Oil recovery mechanisms and asphaltene precipitation phenomenon in immiscible and miscible CO2 flooding processes

Fuel 109 (2013) 157–166

Contents lists available at SciVerse ScienceDirect

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

Oil recovery mechanisms and asphaltene precipitation phenomenon in immiscible and miscible CO2 flooding processes Meng Cao, Yongan Gu ⇑ Petroleum Technology Research Centre (PTRC), Petroleum Systems Engineering, Faculty of Engineering and Applied Science, University of Regina, Regina, Saskatchewan, Canada S4S 0A2

h i g h l i g h t s " Three different CO2 flooding processes were studied. " CO2-based oil recovery mechanisms were analyzed. " CO2-induced asphaltene precipitation phenomenon was examined. " Temperature effect on CO2-based oil recovery was investigated.

a r t i c l e

i n f o

Article history: Received 11 October 2012 Received in revised form 7 January 2013 Accepted 8 January 2013 Available online 29 January 2013 Keywords: CO2-EOR Light crude oil–CO2 system Immiscible and miscible CO2 flooding Asphaltene precipitation and deposition Reservoir permeability reduction

a b s t r a c t In this paper, oil recovery mechanisms and asphaltene precipitation phenomenon of immiscible and miscible CO2 flooding processes in the tight sandstone reservoir core plugs are experimentally studied. First, the vanishing interfacial tension (VIT) technique is applied to determine the minimum miscibility pressure (MMP) between the original light crude oil and CO2. Second, a total of five coreflood tests are performed at the actual reservoir temperature. It is found that the oil recovery factor (RF) increases monotonically as the injection pressure increases during the immiscible CO2 flooding. The increased oil RF is attributed to the increased CO2 solubility in oil, reduced oil viscosity, increased CO2 viscosity, and reduced equilibrium interfacial tension (IFT) of the light crude oil–CO2 system. Once the injection pressure exceeds the MMP, the oil RF increases slightly and will reach an almost constant maximum value. In this case, it is the multi-contact miscibility (MCM) and the extremely low equilibrium IFT that jointly make the ultimate oil RF high and unchanged. Moreover, the oil RF after CO2 breakthrough (BT) increases slightly in the immiscible case but substantially in the miscible case. This is due to the strong light-hydrocarbons extraction by CO2 and the miscible displacement in the second case. On the other hand, the measured average asphaltene content of the produced oil and the measured oil effective permeability reduction are found to be higher in the immiscible flooding process. They both reach some lower values in the miscible case. Finally, the experimental results obtained in this study at the actual reservoir temperature of Tres = 53.0 °C are compared with those attained previously at the laboratory temperature of Tlab = 27.0 °C to examine the temperature effect on the CO2-based oil recovery process. Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction Enhanced oil recovery (EOR) processes have become increasingly important to the petroleum industry. After the primary and secondary oil recovery, a typical residual oil saturation in a light or medium oil reservoir is still in the range of 50–60% of the original-oil-in-place (OOIP) [1]. Carbon dioxide (CO2) tertiary or even secondary oil recovery can be applied to further exploit the conventional oil reserves. After the secondary water flooding, many light and medium oil reservoirs are suitable for the miscible or even ⇑ Corresponding author. Tel.: +1 306 585 4630; fax: +1 306 585 4855. E-mail address: [email protected] (Y. Gu). 0016-2361/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.fuel.2013.01.018

immiscible CO2 flooding. It is worthwhile to note that CO2 flooding can not only effectively recover the residual oil but also considerably reduce greenhouse gas emission [2]. Since the 1950s, a number of laboratory [3], field [4], and numerical [5] studies have shown that CO2 can be an efficient oil-dissolving and displacing solvent in the reservoir formations. It has been found from a laboratory study that the ultimate oil recovery factor of a CO2 secondary flood is over 60% of the OOIP, which is significantly higher than the average oil recovery factor of 44% of the OOIP for a secondary water flood [6]. On the other hand, a miscible CO2 tertiary flood can recover 8–25% of the OOIP in the field-scale pilot tests [7,8]. In general, the injected CO2 cannot achieve the first-contact miscibility (FCM) with the reservoir crude oil. However, it can

158

M. Cao, Y. Gu / Fuel 109 (2013) 157–166

gradually develop the so-called dynamic miscibility with the residual oil through the multiple contacts under the actual reservoir conditions [9], which is also referred to as the multi-contact miscibility (MCM). Miscible CO2 flooding has been proven to be an effective and economical means of enhancing or improving the oil recovery. The foremost technical issue in optimization or design of a CO2 flooding project is to determine the minimum miscibility pressure (MMP) between the crude oil and CO2 at the reservoir temperature [10]. The MMP of a given crude oil–CO2 system is defined as the minimum operating pressure at which CO2 can reach the dynamic or multi-contact miscibility with the crude oil [9]. Among all the existing experimental methods, the slim-tube test is most commonly used for the determination of the MMP [11,12]. The rising-bubble apparatus (RBA) is recognized as a faster and visual alternative to the slim-tube test [13,14]. More recently, a new experimental approach named the vanishing interfacial tension (VIT) technique has been utilized to determine the miscibility conditions of different crude oil–CO2 systems [15,16]. The VIT technique is based on the concept that the interfacial tension (IFT) between the crude oil and solvent (e.g., CO2) involved must approach zero when these two phases become miscible. Its applicability, reliability, and accuracy have been discussed in the literature [15–17]. The commonly-recognized oil recovery mechanisms for CO2 flooding include the oil viscosity reduction, oil-swelling effect, the IFT reduction, light-hydrocarbons extraction, immiscible and miscible displacements [3,18,19]. These mechanisms can play more or less important roles, mainly depending on whether the CO2 displacement is immiscible or miscible. For example, the oil viscosity reduction, IFT reduction, and immiscible displacement are more important oil recovery mechanisms for the immiscible CO2 flooding process, whereas the oil-swelling effect, light-hydrocarbons extraction, and miscible displacement play more important roles in the miscible CO2 flooding process [20]. After CO2 is injected into an oil reservoir, it contacts and interacts with the reservoir oil and thus changes the reservoir equilibrium conditions and fluid properties, which may lead to the precipitation of the heavy organic solids, primarily asphaltenes [21]. Asphaltenes are the heaviest and most complicated fraction in a crude oil sample and consist of condensed polynuclear aromatics, small amounts of heteroatoms (e.g., S, N, and O), and some traces of metal elements (e.g., nickel and vanadium). They are generally characterized as insoluble materials in a low boiling-point normal alkane (e.g., n-pentane or n-heptane) [22]. Asphaltene precipitation may cause some serious production problems, such as reservoir permeability reduction, wettability alteration, and formation damage. In particular, asphaltene precipitation problems are more common and serious in undersaturated tight light oil reservoirs, though generally the light crude oil has an extremely low asphaltene content [23]. For example, production of Hass–Messaoud light crude oil in Algeria with asphaltene content of only 0.15 wt.% causes numerous operating problems due to asphaltene precipitation and deposition onto the reservoir formation, downhole wellbore, and surface facilities [24]. This is because of small pore throats in the tight light oil reservoir and a significant reduction of the resin–asphaltene ratio in the light crude oil as a result of substantial CO2 dissolution in it [25]. There are many tight sandstone formations in Canada, which contain a tremendous amount of the residual light crude oil of superior quality. There is a huge potential to apply various CO2based oil recovery processes to further exploit the conventional oil reserves from such unconventional resources. At present, some oil producers are seriously considering CO2 secondary or tertiary flooding process to recover the light crude oil from these tight sandstone formations. However, it is neither well studied nor fully understood what major oil recovery mechanisms are involved and

