Laboratory investigation of oil recovery by CO2 foam in a fractured carbonate reservoir using CO2-Soluble surfactants

Accepted Manuscript Laboratory investigation of oil recovery by CO2 foam in a fractured carbonate reservoir using CO2-Soluble surfactants Guangwei Ren, Quoc P. Nguyen, Hon Chung Lau PII:

S0920-4105(18)30359-0

DOI:

10.1016/j.petrol.2018.04.053

Reference:

PETROL 4905

To appear in:

Journal of Petroleum Science and Engineering

Received Date: 26 February 2018 Revised Date:

10 April 2018

Accepted Date: 24 April 2018

Please cite this article as: Ren, G., Nguyen, Q.P., Lau, H.C., Laboratory investigation of oil recovery by CO2 foam in a fractured carbonate reservoir using CO2-Soluble surfactants, Journal of Petroleum Science and Engineering (2018), doi: 10.1016/j.petrol.2018.04.053. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Laboratory Investigation of Oil Recovery by CO2 Foam in a Fractured Carbonate Reservoir Using CO2-Soluble Surfactants

1

Guangwei Ren 1,2 *, Quoc P. Nguyen 1, Hon Chung LAU 3

RI PT

* Corresponding author ([email protected]) 1 University of Texas at Austin 2 currently at Total E&P R&T USA 3 National University of Singapore

Abstract

EP

TE D

M AN U

SC

Miscible CO2 flooding has been used as an EOR method for carbonate reservoirs which hold around 60% of the world’s oil reserves. However, natural fractures, unfavorable mobility ratio and gravity segregation in carbonate reservoirs often lead to premature CO2 breakthrough and bypassed oil. To remedy this situation, CO2 foam has been used to reduce the mobility of the injected CO2. Typically, this employed a water soluble surfactant for foam propagation. However, surfactant transport in the aqueous phase was often hindered by surfactant adsorption and undesirable chemical reactions with reservoir minerals. In this study, we investigated whether CO2-soluble surfactants were more effective than water-soluble surfactant in oil recovery of fractured carbonate reservoirs under miscible conditions. A series of corefloods were conducted to determine the oil recovery factor (RF), speed of foam propagation and foam strength in artificially fractured carbonate cores at 35oC (308.15 K) and 1500 psi (1.034*107 Pa) which was above minimum miscibility pressure. Silurian Dolomite outcrop with permeability of 150 md and West Texas Wasson crude were used. The cores were intermediate-wet indicated by both qualitative and quantitative tests. Three different surfactants were compared including an anionic water-soluble surfactant and other two nonionic CO2 soluble surfactants (2-ethyl-1-hexanol with different ethylene oxide groups) with distinct degree of solubility in CO2. Phase behavior experiments indicated these surfactants did not lower the interfacial tension significantly between the crude and water. RF of CO2 flooding was only 24% due to the heterogeneous nature of the fractured core. Co-injection of CO2 and water increased the RF to 35%, which was further increased to 54% when a water-soluble only surfactant presented. However, use of CO2 foam by the two CO2-soluble surfactants increased the RF to 71% and 92% respectively, with a higher RF for the surfactant that partitioned more to the CO2 phase. Also, pressure drop in different sections of the core confirmed that the surfactant which partitioned more into the CO2 phase gave a fasterpropagating and stronger foam. These results educated that the partitioning of surfactant into the CO2 phase has several advantages. First, it allows surfactant to be transported in the CO2 phase ahead of the aqueous phase thus leading to faster foam propagation. Second, it generated a stronger foam. The combined effect of the two leaded to higher RF in current scenarios. Several hypotheses based on literatures were raised and listed to further interpret the observations. Our results also reinforce that the so-called optimal CO2 soluble surfactant is case dependent and is the function of injection strategy, reservoir environment, and operation pressure or rates as well as other specific conditions. One could tailor a surfactant with suitable solubility in the CO2 phase to optimize oil recovery in fractured carbonates. We believed the results were encouraging enough to warrant further R&D and eventual field piloting.

