CO2 and ethane have shown that increased solution GOR at same saturation pressure results in reduced recovery factor. At 500 psi saturation pressure, the solution GOR was 9.1 m3/m3 and 28 m3/m3 with methane and CO2 respectively (at standard conditions). In high rate tests using similar decline rates, the recovery factor was 26.1% with methane and around 19.6% with CO2. In slow depletion tests the recovery factor declined from 13.1% with methane to 5.1% with CO2. It appears that the release of larger volume of gas at high solution GOR tends to form a continuous gas phase.

C2H6

30 25

Recovery Factor, %

515

20 15 10 5

5.5. Effect of flow direction in vertical depletions 0 0

5

10

15

20

25

30

Average Drawdown Pressure (psi) Fig. 23. Effect of drawdown pressure on solution gas drive recovery factors using three different solution gases.

against average drawdown pressure. This figure represents an obvious trend of an increase in oil recovery factor with an increasing average drawdown pressure. As a consequence, the pressure gradient that develops in the onsite fields would be comparable with the pressure gradient involved in these tests. Thus the recovery factors anticipated in the field would be similar to those reported in high pressure rate depletions tests. Thus, the data obtained with different gases and varying pressure depletion rate at the production port fits the same trend. As a result, it can be concluded that the recovery factor is a function of the pressure decline that develops in the system as a result of pressure reduction at the production end. 5.4. Effect of solution gas oil ratio (GOR) Li et al. [10] and Tang and Firoozabadi [39] examined the effect of solution GOR in primary depletion tests. The GOR was varied by using different saturation pressures. They found that increased GOR resulted in faster bubble nucleation, greater critical gas saturation and marginally greater recovery in oil. This is consistent with the effect of saturation pressure discussed above. The solution GOR can also be changed without changing the saturation pressure

In vertical depletions the direction of flow can be upward or downward. Intuitively, one would expect the oil recovery to be higher when the flow is downward due to possible contribution of gravity drainage. However, laboratory tests in two meter long sand-packs showed that this is not always the case [40]. It was found that the recovery was totally insensitive to flow direction in fast depletions. In slow depletions, the recovery was indeed higher with downward flow. At intermediate depletion rates, the recovery was significantly higher with flow in upward direction. It appears that in upward flow at intermediate depletion rate, the gravitational forces helped in mobilization of bubbles, which promoted foamy oil flow, while in downward flow the gravitational forces worked against mobilization. In fast displacements, the viscous forces were strong enough to generate foamy flow in both direction and the contribution of gravity was insignificant. In slow depletions, gravity was ineffective in creating foamy flow in upward direction due to very weak viscous forces. In absence of dispersed flow, the gravity drainage provided a small additional recovery. This observation of higher recovery with upward flow at intermediate rates of depletion supports the idea that mobilization of growing bubbles before they form a continuous gas phase is the key to realizing benefits of foamy oil flow. 5.6. Effect of pressure depletion rate Based on the study conducted by Lu et al. [44], it is apparent that the pressure decline rate imposed at the production end has a profound effect on the efficiency of solution gas drive. The

Fig. 24. Effect of pressure depletion rate on cumulative oil recovery in mineral oil-methane system.

516

B.S. Abusahmin et al. / Fuel 197 (2017) 497–517

Fig. 25. Comparison profiles of oil recovery with different solvent gases blended with mineral oil at higher (very fast) depletion rates.