how asphaltene precipitation affects the CO2-based oil recovery processes in a tight light oil reservoir under the actual reservoir conditions. In this paper, first, the equilibrium IFTs between the light crude oil and CO2 are measured at fifteen different equilibrium pressures and the actual reservoir temperature. The MMP is determined by applying the VIT technique. Then a series of five CO2 coreflood tests are undertaken at five different injection pressures by using composite sandstone reservoir core plugs under the immiscible, near-miscible, and miscible conditions. The oil recovery factor, producing gas–oil ratio (GOR), and asphaltene content of CO2-produced oil are measured at a different PV of injected CO2 in each coreflood test. Furthermore, the total oil recovery factor, average asphaltene content of CO2-produced oil, and oil effective permeability reduction are measured and compared at five different injection pressures to study the CO2-based oil recovery mechanisms and CO2-induced asphaltene precipitation phenomenon in a tight light oil reservoir. 2. Experimental section 2.1. Materials The original light crude oil was collected from the Pembina Cardium oilfield in Alberta, Canada. The obtained original crude oil was cleaned by using a centrifuge (Allegra X-30 Series, Beckman Coulter, USA) to remove any sands and brine. The density and viscosity of the cleaned light crude oil were measured to be qoil ¼ 842:9 kg=m3 and loil ¼ 7:92 cP at the atmospheric pressure and T = 22.0 °C, respectively. The asphaltene content of the cleaned light crude oil was measured to be wasp = 0.26 wt.% (n-pentane insoluble) by using the standard ASTM D2007-03 method [26] and filter papers (Whatman No. 5, England) with a pore size of 2.5 lm. The compositional analysis result of the light crude oil was obtained by using the standard ASTM D86 [27] and is given in Table 1. The reservoir brine samples were collected from the same oilfield, cleaned, and analyzed. Its detailed physicochemical properties are listed in Table 2. A number of tight sandstone reser-

Table 1 Compositional analysis result of Pembina Cardium original light crude oil (qoil = 842.9 kg/m3 and loil = 7.92 cP at the atmospheric pressure and T = 22.0 °C) with the asphaltene content of wasp = 0.26 wt.% (n-pentane insoluble). Carbon no.

mol.%

Carbon no.

mol.%

C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 C15 C16 C17 C18 C19 C20 C21 C22 C23 C24 C25 C26

0.00 0.00 0.20 1.17 3.67 5.01 10.67 7.20 7.61 6.95 5.75 5.01 4.59 3.97 3.74 2.98 3.08 3.09 2.01 2.07 1.96 1.14 1.55 1.28 1.27 1.14

C27 C28 C29 C30 C31 C32 C33 C34 C35 C36 C37 C38 C39 C40 C41 C42 C43 C44 C45 C46 C47 C48 C49 C50+ Total

1.07 0.94 0.89 0.64 0.68 0.59 0.46 0.38 0.54 0.47 0.30 0.27 0.37 0.28 0.27 0.22 0.22 0.20 0.20 0.15 0.14 0.13 0.13 3.35 100.00

159

M. Cao, Y. Gu / Fuel 109 (2013) 157–166 Table 2 Physical and chemical properties of the cleaned Pembina Cardium reservoir brine at P = 1 atm. Temperature (°C) Density (g/cc) Viscosity (mPa s) pH at 20.0 °C Specific conductivity (lS cm1) Refractive index at 25.0 °C Chloride (mg/L) Sulfate (mg/L) Total dissolved solids (mg/L) Potassium (mg/L) Sodium (mg/L) Calcium (mg/L) Magnesium (mg/L) Iron (mg/L) Manganese (mg/L) Barium (mg/L)

15 1.003 1.17 8.08 7250 1.3339 1130 4.0 4323 at 180 °C 17 1690 17 11 0.021 <0.001 9.20

20 1.002 1.02

40 0.996 0.67

was rinsed with n-pentane until the precipitant remained colorless after it passed through the filter papers. The precipitated asphaltenes with the filter papers were slowly dried at T = 100.0 °C in an oven (OF-12G, JEIO TECH Ltd., Korea) until their weight did not change from the reading of the electric balance. With the measured weight change of the filter papers before and after filtration, the asphaltene content of the oil samples was determined accordingly. 2.3. IFT measurement

voir core plugs were collected from several wells located in the Pembina Cardium oilfield at the reservoir depths of 1600– 1648 m. The purity of carbon dioxide (Praxair, Canada) used in this study is equal to 99.998 mol.% and the purity of n-pentane (VWR International, Canada) is equal to 99.76 mol.%.