AC C

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 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54

Keywords: CO2 soluble surfactants, partition coefficient, CO2 foam, fractured carbonate reservoir, EOR 1. Introduction and Background Carbonate reservoirs holds around 60% of the world’s oil reserves (Aknar et al. 2000) and 60% of total miscible-flooding EOR will occur in carbonate reservoirs (Baily 1984). It is sensitive to stress/strain which often makes the carbonate reservoirs naturally fractured (Chillenger and Ten 1983, Sloan 2003), creating thief zones for injected fluids, and leaving the matrix unswept (Benson et al. 1998, Graue et al. 2000). About 80% of carbonate reservoirs are classified as mixed-wet to preferentially oil-wet which are unfavorable conditions for spontaneous water imbibitions (Allan and Sun 2003, Tabary 2009). In the past, much effort has been given to use liquids to enhance oil recovery in non-water wet fractured reservoirs through wettability alternation followed by imbibitions (Gupta and Mohanty 2008, Chen and Mohanty 2014) or interfacial tension (IFT) reduction (Alshehri and Kovscek 2014) or both (Chen and Mohanty 2015). Also, miscible solvent, such as Heptane, Pentane and Decalin, provided

1

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

another choice (Kahrobaei et al. 2012, Chahardowli et al. 2013). However, those solutions with liquid agents are be challenged economically especially in reservoirs with fracture network. Traditionally, gas injection has been considered as inefficient method for enhancing oil recovery for oil or mixed wet naturally fractured reservoirs. Several mechanisms have been proposed during gas injection in these reservoirs, including gas-oil gravity drainage (GOGD) (Farajzadeh et al.2012), gas diffusion(LeGallo, et al. 1997, Eide et al. 2015), light component vaporization (or stripping effect or extraction) (Thiebot and Sakthikumar 1991, Morel et al. 1993), oil viscosity reduction (Boerrigter et al. 2007, Alavian and Whitson 2010), oil swelling (Denney 2008), IFT reduction (Karimaie and Torsater 2010b) and effect of miscibility (Kamali et al. 2011, Suicmez et al. 2011). However, most of them are slow processes and uneconomical. Foam is discontinuous gas phase dispersed within a continuous liquid phase. Bubbles are separated by thin liquid films called lamellae. Lamellae are stabilized by adsorption of surfactant at the gas-liquid interfaces, which inhibits the coalescence of bubbles (Rossen 1996, Nguyen et al. 2000). Foam reduces gas mobility in two manners, either by a drag imposed on flowing bubbles as a result of viscous shear stresses in the thin films which exist between the pore walls and the gas-liquid interface (Hirasaki and Lawson 1985), or by the forces required to push lamellae through the constricted pore throats (Falls et al. 1989). Strictly speaking, supercritical (Sc) or liquid CO2 will produce emulsion-like mixture. Here, we still use “Foam” to be consistent. The co-injection of CO2 and a foaming agent has been proven to reduce gas mobility and effects of heterogeneity and therefore increase sweep efficiency in fracture free media (Grigg et al. 2002). The CO2 foam has been well studied in laboratory (Mukherjee et al. 2014) and successfully applied to unfractured reservoirs (Norris et al. 2014). Foam is often proposed as a solution to overcome heterogeneity and divert fluids to lower permeability zones (Casteel and Djabbarah 1988, Bertin et al. 1999, Li et al. 2012). Only a few studies of foam transport in fractured system have been reported, rarely in field (Ocampo-Florez et al. 2014), but most in laboratory scale. We limit our review in the scope of foam stabilized by surfactants only, but not other