recovery factor declines with decreasing pressure rate in all systems but the extent of change and how quickly this change occurs differs with the gas involved. It is therefore useful to examine this aspect and hence these characteristic were examined in the depletion tests carried in this study. The effect of rate of pressure decline on oil recovery in mineral oil saturated with methane system is depicted in Fig. 24. The depletion starting from 2068.43 kPa (300psi) saturation pressure is also included for comparison. In this system, the reduction in performance with decreasing pressure decline rate is more gradual. There is only a small reduction in cumulative oil by declining to 0.274 psi/min from 0.406 psi/min. Even at the slowest decline rate of 0.021 psi/min, over 200 ml of oil is produced by foamy solution gas drive. It is also interesting that the oil recovery for run starting from 300 psi saturation pressure is similar to that in the run starting at higher saturation pressure at similar depletion rate. The reduction in pressure depletion rate in tests with CO2 saturated oil has observed more dramatic effect, wherein the final cumulative is reduced from 0.434 psi/ min to 0.220 psi/min. Similarly, the effect of pressure decline rate on oil recovery in tests with ethane saturated oil is very similar to that observed with CO2 saturated oil. The performance deteriorates quickly with decreasing pressure decline rate. Both CO2 and ethane give much higher solution GOR and both show faster disappearance of foamy oil flow effects with decreasing pressure decline rates. This strongly suggests that high solution GOR is not good for maintaining dispersed gas flow. It has been noted by other investigators that foam quality involved in foamy oil flow is much lower than aqueous foams. The liquid lamellae involved in foamy oil becomes unstable before draining down to lesser thickness. When the solution GOR is high, the larger volume of released gas tends to drive the foam quality higher. This phenomena further makes the foam very unstable. 5.7. Influence of solution gas The recovery performance of fast decline tests with methane, ethane and CO2 saturated mineral oil systems are presented in Fig. 25. It is of our interest to see the comparison among all of the systems mentioned above in terms of the overall performance at different depletion rates. It emphasizes the influence of higher solubility of CO2 and ethane in slow, medium and fast pressure depletion rates. The decline rates were almost similar to each other and the only difference is the solution gas oil ratio. Methane which

is the least soluble gas, contributes the highest oil recovery factor. Carbon dioxide, that is more soluble than methane and was conducted at the same saturation pressure which is at 3447 kPa (500 psi). It was noticed that CO2 oozes out of the solution at a higher pressure compared to methane. However, oil production rate was inferior to that observed with methane. Ethane is more soluble than carbon dioxide, which provides higher recovery factor than carbon dioxide, when saturated at a pressure of 2068.43 kPa (300 psi). Based on the various depletions tests, it was concluded that the solution gas oil ratio may not be the only factor involved in controlling the performance and the nature of gas involved in it.

6. Conclusions Although the understanding of what is really involved in foamy oil flow has greatly improved over the last decade and it is now accepted as a significant mechanism in production of heavy oil, several issues remain unresolved. The relationship between the interfacial properties of oil and the foamy oil flow is still not well understood. To study and assess the effect of number of parameters that are mainly related to oil recovery trends, as well as the production rates in high viscous reservoirs under foamy solution gas drive behavior, several experimental depletion tests were carried. The prominent factors that are studied comprised of refined mineral oil versus crude oil, saturation pressure, drawdown pressure, pressure depletion rate and gas oil ratio. Gas oil ratio was obtained at different saturation pressures by having a mixture of dead oil blended with gases such as carbon dioxide, ethane and methane to form what so called live oil. Each high viscous oil type/system was completely characterized by evaluating different parameters namely live oil viscosity, oil compressibility and gas oil ratio. The major outcome of the depletion tests carried in this research study is that the decreasing pressure depletion rate resulted in lower overall performance in all these tests. At the same rate of pressure depletion, higher oil recovery was obtained with methane saturated oil compared to carbon dioxide saturated oil. Increased solution gas-oil-ratio does not affect the performance of foamy solution gas. In addition it was noticed that the oil recovery factor did not decrease significantly when the saturation pressure was decreased from 500 psi to 300 psi. At 3447 kPa, (500 psi) saturation pressure, the solution GOR was 9.1 m3/m3, 28 m3/m3