2.2. Asphaltene content measurement The asphaltenes were precipitated from the original crude oil or the produced oil samples and measured by using the standard ASTM D2007-03 method. More specifically, one volume of the oil sample was mixed with 40 volumes of liquid n-pentane, which was used as a precipitant. The crude oil–precipitant mixture was agitated by using a magnetic stirrer (SP46925, Barnstead/Thermolyne Corporation, USA) for 24 h. Filter papers with pore size of 2.5 lm were weighed by using a high-precision electric balance (PR 210, Cole–Parmer, Canada) with an accuracy of 1 mg before they were used to filter the crude oil–precipitant mixture. The filter cake, which was primarily composed of precipitated asphaltenes,

Fig. 1 shows the schematic diagram of the experimental setup used for measuring the equilibrium IFT between the light crude oil and CO2 by applying the axisymmetric drop shape analysis (ADSA) technique for the pendant drop case [28]. The major component of this experimental set-up was a see-through windowed high-pressure IFT cell (IFT-10, Temco, USA). A stainless steel syringe needle was installed at the top of the IFT cell and used to form a pendant oil drop. The light crude oil and CO2 were stored in two transfer cylinders (500-10-P-316-2, DBR, Canada) and heated to the specified reservoir temperature of Tres = 53.0 °C. Afterwards, the light crude oil was introduced from the transfer cylinder to the syringe needle by using a programmable syringe pump (100DX, ISCO Inc., USA). A light source and a glass diffuser were used to provide uniform illumination for the pendant oil drop. A microscope camera was used to capture the sequential digital images of the dynamic pendant oil drop surrounded by CO2 inside the IFT cell at different times. The high-pressure IFT cell was positioned horizontally between the light source and microscope camera. The entire ADSA system and IFT cell were placed on a vibration-free table (RS4000, Newport, USA). The digital image of the dynamic pendant oil drop surrounded by CO2 at any time was acquired in a tagged image file format (TIFF) by using a digital frame grabber (Ultra II, Coreco Imaging, Canada) and stored in a DELL personal computer. The high-pressure IFT cell was first filled with CO2 at a pre-specified pressure and a constant temperature. After the pressure and temperature inside the IFT cell reached their stable values, the light crude oil was introduced from the original light crude oil sample

Hydraulic oil

CO2

Crude oil

Positive displacement pump

Syringe pump P

Pendant Light source

Temperature controller

oil drop

High-pressure IFT cell

Microscope & camera Personal computer

Vibration-free table

Fig. 1. Schematic diagram of the experimental set-up used for measuring the equilibrium interfacial tension (IFT) between the light crude oil and CO2 by applying the axisymmetric drop shape analysis (ADSA) technique for the pendant drop case.

160

M. Cao, Y. Gu / Fuel 109 (2013) 157–166

the sandstone reservoir core plugs were cleaned by using a Dean–Stark extractor (09-556D, Fisher Scientific, Canada) for 4– 7 days. An automatic displacement pump (PMP-1000-1-10-MB, DBR, Canada) was used to displace the crude oil, reservoir brine or CO2 through the composite reservoir core plugs inside a coreholder (RCHR-2.0, Temco, USA). The tap water was pumped by using the syringe pump to apply the so-called overburden pressure, which was always kept 3–5 MPa higher than the inlet pressure (i.e., injection or test pressure) of the coreholder. The composite reservoir core plugs used in the five CO2 coreflood tests were L = 8–10 in. long and D = 2 in. in diameter. Four high-pressure cylinders (500-10-P-316-2, DBR, Canada) were used to store and deliver the crude oil, reservoir brine, CO2, and tap water, respectively. These four transfer cylinders and the high-pressure coreholder were placed inside an air bath. A thermocouple heating gun (HG 1100, Thankita, USA) and a temperature controller (Standard-89000-00, Cole–Parmer, Canada) were used to heat the air bath and keep it at the constant reservoir temperature of Tres = 53.0 °C. A back-pressure regulator (BPR-50, Temco, USA) was used to maintain the pre-specified injection pressure during each CO2 flooding test and the BPR pressure was always kept 0.5–1.0 MPa lower than the injection pressure. During the reservoir brine, original light crude oil, and CO2 injection processes, the differential pressure between the inlet and outlet of the coreholder was measured by using a digital differential pressure

cylinder to the IFT cell to form a pendant oil drop at the tip of the syringe needle. Once a well-shaped pendant oil drop was formed and surrounded by CO2, the sequential digital images of the dynamic pendant oil drop at different times were acquired and stored automatically in the personal computer. Then the ADSA program for the pendant drop case was executed to determine the dynamic IFT between the dynamic pendant oil drop and CO2 phase at any time. The IFT measurement was repeated for at least three different pendant oil drops to ensure a satisfactory repeatability of ±0.05 mJ/m2 at each pre-specified pressure and constant temperature. In this study, the crude oil–CO2 dynamic and equilibrium IFTs were measured at a constant reservoir temperature of Tres = 53.0 °C and fifteen different equilibrium pressures in the range of Peq = 1.7–19.1 MPa. Only the average value of the equilibrium IFTs of three repeated IFT measurements at each equilibrium pressure and the constant reservoir temperature was noted and is presented in this paper. Then the vanishing interfacial tension (VIT) technique was applied to determine the MMP of the light crude oil– CO2 system from the measured equilibrium IFT versus equilibrium pressure data. 2.4. Coreflood test A schematic diagram of the high-pressure coreflood apparatus used in CO2 coreflood tests is shown in Fig. 2. Prior to each test, Fan

Fan

Air bath

Crude oil

CO2 Automatic pump Brine Tap water Differential pressure transducer Thermocouple

P Coreholder P1

Temperature controller

High-pressure coreholder P2

Heating gun

Syringe pump

P

Back-pressure regulator

Personal computer Syringe pump Gas flow metre Gas sample cylinder

To atmosphere

Oil sample collector Fig. 2. Schematic diagram of the high-pressure CO2 coreflood apparatus.