agents, such as nano-particles (Aroonsri et al. 2013) or aimed by polymer/gel (Su et al. 2017, Qu et al. 2017). Also, for EOR applications, it is of particular interest to examine foam performance in the presence of oil. Studies of foam performance in fractured system without presence of oleic phase (Chambers and Radke 1991, Kovscek et al. 1995, Yan et al. 2009, Buchgraber et al. 2012, Pancharoen et al. 2012, Fernø et al. 2014, Gauteplass et al. 2015) or in tight reservoir system (Liu et al. 2017) are out of scope here. Moreover, note that modeling studies to reveal and demonstrate foam effects on diverting fluids from fracture to matrix or enhancing CO2 diffusion or GOGD process also are out of current experimental scope, but which could be found in the literatures (Abbaszadeh et al. 2010, Abbaszadeh and Ren 2013, AlMaqbali, et al. 2015, AlMaqbali, et al. 2017, Farajzadeh et al.2012, Ma et al. 2018, Seyed et al. 2007, Zuta et al. 2010a, Zuta et al. 2010b, Zuta and Fjelde 2011). One application of foam in a fractured system is to strengthen and improve the recovery mechanisms mentioned above, such as molecular diffusion, vaporization and GOGD, but without relying on fluids diversion by foam. Zuta et al. (2008, 2010, 2010a) performed a series of experiments and simulations to investigate the transport and retention phenomena of CO2-foaming agents as well as their effects on the transport of CO2 and oil production in a fractured carbonate media because the high viscous phases and interfaces formed by foam may enhance mass transfer of CO2 into water/oil and ultimate oil recovery (Farajzadeh et al. 2007). They found co-injecting CO2 gas and aqueous foaming agent solution gave slightly higher oil recovery than CO2 injection. No significant CO2 mobility reduction has been reported. They concluded molecular diffusion and retention were the dominate mechanisms for transport of CO2 and surfactant from fracture to matrix, which altered the rock wettability and IFT of CO2/oil and surfactant solution system. In turn, capillary imbibition of CO2 and aqueous foaming agent from fracture to matrix was promoted. However, diffusion rate was time-dependent and lab experiments were conducted in hundreds of days. The time frame during a project life may be challenged if diffusion dominated transportation of CO2 for oil production (Grogan and Pinczewski 1987, Do and Pinczewski 1991). Alternatively, another more efficient application is to employ foam as blocking agent in the fracture to reduce the effective fractures conductivity by increasing flow resistance, which could divert fluid flow (aqueous or gaseous or both) to the matrix and thus significantly improve sweep efficiency and oil recovery. Alkan et al. (1991) conducted dynamic experiments to investigate the effect of immiscible CO2 foam on gas flow and displacement in artificially fractured limestone rock, which reduced fluids mobility to a half compared with those obtained from unfractured cores. Panahi et al. (2004) studied the beneficial effects of pre-generated CO2 foam on the oil recovery with composite heterogeneous and fractured systems. Presence of foam could decrease the negative effects of longitudinal fractures and gave 20% of additional oil recovery over gas flooding. Lopera Castro et al. (2009) reported experiments in which the fractures were closed by injecting foam in comparison with closure by increasing overburden pressure. They found that foam was an effective divergent agent to increase the recovery on average by up to 10% over the fracture blockage by applying pressure. Zuta et al. (2009, 2010b) investigated the effect of mode

AC C

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 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56