B.S. Abusahmin et al. / Fuel 197 (2017) 497–517

and 33 m3/m3 with methane, CO2 and ethane gas respectively (at standard conditions), whereas solution GOR of methane saturated with crude oil were found to be 11 m3/m3. The degree of supersaturation depends on the depletion rate and it’s inversely proportional to the depletion rate i.e. it declines as the depletion rate increases. Under similar tests conditions, the cumulative oil production behavior observed with mineral oil and crude oil was essentially the same. Both mineral and crude oil systems displayed similar decline in the oil recovery performance with decreasing pressure depletion rate. In high depletion rate tests using similar decline rates, the recovery factor was 26.1% with methane and only 19.6% with CO2. In slow depletion tests the recovery factor declined from 13.1% with methane to 5.1% with CO2. The outcomes of primary depletion tests also indicated that ethane saturated oil provided the lowest recovery, even though it had the highest solution GOR whereas methane saturated oil (which had the lowest solution GOR of 9.1 m3/m3) delivers the highest recovery factor. Acknowledgements The authors acknowledge the Libya education funding, research facilitator at the University of Calgary and Shell Canada for providing samples for conducting the research laboratory experiments. References [1] Sheng J, Maini B, Hayes R, Tortike W. A non-equilibrium model to calculate foamy oil properties. J Can Pet Technol 1999;38:38–45. [2] Maini BB. Foamy-oil flow. J Petrol Technol 2001;53:54–64. [3] Tang G-Q, Leung T, Castanier LM, Sahni A, Gadelle F, Kumar M, et al. An investigation of the effect of oil composition on heavy oil solution-gas drive. In: SPE Annual Technical Conference and Exhibition. Society of Petroleum Engineers; 2003. [4] Liu P, Wu Y, Li X. Experimental study on the stability of the foamy oil in developing heavy oil reservoirs. Fuel 2013;111:12–9. [5] Zhang Y-Y, Sun X-F, Duan X-W, Li X-M. Diffusion coefficients of natural gas in foamy oil systems under high pressures. Petrol Sci 2015;12:293–303. [6] Zhou X, Zeng F, Zhang L, Wang H. Foamy oil flow in heavy oil–solvent systems tested by pressure depletion in a sandpack. Fuel 2016;171:210–23. [7] Lu T, Li Z, Li S, Wang P, Wang Z, Liu S. Enhanced heavy oil recovery after solution gas drive by water flooding. J Petrol Sci Eng 2016;137:113–24. [8] Turta A, Maini B, Jackson C. Mobility of gas-in-oil dispersions in enhanced solution gas drive (foamy oil) exploitation of heavy oil reservoirs. J Can Pet Technol 2003;42:48–55. [9] Wang R, Chen Z, Qin J, Zhao M. Performance of drainage experiments with Orinoco belt heavy oil in a long laboratory core in simulated reservoir conditions. Soc Petrol Eng 2008;13:474–9. [10] Li S, Li Z, Wang Z. Experimental study on the performance of foamy oil flow under different solution gas–oil ratios. RSC Adv 2015;5:66797–806. [11] Sun X, Dong M, Zhang Y, Maini BB. Enhanced heavy oil recovery in thin reservoirs using foamy oil-assisted methane huff-n-puff method. Fuel 2015;159:962–73. [12] Maini BB, Busahmin B. Foamy oil flow and its role in heavy oil production. In: Porous Media and Its Applications in Science, Engineering, and Industry: 3rd International Conference. AIP Publishing; 2010. p. 103–8. [13] Chen ZJ, Sun J, Wang R, Wu X. A pseudobubblepoint model and its simulation for foamy oil in porous media. Soc Petrol Eng 2015;20:239–47. [14] Sheng J, Maini B, Hayes R, Tortike W. Experimental study of foamy oil stability. J Can Pet Technol 1997;36:31–7. [15] Ostos A, Maini B. An integrated experimental study of foamy oil flow during solution gas drive. J Can Pet Technol 2005;44:43–50.