161

M. Cao, Y. Gu / Fuel 109 (2013) 157–166

indicator (Type PM, Heise, USA). The cumulative amount of gases released from the produced oil and during the CO2 breakthrough was measured by using a gas flow meter (GFM 17, Aalborg, USA). The differential pressure data and cumulative amounts of produced oil and gases were measured, recorded, and stored automatically in a personal computer at a preset time interval of 10 s. The general procedure for preparing each CO2 coreflood test is briefly described as follows. The sandstone reservoir core plugs were placed in series inside the Dean–Stark extractor and cleaned with toluene, methanol, and chloroform in sequence to remove hydrocarbons, salts, and clays, respectively. After the sandstone reservoir core plugs were cleaned and dried, they were assembled in series in the horizontal coreholder and vacuumed for 24 h. Then, the cleaned reservoir brine was injected to measure the porosity of the composite reservoir core plugs. Afterward, the cleaned reservoir brine was injected at different flow rates (qw ¼ 0:1—0:5 cc=min) to measure the absolute permeability of the composite reservoir core plugs. As given in Table 3, the measured porosity was in the range of / = 12.7–16.1% and the measured absolute permeability was in the range of k = 0.8–1.7 mD. In a CO2 secondary oil recovery process (Tests #1–5), precisely speaking, which is CO2 secondary flooding at the initial connate water saturation, the original light crude oil was injected through the brine-saturated composite core plugs until no more water was produced and the connate water saturation was achieved. It is worthwhile to note that in this study, the composite core plugs were saturated first with the reservoir brine and then with the original crude oil to finally reach the so-called connate water saturation and initial oil saturation at the temperature of T = 22.0 °C. In this way, a high initial oil saturation was obtained to model the actual reservoir case. The connate water saturation was found to be Swc = 23.3–39.8% and the initial oil saturation was in the range of Soi = 60.2–76.7%. After the connate water saturation and initial oil saturation were achieved, the entire coreflood apparatus was placed inside the airbath, where it was heated and maintained at the constant reservoir temperature of Tres = 53.0 °C for at least two days. Then, a total of 3.0 PV of the original light crude oil was injected initially at qoil = 0.1 cc/min to pressurize the composite core plugs and ensure that the pre-specified injection pressure was reached and that a stable differential pressure between the inlet and outlet of the coreholder was achieved. During CO2 flooding, a constant volume injection rate of CO2, qCO2 ¼ 0:4 cc=min, was used at each pre-specified injection pressure and Tres = 53.0 °C. CO2 injection was terminated after a total of 2.0 PV was injected and no more oil was produced. A digital video camera was used to record the cumulative volume of the produced oil. The produced oil sample was first visually examined and then centrifuged to separate the oil and water phases. No connate water was produced and found in the visual examination and centrifuging process of the produced oil sample in any CO2 secondary oil recovery test. The cumulative volume of the produced gas was measured and re-

corded by using the gas flow meter. After each CO2 flooding test, a total of 3.0 PV of the original light crude oil was re-injected at qoil ¼ 0:1 cc=min to re-pressurize the composite core plugs and ensure that another stable differential pressure across the coreholder was established. In this work, the respective oil effective permeabilities were calculated from the measured stable differential pressures between the inlet and outlet of the coreholder during the initial original light crude oil injection before CO2 flooding and during its final reinjection after CO2 flooding at qoil ¼ 0:1 cc=min and Tres = 53.0 °C. It was found that the measured respective differential pressures before and after CO2 flooding could reach their stable values during the initial oil injection (DP1) and during the final oil reinjection (DP2) after a total of 3.0 PV of the original light crude oil was injected, which are listed in Table 3. The differential pressure measured during the final oil reinjection after CO2 flooding was always higher than that measured during the initial oil injection before CO2 flooding, i.e., DP2 > DP1. They were then used to calculate the oil effective permeabilities by applying Darcy’s law in Eqs. (1) and (2), respectively. Lastly, the oil effective permeability reduction in percentage for each CO2-coreflood test was obtained from Eq. (3) and is given in Table 3:

ko1 ¼

qoil lo L ; ADP1

ð1Þ

ko2 ¼

qoil lo L ; A DP 2

ð2Þ

  Dko ko1  ko2 DP 1  100%: ¼ ¼ 1 ko ko1 DP 2

ð3Þ

In Eqs. (1)–(3), ko1 and ko2 represent the oil effective permeabilities before and after CO2 flooding; qoil is the constant volume injection rate and lo is the viscosity of the original light crude oil; A is the cross-sectional area and L is the length of the composite core plugs. 3. Results and discussion 3.1. Equilibrium IFT and MMP In this study, the measured equilibrium IFTs between the light crude oil and CO2 at 15 different equilibrium pressures of Peq = 1.7–19.1 MPa and a constant reservoir temperature of Tres = 53.0 °C are plotted in Fig. 3. It is found that the measured equilibrium IFT is reduced almost linearly with the equilibrium pressure in two distinct pressure ranges: Range I (Peq = 1.7– 8.3 MPa) and Range II (Peq = 8.3–19.1 MPa). In Range I, the equilibrium IFT reduction is solely attributed to the increased solubility of CO2 in the original light crude oil at an increased equilibrium pres-

Table 3 The physical properties of the composite sandstone reservoir core plugs, experimental conditions, total oil recovery factors and oil effective permeability reduction data for five coreflood tests at the actual reservoir temperature of Tres = 53.0 °C. Test no.

Pinj (MPa)

/ (%)

k (mD)

Soi (%)

Swc (%)

CO2 RF (%)

wasp (wt.%)

DP1 (kPa)

DP2 (kPa)

Dko/ko

1 2 3 4 5

7.2 9.2 10.4 12.1 14.0

16.1 15.9 14.1 12.7 13.0

1.7 1.0 0.8 1.5 1.7

63.9 70.7 60.2 76.7 70.5

36.1 29.3 39.8 23.3 29.5

63.1 69.0 81.0 85.3 87.0

0.19 0.20 0.11 0.12 0.14

1,911.3 4,403.1 610.0 1,314.1 404.8

2,643.7 5,230.2 708.3 1,471.3 452.9

27.70 15.81 13.88 10.68 10.62

Notes: Pinj: CO2 injection pressure; /: porosity of the composite sandstone reservoir core plugs; k: absolute permeability of the composite sandstone reservoir core plugs; Soi: initial oil saturation; Swc: initial connate water saturation; CO2 RF: total oil recovery factor at 2.0 PV of injected CO2; wasp: average asphaltene content of CO2-produced oil (npentane insoluble); DP1: stable differential pressure between the inlet and outlet of the coreholder during the initial original light crude oil injection before CO2 flooding; DP2: stable differential pressure between the inlet and outlet of the coreholder during the final original light crude oil reinjection after CO2 flooding; Dko/ko: oil effective permeability reduction in percentage after CO2 flooding.

162

M. Cao, Y. Gu / Fuel 109 (2013) 157–166

Fig. 3. Measured equilibrium interfacial tensions (IFTs) of the Pembina Cardium light crude dead oil–pure CO2 system at different equilibrium pressures and Tres = 53.0 °C.