2

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

of injection on oil recovery during CO2-foam flooding in fractured chalks, and showed that CO2-foam can recover more oil than pure CO2 and WAG, which served as a blocking agent and increased the effective viscosity of the CO2. Skoreyko et al. (2011) conducted experiments in two different fracture systems. The first one was a simple split core where a 1.34 mm fracture was axially and longitudinally filled with metal balls. Natural depletion was followed by nitrogen displacement, then in turn pre-generated foam was injected at constant pressure. They concluded that gas drive due to foam has caused the surfactant solution to be driven through the matrix, thereby causing a reduction of IFT between water and oil, which mobilized the residual oil. Another two experiments were conducted with in-situ generated foam by co-injection of surfactant solution and nitrogen on the second fracture system. The Indiana limestone core (2md) was split in half and it had vuggs by drilling different number of holes, either 19 or 29, filled with sand with 710 µm diameter. They observed complicated behavior and a trend that pressure drop increased with injection pore volume with variable foam qualities. Haugen et al. (2012) investigated the possibility to reduce the fracture transmissibility and divert flow from open fractures into matrix in an oil-wet limestone. They found that pre-generated foam caused an increase in the differential pressure and diverted fluid to matrix, yielding a significant increase in oil recovery (80% OOIP) compared with individual surfactant solution or gas injection (10% OOIP). Conn et al. (2014) visualized the oil displacement in an oil wet 2D microfluidic device which was designed with a central fracture flanked by high-permeability and low-permeability zones. The measured oil saturation and pressure drop indicated an increase in apparent viscosity for pre-generated foam. Foam was shown to more effectively mobilize trapped oil by increasing the flow resistance in the fracture and highpermeability zones and by diverting the surfactant solution into adjacent low-permeability zones. Haugen et al. (2014) experimentally investigated tertiary pure CO2, and pre-generated N2- and CO2-foam injections for enhanced oil recovery in fractured, oil-wet limestone core plugs. Fracture system was built by splitting core along length then assembling back after primary drainage with oil. Miscible CO2 and CO2-foam were compared with immiscible CO2and N2-foam subsequent to waterfloods, in fractured rocks with different wettability. They found immiscible foam was less efficient (30 pore volumes injected) compared to miscible foam (2 pore volumes injected) even though both of them could achieve high ultimate recovery. Those results were attributed to diminished capillary holdup between water and oil by diverted surfactant via N2 foam or between oil and diverted CO2 for miscible CO2 foam scenario. Continuous studies were conducted by Steinsbø et al. (2015), who focused on the performance of secondary and tertiary miscible CO2 and CO2-foam in strong water-wet fractured carbonate core. Two kinds of carbonate were used, chalk and limestone, and two different fracture systems were developed. In “single fracture”, split core was assembled back after primary drainage, with support by polyoxymethylene (POM) spacer in 1000 µm, in which pregeneration was employed. In “Multiple-fracture”, three different pieces of slugs, unfractured, vertically fractured and horizontally fractured, were placed in sequence without POM spacer, where foam was generated in-situ. They reported that fractures dramatically decreased oil recovery for secondary supercritical CO2 injection. CO2 foam accelerated oil production, increased total oil recovery, and required less CO2 injection. With similar core configurations, Fernø et al. (2015) compared EOR performance of miscible CO2 and liquid CO2 foam in fractured carbonate system. A significant oil recovery was observed during CO2 injection mainly driven by diffusion. Pregenerated CO2 foam accelerated oil recovery comparing to pure CO2 injection with adding a viscous displacement. Telmadarreie and Trivedi (2015) investigated the pore level phenomena through specially designed fractured micromodel during CO2 foam and polymer enhanced foam (PEF) on heavy oil recovery after solvent injection. They found that not only injected surfactant solution can be diverted into matrix to recover oil through emulsification and IFT reduction in early stage, but also with stronger foam in later stage, small bubble began move into the matrix to displace the trapped oil. Emulsified oil can also be carried in the lamella before bubble ruptured. Gauteplass et al. (2015) evaluated N2 foam injection for oil recovery in fractured micromodel system and foam displacement efficiency was compared with gas injection. Foam generation by snap-off was observed both in the interior of porous network and at permeability discontinuities between fracture and porous matrix, which was important for in situ foam generation and created different bubble textures. They concluded foam injection significantly enhanced sweep efficiency and successfully diverted gas from a high permeable fracture to a low permeable matrix. Moreover, promising combined methods more than one individual mechanism, such as enhancing capillary imbibitions through wettability modifier (WM), eliminating adverse capillary force by ultra low IFT reduction additives, and improving viscous force by foam, have begun to be investigated. Recently, low tension gas (LTG) concept was proposed (Nguyen et al. 2015), employing a mixture of low IFT formula and foam booster to achieve ultra-low IFT and mobility control simultaneously. Chevallier et al. (2015) screened several LTG formula and investigated their properties to mobilize oil in the poorly-swept matrix of oil-wet “fractured” carbonate, during which pre-generated N2 foam produced 93% OOIP. However, the new coreflooding setup, so-called waterflood fractured reservoir, mimicing fracture and matrix with high permeability foamer and low permeability core, may be less convincing. More representative oil wet fractured core system was employed by Dong et al. (2017, 2018) with a