517

[16] Shahvali M, Pooladi-Darvish M. Dynamic modelling of solution-gas drive in heavy oils. J Can Pet Technol 2009;48:39–46. [17] Bondino I, McDougall SR, Hamon G. Pore-scale modelling of the effect of viscous pressure gradients during heavy oil depletion experiments. J Can Pet Technol 2011;50:45–55. [18] Sheng J. Enhanced oil recovery field case studies. Gulf Professional Publishing; 2013. [19] Liu P, Li W, Hao M, Wu Y, Li X. Quantitative evaluation of factors affecting foamy oil recovery in the development of heavy hydrocarbon reservoirs. Int J Hydrogen Energy 2016;41:15624–31. [20] Karri RR, Srinivasan T, Venkateswarlu C. Mathematical and kinetic modeling of biofilm reactor based on ant colony optimization. Process Biochem 2010;45:961–72. [21] Busahmin BS, Maini BB. Effect of solution-gas-oil-ratio on performance of solution gas drive in foamy heavy oil systems. In: Canadian Unconventional Resources and International Petroleum Conference. Society of Petroleum Engineers; 2010. [22] Bondino I, McDougall S, Hamon G. A pore-scale modelling approach to the interpretation of heavy oil pressure depletion experiments. J Petrol Sci Eng 2009;65:14–22. [23] Kumar R. Solution-gas drive in heavy oil: gas mobility and kinetics of bubble growth [Dissertation]. Canada: University of Calgary; 1999. [24] Firoozabadi A. Mechanisms of solution gas drive in heavy oil reservoirs. J Can Pet Technol 2001;40:15–20. [25] Sun X, Zhang Y, Cui G, Duan X, Zhao C. Feasibility study of enhanced foamy oil recovery of the Orinoco Belt using natural gas. J Petrol Sci Eng 2014;122:94–107. [26] Li S, Li Z, Lu T, Li B. Experimental study on foamy oil flow in porous media with Orinoco Belt heavy oil. Energy Fuels 2012;26:6332–42. [27] Rose W. Myths about later-day extensions of Darcy’s law. J Petrol Sci Eng 2000;26:187–98. [28] Bera A, Babadagli T. Relative permeability of foamy oil for different types of dissolved gases. SPE Reservoir Eval Eng 2016. [29] Brito A, Cabello R, Guzmán N, Marcano L, Márquez J, Trujillo J. Hydrodynamic study of multiphase flow transport of highly viscous foamy fluids. J Petrol Sci Eng 2015;135:367–74. [30] Busahmin B, Maini B, Hejazi H, Sharifi A, Tavallali M. Application of foamy mineral oil flow under solution gas drive to a field crude oil. Int J Contemp Energy 2016;2. [31] Sheng J, Maini B, Hayes R, Tortike W. Critical review of foamy oil flow. Transp Porous Media 1999;35:157–87. [32] Smith GE. Fluid flow and sand production in heavy-oil reservoirs under solution-gas drive. SPE Prod Eng 1988;3:169–80. [33] Maurice B. Heavy-oil production enhancement by encouraging sand production. Soc Petrol Eng 2000;59276. [34] Wong R, Maini B. Gas bubble growth in heavy oil-filled sand packs under undrained unloading. J Petrol Sci Eng 2007;55:259–70. [35] Dai L, Zhang Y. Experimental study of oil–water two-phase flow in a capillary model. J Petrol Sci Eng 2013;108:96–106. [36] Karri RR, Wang X, Gerritsen H. Ensemble based prediction of water levels and residual currents in Singapore regional waters for operational forecasting. Environ Model Software 2014;54:24–38. [37] Akin S, Kovscek AR. Heavy-oil solution gas drive: a laboratory study. J Petrol Sci Eng 2002;35:33–48. [38] Karri RR. Evaluating and estimating the complex dynamic phenomena in nonlinear chemical systems. Int J Chem Reactor Eng 2011;9. [39] Tang G, Firoozabadi A. Effect of GOR, temperature, and initial water saturation on solution-gas drive in heavy-oil reservoirs. Soc Petrol Eng 2005;10:34–43. [40] Ostos AN. Effect of capillary number on performance of solution gas drive in heavy oil reservoirs [Dissertation]. Canada: University of Calgary; 2003. [41] Xu S. Investigation of mechanisms involved in generation of foamy oil flow [Dissertation]. Canada: University of Calgary; 2007. [42] Karri RR, Damaraju PR, Chimmiri V. Soft sensor based nonlinear control of a chaotic reactor. IFAC Proc Vol 2009;42:537–43. [43] Lago M, Huerta M, Gomes R. Visualization study during depletion experiments of Venezuelan heavy oils using glass micromodels. J Can Pet Technol 2002;41. [44] Lu T, Li Z, Li S, Li B, Liu S. Performances of different recovery methods for Orinoco Belt heavy oil after solution gas drive. Energy Fuels 2013;27:3499–507.