sure. In Range II, the pendant oil drop that was finally formed at the tip of the syringe needle was mainly composed of relatively heavy hydrocarbons of the original light crude oil after the initial quick and subsequent slow light-hydrocarbons extraction [29]. The measured equilibrium IFT was reduced marginally and ultimately it reached its lowest value of ceq ¼ 1:12 mJ=m2 at the equilibrium pressure of Peq = 19.1 MPa. The above experimental data indicate that different groups and amounts of light to intermediate hydrocarbons are extracted from the original light crude oil to CO2 phase. In fact, the measured equilibrium IFT is between different intermediate and heavy hydrocarbons of the original light crude oil and CO2 in the second and possibly first equilibrium pressure ranges. On the basis of the measured data in Fig. 3, the equilibrium IFT ceq ðmJ=m2 Þ is correlated to the equilibrium pressure Peq (MPa) by applying the linear regression in the above-described two equilibrium pressure ranges, respectively:

ceq ¼ 2:04 Peq þ 21:63 ð1:7 MPa 6 Peq 6 8:3 MPa; R2 ¼ 0:982Þ; ð4Þ

ceq ¼ 0:31P eq þ 7:15 ð8:3 MPa 6 Peq 6 19:1 MPa; R2 ¼ 0:981Þ: ð5Þ For this Pembina Cardium light crude oil–CO2 system, the linear regression equation of the measured equilibrium IFT versus equilibrium pressure data for Range I intersects with the abscissa (i.e., ceq = 0) at Peq = 10.6 MPa. Therefore, the MMP of this light crude oil–CO2 system at Tres = 53.0 °C is determined to be 10.6 MPa. In addition, the miscibility pressure between the intermediate to heavy hydrocarbons of this light crude oil and CO2 is found to be Pmax = 23.1 MPa from the linear regression equation for Range II. 3.2. Oil RF and producing GOR A total of five CO2 coreflood tests for CO2 secondary oil recovery were carried out by using the Pembina Cardium sandstone core plugs in series at different injection pressures and Tres = 53.0 °C under the immiscible, near-miscible, and miscible conditions. More specifically, Tests #1 and #2 are immiscible CO2 flooding processes, Test #3 is a near-miscible CO2 flooding process, whereas Tests #4 and #5 are miscible CO2 flooding processes. A constant CO2 volume injection rate of qCO2 = 0.4 cc/min was used to examine the effect of the injection pressure on the secondary oil recovery process. The detailed physical properties of the composite core

plugs for the five coreflood tests are summarized in Table 3. The injection pressures of the five CO2 coreflood tests were chosen in the range of 7.2 MPa 6 Pinj 6 14.0 MPa. Moreover, it is worthwhile to note that after the connate water saturation was reached at the end of the initial original light crude oil injection process, it remained unchanged during each CO2 coreflood test. In this study, no water was produced during the CO2 secondary oil recovery process and during the final original light crude oil reinjection process. The oil recovery factor (RF) at any PV of injected CO2 under the coreflood test conditions is defined as the ratio of the volume of the produced oil at any time to that of the initial original light crude oil in the composite core plugs. Fig. 4a shows the measured oil RF versus the PV of injected CO2 for the coreflood tests at five different injection pressures. As expected, the oil RF is increased with the PV of injected CO2 at each injection pressure and finally reaches its maximum value after at most 1.3 PV of CO2 is injected. In this study, each CO2 coreflood test was terminated at 2.0 PV of injected CO2 when no more oil was produced. A quick increase of the initial oil RF occurs at Pinj = 7.2 MPa was mainly attributed to the gas displacement. At Pinj = 9.2 MPa, the smallest increase of the initial oil RF is found and attributed to the increased solubility of supercritical CO2 in the light crude oil. In this case, a large amount of injected supercritical CO2 was dissolved into the light crude oil especially at the beginning, similar to a ‘‘soaking’’ process. This is why the initial oil RF was rather low at a small PV in Test #2. When the injection pressure is higher than the MMP, a relatively larger increase of the initial oil RF is achieved in comparison with that at Pinj = 9.2 MPa. Also, a substantial increase of the total oil RF is obtained at the end of each miscible CO2 flooding process. The measured cumulative producing gas–oil ratio (GOR) versus the PV of injected CO2 is shown in Fig. 4b. It is found that in general, the producing GOR was extremely low at a low injection pressure before CO2 breakthrough (BT) but increased drastically after CO2 BT. Early CO2 BT (0.4 PV) was observed at Pinj = 7.2 MPa as CO2 was in a gas phase at the lowest injection pressure. When the injection pressure was increased to 9.2 MPa, CO2 BT was significantly delayed to 0.9 PV of injected CO2. In this case, CO2 changed into a supercritical state and a large amount of injected supercritical CO2 was dissolved into the light crude oil, similar to a ‘‘soaking’’ process. When the injection pressure was higher than 9.2 MPa, CO2 BT occurred at 0.3–0.6 PV of injected CO2. Furthermore, CO2 secondary oil recovery process was almost completed at CO2 BT and only a small amount of the remaining oil was produced after CO2 BT under the immiscible conditions. On the other hand, a relatively lower oil RF was achieved at CO2 BT but a much larger amount of the remaining oil was recovered after CO2 BT under the miscible conditions. The producing GOR data in Fig. 4b are in an excellent agreement with the measured oil RFs in Fig. 4a. In summary, a higher injection pressure leads a higher oil RF and a higher producing GOR at 2.0 PV of injected CO2. 3.3. Asphaltene precipitation Fig. 5 shows the measured asphaltene contents of CO2-produced oil versus the PV of injected CO2 for the coreflood tests at five different injection pressures. The measured asphaltene contents of the produced oil samples are all lower than 0.26 wt.% (npentane insoluble) of the original light crude oil. A lower asphaltene content of CO2-produced oil means that some asphaltenes are precipitated and deposited onto the reservoir cores. In this study, only 2–3 produced oil samples collected before 1.0 PV were nearly sufficient for measuring their asphaltene contents. There was not enough produced oil sample collected after 1.0 PV in each test as a much smaller amount of produced oil sample was obtained at a higher PV. It can be clearly seen from the figure that

M. Cao, Y. Gu / Fuel 109 (2013) 157–166

Oil recovery factor RF (%)

100

80

60

40 Test #5 at Pinj = 14.0 MPa Test #4 at Pinj = 12.1 MPa Test #3 at Pinj = 10.4 MPa Test #2 at Pinj = 9.2 MPa Test #1 at Pinj = 7.2 MPa

20

0 0.0

0.5

1.0

1.5

2.0

Pore volume (PV) of injected CO 2 Fig. 4a. Measured oil recovery factor (RF) of CO2 secondary flooding versus the pore volume (PV) of injected CO2 in each coreflood test at qCO2 ¼ 0:4 cc=min and Tres = 53.0 °C.

Producing GOR (ml CO2 /ml oil)

500

400

Test #5 at Pinj = 14.0 MPa Test #4 at Pinj = 12.1 MPa Test #3 at Pinj = 10.4 MPa Test #2 at Pinj = 9.2 MPa Test #1 at Pinj = 7.2 MPa

300

200

100

0 0.0

0.5

1.0

1.5

2.0

Pore volume (PV) of injected CO2 Fig. 4b. Measured cumulative producing GOR of CO2 secondary flooding versus the pore volume (PV) of injected CO2 in each coreflood test at qCO2 ¼ 0:4 cc=min and Tres = 53.0 °C.