AC C

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 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56

3

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

three components LTG formula, in which a well defined fracture as created by splitting core lengthwise and controlled of aperture by applying specific confining pressure. IFT measurements and bulk foam tests were also conducted to study the salinity gradient effect to ultralow-IFT foam performance. Results indicated surfactant solution was diverted into matrix and 72% oil recovery was achieved because of mobility control by foam, capillary pressure reduction by low IFT formula and enhanced water phase mobility in matrix. Xiao et al. (2018) presented three characteristic zones identified in micromodel including three regions (high permeability fracture region, low permeability) during displacement of crude oil via surfactant-stabilized foam. a series of pore-level dynamics in these zones were described, such as foam coalescence in front zone, surface wettability alternation in transition zone and strong foam bank zone. They visualized and discussed observed foam “phase separation”, i.e. fluid redistribution, in those three regions. Most recently, Chevallier et al. (2018) presented laboratory studies to apply a new formula in combination of low-IFT and foam boosting surfactants on an artificially fractured reservoir core system. With facilitating from auxiliary tests, including phase behavior with live oil and foam stability under reservoir conditions, coreflooding experiment showed good control by foam of aqueous solution mobility in fracture and efficient imbibitions of aqueous solution from fracture to matrix. Bourbiaux et al. (2017) presented an experimental study to evaluate the impacts of foam agents and WMs, implemented separately or jointly, in nonwater wet fractured carbonate system with pre-generated air foam. They found that joint agents could accelerate the oil recovery more than 100 times relative to spontaneous imbibition, but only aqueous phase penetrated into matrix without foam. Dimensional analysis indicated key roles of fracture-to-matrix permeability ratio, total flow rate and oil-to-foam viscosity ratio. Fredriksen et al. (2018) performed an integrated enhanced oil recovery approach using surfactant pre-flooding to change wettability and improve tertiary CO2 foam injections. Surfactant pre-floods, prior to CO2 foam injection, altered the wettability of fracture surface towards to weakly water-wet to reduce the capillary threshold pressure for foam generation in matrix and create capillary contact between matrix blocks. 32% OOIP was additionally recovered by CO2 foam after secondary water flooding. In the past, most of the foaming agents employed by conventional foam were only water soluble. CO2 is a poor solvent for hydrophilic molecules and polar compounds, in which most of conventional surfactants were insoluble or only slightly soluble at moderate pressure (Consani and Smith 1990 ). Although some early CO2 soluble surfactants including fluorinated or silicone based hydrophobes were used successfully in supercritical fluid research, some inevitable drawbacks, such as cost and toxicity concerns have impeded their use in commercial applications (Johnston et al. 1996). Therefore, efforts have been made to obtain low toxicity and less expensive CO2-soluble hydrocarbon based surfactants, including: hydrocarbon polymers (Sarbu et al. 2000), oxygenated hydrocarbon ionic surfactants (Fan et al. 2005), and nonionic surfactants (Liu et al. 2001, Adkins et al. 2010a, Adkins et al. 2010b, Chen et al. 2010, Xing et al. 2012). Recently, Sanders et al. (2010) reported the design and synthesis of a new class of twin-tailed surfactants based on glycerin and designed for supercritical CO2-water interface. McLendon et al. (2012) reported the efficiency of a branched nonylphenol ethoxylate as a CO2 foaming agent based on coreflood experiments. Ethoxylated amine surfactants, which are switchable from nonionic (unprotonated amine) state in dry CO2 to cationic (protonated amine) in the presence of an aqueous phase with a pH below 6, have been characterized for foam generation up to 120OC in the presence of high salinity brine of 22% TDS (Elhag et al. 2014). Meanwhile, several new CO2-philic surfactants have been synthesized using maleic anhydride with 4-tert-butylbenzyl alcohol (Sagir et al. 2014a), or dipropylene tertiary butyl alcohol (Sagir et al. 2014c), or 2-butyl-1-octanol (Sagir et al. 2014d). Through coreflooding with Berea sandstone, mobility reduction factor around 3.1 was obtained by the incorporation of 0.5% of these new CO2-philic foaming agent (Sagir et al. 2014b). The novel foaming concept in which the surfactant can partition between the CO2 and aqueous phase (Le et al. 2008, Ren et al. 2013, Ren and Nguyen 2016) exhibited several advantages over conventional foam, including lower surfactant adsorption, more indepth robust foam, and higher injectivity. In addition, it enabled the possibility of continuous injection of CO2 with dissolved surfactant to generate foam in-situ when the injected CO2 mixes with formation brine. Field scale simulation demonstrated much better capacity to mitigate gravity segregation effect with novel CO2 soluble surfactant foam (Ren et al. 2013, Ren and Nguyen 2016). Chen et al.(2015) also evaluated an nonionic surfactant with a high degree of ethoxylation with respect to cloud point temperature, CO2/brine partition coefficient and IFT, foam apparent viscosity in sand or glass bead pack, and oil/brine partition coefficient. A field trail was carried out in west Texas where surfactant injection in CO2 phase was used to create a CO2-in-water emulsion or foam to improve vertical conformance and create in-depth mobility control (Sanders et al. 2012). The immature description of CO2 foam performance in the fractured carbonate media illustrated the need for further studies. Therefore, in this paper, we follow the similar idea employing foam as blocking agent, to divert the injected fluids into the low permeable matrix and improve the process efficiency. Several novel CO2 soluble surfactants were employed, whose properties, including solubility in CO2 and partition coefficient between CO2 and brine, have been studied (Ren et al. 2014) and significant advantages over conventional aqueous soluble surfactant