163

the measured asphaltene content of the produced oil is decreased with the PV of injected CO2. The highest asphaltene content of the produced oil was always found at the beginning of CO2 injection. A small amount of asphaltenes was present in the produced oil at the end, especially during near-miscible and miscible CO2 flooding processes (Tests #3–5). This indicates that more and more asphaltenes are precipitated and left in the porous media as CO2 flooding process proceeds. In this work, it is speculated that more precipitated asphaltenes are left in the reservoir but more likely produced during the final original crude oil reinjection after CO2 flooding. In this regard, a small amount of the remaining oil in the composite core plugs after CO2 flooding was purposely collected at the beginning of the final original light crude oil reinjection in Test #4 (Pinj = 12.1 MPa). Figs. 6a and b show the respective compositional analysis results and grouped carbon number distributions of the remaining oil and the original light crude oil. It can be clearly seen from these two figures that the light hydrocarbons (C3–C5) are completely extracted and the intermediate hydrocarbons (C6–C8) are also partially extracted from the original light crude oil due to strong light-hydrocarbons extraction by CO2 at the high injection pressure. Therefore, more heavy hydrocarbons are present in the remaining oil. The remaining oil has a much higher mole percentage of the heavy hydrocarbons (C30+) and a much higher molecular weight (MW) than those of the original light crude oil. Moreover, Fig. 7 shows that a relatively higher average asphaltene content of the produced oil is found during an immiscible CO2 flooding process, whereas a relatively lower value is obtained during a miscible CO2 flooding process. In the immiscible case, a less amount of asphaltenes is precipitated due to a relatively low injection pressure and thus more asphaltenes remain in the produced oil. On the other hand, when the injection pressure is much higher in the miscible CO2 flooding process, the distances between the asphaltene molecules are much shortened. As a result, they tend to coagulate and precipitate so that the average asphaltene content of the produced oil is relatively lower at a higher injection pressure. In summary, less asphaltene precipitation is observed due to the low injections pressures under the immiscible conditions, whereas the shortened distances between the asphaltene molecules cause more asphaltene precipitation at a higher injection pressure in the miscible case [30]. In this study, the measured stable differential pressures before and after CO2 flooding were used to calculate the oil effective permeabilities from Darcy’s law, respectively. The calculated oil effective permeability reduction in percentage for each test is given in Table 3 and also plotted in Fig. 7. It can be seen from the figure that the oil effective permeability reduction quickly becomes small as the injection pressure increases from 7.2 to 10.4 MPa. At Pinj P 12.1 MPa, the oil effective permeability reduction is small and reaches an almost constant value. It is worthwhile to mention that the reservoir core plugs may change from water-wet to oil-wet after CO2 flooding due to asphaltene precipitation and deposition [31]. It is speculated that both the asphaltene precipitation and wettability alteration jointly contribute to a larger oil effective permeability reduction in an immiscible CO2 flooding case. In a miscible CO2 flooding case, Figs. 6a and b show that more heavy hydrocarbons, including the precipitated asphaltenes, are left in the reservoir but more likely produced during the final original crude oil reinjection after CO2 flooding. As a result, a smaller oil effective permeability reduction occurs in the second case. 3.4. CO2 oil recovery mechanisms

Fig. 5. Measured asphaltene content of CO2-produced oil versus the pore volume (PV) of injected CO2 in each coreflood test at qCO2 ¼ 0:4 cc=min and Tres = 53.0 °C.

The total oil RF of each CO2 coreflood test at 2.0 PV of injected CO2 versus the injection pressure is summarized in Table 3 and also plotted in Fig. 7. More specifically, when the injection pressure

164

M. Cao, Y. Gu / Fuel 109 (2013) 157–166

15 Original oil (C30+ = 9.97 mol.%, MW = 212.1 g/mol) Remaining oil (C30+ = 15.21 mol.%, MW = 247.3 g/mol)

Mole percentage (mol.%)

12

9

6

3

0

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Carbon number Fig. 6a. Compositional analysis results of the original light crude oil and the remaining oil collected in Test #4 at Pinj = 12.1 MPa, qCO2 ¼ 0:4 cc=min, and Tres = 53.0 °C.

50

Mole percentage (mol.%)

Original oil Remaining oil 40

30

20

10

0

CC 3–C 3-C55

-C88 CC66–C

CC99–C -C17 17

C1818-C –C C 2929

CC30+ 30+

Carbon number Fig. 6b. Grouped carbon number distributions of the original light crude oil and the remaining oil collected in Test #4 at Pinj = 12.1 MPa, qCO2 ¼ 0:4 cc=min, and Tres = 53.0 °C.

is equal to 7.2 MPa, the total oil RF is the lowest. At this pressure, the injected CO2 is in a gas phase with an extremely low viscosity, which leads to an early CO2 BT (0.4 PV) and a relatively high equilibrium IFT (ceq ¼ 6:94 mJ=m2 ) of the light crude oil–CO2 system. When the injection pressure is in the range of Pinj = 9.2– 12.1 MPa, the total oil RF increases substantially. This is caused by a reduced oil viscosity due to an increased CO2 solubility in oil, an increased CO2 viscosity, and a reduced equilibrium IFT. The oil RF increases marginally when the injection pressure exceeds 12.1 MPa. Fig. 3 shows that the measured equilibrium IFT between the crude oil and CO2 remains low, even if the equilibrium pressure is higher than the MMP (i.e., 10.6 MPa). Therefore, it is the multi-contact miscibility (MCM) and the low equilibrium IFT that jointly make the total oil RF high and almost constant in the miscible case [29].