AC C

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 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56

4

ACCEPTED MANUSCRIPT

SC

RI PT

(CD 1045) with respect to foam propagation, injectivity and sweep efficiency have been demonstrated in relatively homogeneous system without presence of oil (Ren et al. 2013, Ren and Nguyen 2016). Here, the study of effect of surfactant transfer between CO2, brine, and oil on oil displacement efficiency during CO2 flooding has been extended to fractured reservoir with variable auxiliary experiments. Co-injection of water and gas and pure CO2 flooding were also conducted as baseline. Both qualitative and quantitative assessments of rock wettability were fulfilled to reveal the foam generation environment. Typical phase behavior test was shown to examine whether the foaming agent used would generate macroemulsion or microemulsion, which could help us to interpret the displacement mechanism during coreflooding. Bulk foaming stability test displayed the oil tolerance with distinct surfactants qualitatively. Miscibility of CO2 with oil sample were examined experimentally with slim tube test and in simulation with CMG/WINPROP. Furthermore, pure CO2 floodings in single matrix (fracture free) core and fractured system were also conducted to demonstrate the impacts of heterogeneity even though miscible condition was achieved. At last, the performances of distinct CO2 soluble surfactants were evaluated as well as an aqueous soluble surfactant with different foam injection qualities. To our knowledge, it is the first time the novel CO2 soluble surfactant foam performances were examined comprehensively in non-water wet carbonate fractured system and compared with commercial aqueous soluble surfactant to demonstrate the role of surfactant partition in CO2 phase.

2. Materials and Methods

TE D

M AN U

2.1 Material Wasson crude oil from West Texas was used. Its viscosity was 7.1 cp (7.1*10-3 Pa.S @35 °C) measured with a Contraves Low Shear 30 rotational viscometer and density was 0.86 g/cm3 (860 kg/m3) with a model LG-3540-106 Gay-Lussac type specific gravity bottle manufactured by Wilmad Labglass. CD 1045 (anionic), and two novel nonionic CO2 soluble surfactants (2-ethyl-1-hexanol), 2EH-PO5-EO9, and 2EH-PO5-EO15 were used for the phase behavior and coreflooding experiments. As shown in previous study (Ren et al. 2014), the partition coefficients of this type of novel surfactants increased with decrease of hydrophilicity, i.e., the former holds 15 times higher partition coefficient than the latter. Therefore, following the prior designation (Ren et al. 2013), we named them 15S (2EH-PO5-EO9) and S (2EH-PO5-EO15) respectively. Silurian Dolomite outcrop (~150 md or 1.48*10-13 m2) was used for the wettability test and coreflood experiments. Synthetic brine with varied concentrations was made with 99.99% pure NaCl. Industrial-grade liquid carbon dioxide was purchased from Matheson Gas, delivered in cylinders with 900 psig (6.2*106 Pa) initial pressure. The density and viscosity of supercritical CO2 under current experimental conditions (1500 psi, 35oC) are 708 kg/m3 and 0.0582 cp (58.2*10-6 PaS) respectively. 60 ft. (18.288 m) of 3/8 in. (0.009525 m) stainless steel tubing with wall thickness of 0.065 in. (0.001651 m) was used for the coil tubing in the slim tube experiment, which was packed with 100 mesh glass beads.

EP

2.2 Experimental methods 2.2.1 Slim Tube The goal of the slim tube experiment was to determine the minimum miscibility pressure (MMP) of the Wasson crude oil with CO2. In the published literatures (Rutherford 1962, Yelling and Metcalfe 1980, Randall and Bennion 1988, Wu and Batycky 1990, Elsharkawy et al. 1992), it has been well studied with different setups, as shown in Table.1. Longer tubing length will help stabilize the process and the displacement is much less dependent on injection (Flock and Nouar 1984). The rate of injection is not very critical as long as the length of the system is sufficiently long (Flock and Nouar 1984). The ratio of particle diameter to tube diameter should be less than 0.01 in order to minimize any wall effects (Perkins and Johnston 1963). Due to the impact of setups, different criteria were employed for MMP determination, such as the recovery greater than some fixed value at some injected pore volume or the inflection point of the recovery versus pressure curve (Wu and Batycky 1990) which was employed in this work. This sand-packed tube displacement apparatus was a device to bring multiple equilibrium contacts between concurrently flowing fluids, which did not intend to indicate ultimate recovery, macroscopic sweep efficiency and transition zone length (Flock and Nouar 1984 ). Also, it is possible the measured MMP in one dimensional nonfractured system is lower than that in fractured system due to multi-dimensional flow and molecular diffusion (Uleberg and Høier 2002). Essentially, CO2 miscibility may present adverse effect to foaming ability (Panahi 2004, Chabert et al. 2014).

AC C

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 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53

Apparatus

5

ACCEPTED MANUSCRIPT

As shown in Fig.1, we packed the stainless steel tubing tightly with 100 mesh glass beads, and made the tubing as a coil. The two accumulators with pistons separating the contents driven by a single Quizix pump, for oil and CO2 respectively, were connected to the inlet of the tubing. There were valves separating the two accumulators so that the content of each one can be injected independently. A BPR was connected to the outlet of tubing to regulate the pressure inside the coiled tubing. A graduated cylinder at the outlet of the BPR was used to collect produced oil.