Fig. 8 shows the oil RF at CO2 BT and the total oil RF at 2.0 PV of injected CO2 at each injection pressure. The difference between the oil RF at CO2 BT and total oil RF at the end of each CO2 coreflood test under the miscible conditions is much larger than that under the immiscible conditions, especially at Tres = 53.0 °C. In general, as shown in Fig. 4b, CO2 BT tends to occur earlier as the injection pressure increases, e.g., 0.3 PV in Test #4 and 0.6 PV in Test #5. This leads much lower oil RFs at CO2 BT for these two tests. However, the total oil RFs at 2.0 PV of injected CO2 are still higher for these two tests. This is because under the miscible conditions, the supercritical CO2 can effectively extract light (e.g., C3–C5) to intermediate (e.g., C6–C8) hydrocarbons from the original light crude oil. A sufficient amount of the extracted hydrocarbons exists in the first and possibly second displacement fronts so that the oil bank is miscibly displaced [3]. In the CO2 coreflood test, the extracted light hydrocarbons in the produced oil after CO2 BT can be easily identified because of their lighter colors, in comparison with the dark crude oil produced before CO2 BT. Thus a large increase of the oil RF is achieved after CO2 BT in this case [32]. When the CO2 is in a gas phase or CO2 flooding is under the immiscible conditions, however, weak light-hydrocarbons extraction occurs and contributes to a small increase of the oil RF after CO2 BT. In summary, both the strong light-hydrocarbons extraction and the miscible displacement mechanism contribute to a large increase of the oil RF after CO2 BT under the miscible conditions. Finally, it is well known that the MMP of the crude oil–CO2 system strongly depends on the crude oil chemical composition and molecular weight [33]. As shown in Figs. 6a and b, the remaining oil collected from the composite core plugs after CO2 flooding in Test #4 has a much larger amount of heavy hydrocarbons (C30+) and a much higher molecular weight. This leads to an extremely higher MMP between the remaining heavy hydrocarbons and CO2. As the PV of injected CO2 increases, it becomes more and more difficult for the remaining heavy hydrocarbons and CO2 to achieve their miscibility. This is why CO2 oil recovery usually diminishes after at most 1.3 PV of CO2 is injected in this study, irrespective of the injection pressure tested.

165

M. Cao, Y. Gu / Fuel 109 (2013) 157–166

0.30

100

40

Average asphaltene content of CO2-produced oil wasp (wt.%)

90 0.20

85

0.15

80 75

0.10

70 0.05

0.00

65

6

8

10

12

14

16

60

Total CO 2 oil recovery factor RF (%)

95

Average asphaltene content of produced oil Oil effetive permeability reduction

0.25

30

20

10

Oil effective permeability reduction Δko /ko (%)

Total oil recovery factor

0

Injection pressure Pinj (MPa) Fig. 7. Measured average asphaltene content of CO2-produced oil, total CO2 oil recovery factor (RF), and oil effective permeability reduction versus the injection pressure of each CO2 secondary flooding test at 2.0 PV of injected CO2, qCO2 ¼ 0:4 cc=min, and Tres = 53.0 °C.

MMP1 = 7.6 MPa

MMP2 = 10.6 MPa

Fig. 8. Total oil recovery factor (RF) at 2.0 PV of injected CO2 and oil recovery factor (RF) at CO2 breakthrough (BT) versus the injection pressure of each CO2 secondary flooding test at qCO2 ¼ 0:4 cc=min and two different temperatures.

3.5. Temperature effect In the previous study [29], several coreflood tests with the same crude oil, reservoir brine and core plugs were conducted at a laboratory temperature of Tlab = 27.0 °C. The MMP of the crude oil–CO2 system increases as the test temperature increases, e.g., MMP1 = 7.6 MPa at Tlab = 27.0 °C and MMP2 = 10.6 MPa at Tres = 53.0 °C. Fig. 8 shows the measured total oil RF at the end of each test and oil RF at CO2 BT versus the injection pressure at each temperature. At the lower temperature, the total oil RF is slightly higher than the oil RF at CO2 BT. The largest increase of the total oil RF occurs under the near-miscible conditions, largely due to CO2 gas-to-liquid phase change. At the higher temperature, the total oil RF is much higher than the oil RF at CO2 BT under the miscible conditions. This is because at the lower temperature, liquid CO2 extracts the light hydrocarbons to form

CO2-enriched oil phase. At the higher temperature, nevertheless, the supercritical CO2 has a much stronger ability to extract more and heavier hydrocarbons. As the injection pressure increases at the higher temperature, the extraction ability of CO2 becomes stronger and the total amount of the light to intermediate hydrocarbons to be extracted by CO2 is larger. Thus a longer contact time is needed for the light crude oil and CO2 to reach the miscible conditions. This is why the oil RF before CO2 BT is much lower at the higher temperature because there is no sufficient time to achieve the miscible conditions. As MMP1 is lower at the lower temperature, a higher oil RF is obtained at the lower temperature when the injection pressure is below MMP2. Once the injection pressure exceeds MMP2, however, the total oil RF is much higher at the higher temperature. This indicates that a higher reservoir temperature is preferred as long as the CO2 flooding is miscible.

166

M. Cao, Y. Gu / Fuel 109 (2013) 157–166

4. Conclusions In this paper, it is found that the measured equilibrium interfacial tension (IFT) between the light crude oil and CO2 is reduced almost linearly with the equilibrium pressure in two different pressure ranges. The linear regression equation of the measured equilibrium IFT versus equilibrium pressure data for the first pressure range gives zero equilibrium IFT at the equilibrium pressure of Peq = 10.6 MPa, which is referred to as the minimum miscibility pressure (MMP) obtained from the vanishing interfacial tension (VIT) technique. In addition, a series of five CO2 coreflood tests are conducted to study the oil recovery mechanisms and asphaltene precipitation phenomenon of the immiscible, near-miscible, and miscible CO2 flooding processes in tight sandstone reservoir core plugs. The oil recovery factor (RF) becomes higher at a higher pressure during the immiscible CO2 flooding. Once the pressure exceeds the MMP, the oil RF increases slightly and eventually reaches an almost constant maximum value in the miscible CO2 flooding. Moreover, the measured asphaltene content of the produced oil is reduced with the pore volume (PV) of injected CO2 in each coreflood test due to CO2-induced asphaltene precipitation. A higher average asphaltene content of the produced oil is found under the immiscible conditions, whereas a lower average asphaltene content of the produced oil occurs under the miscible conditions. These experimental findings explain why a near-miscible or miscible CO2 flooding process is always pursued in CO2 field applications if CO2-induced asphaltene precipitation is not of major concern. Finally, the total oil RF is compared at two different temperatures: Tlab = 27.0 °C and Tres = 53.0 °C. The total oil RF is much higher at the higher temperature under the miscible conditions. Therefore, a higher reservoir temperature is desired in order to obtain a higher oil RF, provided that the CO2 flooding is miscible at the actual reservoir temperature. Acknowledgments The authors acknowledge the discovery grant and collaborative research and development (CRD) grant from the Natural Sciences and Engineering Research Council (NSERC) of Canada and an industrial R&D fund from the PennWest Exploration to Dr. Yongan Gu. The authors thank Mr. Andrew Seto and Ms. Suzy Chen from the PennWest Exploration for their technical support. The authors also thank Ms. Xiaoqi Wang at the Saskatchewan Research Council (SRC) and Mr. Shiyang Zhang at SINOPEC for their technical discussion. The authors are grateful to Mr. Weiguo Luo and Mr. Zeya Li at the University of Regina for their technical assistance in the highpressure interfacial tension measurements and in the high-pressure CO2-coreflood tests, respectively. References [1] Moritis G. Special report: EOR/heavy oil survey. Oil Gas J 2006;104(12):37–57. [2] Aycaguer AC, Lev-On M, Winer AM. Reducing carbon dioxide emissions with enhanced oil recovery projects: a life cycle assessment approach. Energy Fuels 2001;15(2):303–8. [3] Holm LW, Josendal VA. Mechanisms of oil displacement by carbon dioxide. J Pet Technol 1974;26(12):1427–36. [4] Brock WR, Bryan LA. Summary results of CO2 EOR field tests, 1972–1987. Presented at SPE joint rocky mountain regional/low-permeability reservoirs symposium and exhibition, denver, CO., March 6–8, 1989; Paper SPE 18977. [5] Ko SCM, Stanton PM, Stephenson DJ. Tertiary recovery potential of CO2 flooding in Joffre Viking pool, Alberta. J Can Pet Technol 1985;24(1):36–43.

[6] Chung FTH, Jones RA, Burchfield TE. Recovery of viscous oil under high pressure by CO2 displacement: a laboratory study. Presented at SPE international meeting on petroleum engineering, Tianjin, China, November 1–4, 1988; Paper SPE 17588. [7] Duncan G. Enhanced recovery engineering: Part 1. World Oil 1994;215(8):95–100. [8] Pyo K, Damian-Diaz N, Powell M, van Nieuwkerk J. CO2 flooding in Joffre Viking pool. Presented at the petroleum society’s canadian international petroleum conference, Calgary, Alberta, June 10–12, 2003; Paper CIPC 2003-109. [9] Green DW, Willhite GP. Enhanced oil recovery. Textbook Series, vol. 6, SPE, Richardson, TX; 1998. [10] Dong M, Huang SS, Dyer SB, Mourits FM. A comparison of CO2 minimum miscibility pressure determinations for Weyburn crude oil. J Pet Sci Eng 2001;31(1):13–22. [11] Flock DL, Nouar A. Parametric analysis on the determination of the minimum miscibility pressure in slim tube displacements. J Can Pet Technol 1984;23(5):80–8. [12] Randall TE, Bennion DB. Recent developments in slim tube testing for hydrocarbon-miscible flood (HCMF) solvent design. J Can Pet Technol 1988;27(6):33–44. [13] Christiansen RL, Haines HK. Rapid measurements of minimum miscibility pressure with the rising-bubble apparatus. SPE Res Eng 1987;2(4):523–7. [14] Thomas FB, Zhou XL, Bennion DB, Bennion DW. A comparative study of RBA, Px, multicontact and slim tube results. J Can Pet Technol 1994;33(2):17–26. [15] Rao DN, Lee JI. Application of the new vanishing interfacial tension technique to evaluate miscibility conditions for the Terra Nova offshore project. J Pet Sci Eng 2002;35(3&4):247–62. [16] Rao DN, Lee JI. Determination of gas–oil miscibility conditions by interfacial tension measurements. J Colloid Interface Sci 2003;262(2):474–82. [17] Nobakht M, Moghadam S, Gu Y. Determination of CO2 minimum miscibility pressure from the measured and predicted equilibrium interfacial tensions. Ind Eng Chem Res 2008;47(22):8918–25. [18] Simon R, Graue DJ. Generalized correlations for predicting solubility, swelling and viscosity behaviour of crude oil–CO2 systems. J Pet Technol 1965;17(1):102–6. [19] Mungan N. Carbon dioxide flooding – fundamentals. J Can Pet Technol 1981;20(1):87–92. [20] Martin DF, Taber JJ. Carbon dioxide flooding. J Pet Technol 1992;44(4):396–400. [21] Kokal SL, Sayegh SG. Asphaltenes: the cholesterol of petroleum. In: Presented at SPE middle east oil show, Manama, Bahrain, March 11–14, 1995; Paper SPE 29787. [22] Speight JG. Handbook of petroleum analysis. New York: John Wiley & Sons, Inc.; 2001. [23] Akbarzadeh K, Hammami A, Kharrat A, Zhang D, Allenson S, Creek J, et al. Asphaltenes – problematic but rich in potential. Oil Rev 2007;19(2):22–43. [24] Sarma HK. Can we ignore asphaltene in a gas injection project for light-oils? Presented at SPE international improved oil recovery conference, Kuala Lumpur, Malaysia, October 20–21, 2003; Paper SPE 84877. [25] Hamouda AA, Chukwudeme EA, Mirza D. Investigating the effect of CO2 flooding on asphaltenic oil recovery and reservoir wettability. Energy Fuels 2009;23(2):1118–27. [26] ASTM D2007-03. Standard test method for characteristics groups in rubber extender and processing oils and other petroleum-derived oils by the clay–gel absorption chromatographic method. West Conshohocken (PA): ASTM International; 2007. [27] ASTM D86. Standard test method for distillation of petroleum products at atmospheric pressure. West Conshohocken (PA): ASTM International; 2003. [28] Cheng P, Li D, Boruvka L, Rotenberg Y, Neumann AW. Automation of axisymmetric drop shape analysis for measurements of interfacial tensions and contact angles. Colloids Surf 1990;43(2):151–67. [29] Wang X, Gu Y. Oil recovery and permeability reduction of a tight sandstone reservoir in immiscible and miscible CO2 flooding processes. Ind Eng Chem Res 2011;50(4):2388–99. [30] He Y, Li D, Hu J, Guo L. Study of asphaltene precipitation in CO2 immiscible process. Journal of Xian Petroleum University (Natural Science Edition) 2011; 26(4): 28–32 [in Chinese]. [31] Escrochi M, Nabipour M, Ayatollahi S, Mehranbod N. Wettability alteration at elevated temperatures: the consequences of asphaltene precipitation. Presented at 2008 SPE international symposium and exhibition on formation damage control, Lafayette, LA, February 13–15, 2008; Paper SPE 112428. [32] Wang X, Gu Y. Four important onset pressures for mutual interactions between each of three crude oils and CO2. J Chem Eng Data 2010;55(10):4390–8. [33] Silva MK, Orr Jr FM. Effect of oil composition on minimum miscibility pressure – Part 1: solubility of hydrocarbons in dense CO2. SPE Res Eng 1987;2(4):468–78.