SC

RI PT

Experimental Procedure Vacuum the glass beads packed tubing followed by fully saturating the coiled tubing with crude oil at 0.5 cc/min (8.3*10-9 m3/S). Then inject CO2 at 0.3 cc/min (5*10-9 m3/S) into the coil to displace the oil. Record the volume of oil recovered at gas breakthrough, and at 1.2 injected PV of CO2. 2 PV of solvent was injected into the tubing at 0.5 cc/min to displace the remaining crude oil. Thereafter, solvent was also cleaned by 2 PV of CO2 injection. All of above cleaning steps were conducted at 2500 psi (1.724*107 Pa). The last step was to open the slim tube outlet to atmosphere pressure and use compressed air to blow and vaporize any remaining solvent out of the tube for 2 hours. Above procedures were repeated for four different pressures to obtain an estimate of the MMP of Wasson crude oil at 35 °C.

TE D

M AN U

2.2.2 Wettability test Previous limited literatures showed that rock wettability may affect foam performance. On one hand, surfactant adsorption may be higher in water-wet (Lescure and Claridge 1986) or oil-wet (Sanchez and Hazlett 1992) depending on surfactant nature and iso-potential of rock. One the other hand, oil-wet environment may be detrimental to foam generation (Aarra et al. 2011). From point of view of lamella creation by snap-off, which required much lower capillary pressure in non-water-wet throats than perfectly water-wet ones, lamella would detach and collapse if pore wall were energetically indifferent to contact with the surfactant solution (Rossen 1996). Sanchez and Hazlett found that, without presence of oil, surfactant solution reversed the wettability of oil-wet bead pack and stabilized the foam. There was no change in wettability and no foam if there was residual oil in an oil-wet beadpack. Other researchers, however, created weak foams in oil-wet rock with oil present (Suffridge et al. 1989, Prieditis and Paulett 1992, Latta et al. 2013), which was attributed to the change of wettability. With effective wettability alternation from intermediate-wet to water-wet, foam performance was not affected (Schramm and Mannhardt 1996, Mannhardt 1999). Recently, Haugen et al. (2014) even reported that the ultimate oil recoveries were higher in oil-wet cores than in water-wet cores, both during CO2 and CO2-foam injections. Indeed, the wettability impact on foam generation and stability deserves further study and bubble generation may be dominated by other mechanisms, such as bubble division and leave behind (Kovscek and Radke 1994), besides snap-off. Here, wettability of employed outcrops was probed by contact angle test qualitatively with standard imbibition cell (Fig.2) and Amott Index measurements quantitatively with a model CRU-5000 centrifuge.

EP

Experimental Procedure For the contact angle test, one drop of crude oil or brine was put onto the surface of the rock separately. Photograghs were taken to record the instantaneous responds without presence of the second liquid phase. After the oil or brine has been fully absorbed into the rock, then put one drop of crude oil onto the brine spot and another one drop of brine onto the oil spot. Photos were also taken to prove our preliminary judgment. Multiple samples from distinct positions on outcrop block were tested to prevent the bias observations. The Amott wettability index measurements were performed using the procedure described in petrophysics text book (Za and Tiab 2003).

AC C

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 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55

2.2.3 Phase Behavior The purpose of the phase behavior experiment was to examine the affinity between used foaming agent and oil samples. As mentioned above, surfactant solution could be diverted into matrix by foam, which may reduce interfacial tension between water and oil (Lopera Castro et al. 2009, Haugen et al. 2012, 2014). However, this really deserves further thoughts. Essentially, surfactant is amphiphile, which attempts to partition into both hydrophilic and hydrophobic phases and retains at interface. Foam stability, which requires surfactant present at water/gas interface, may be impaired by preferential partition of surfactant into oleic phase (Princen and Goddard 1972, Srivastava 2010). Meanwhile, preferential presence of surfactant at water/oil interface may produce additional phase. For example, microemulsion or macroemulsion or other high viscosity phases (liquid crystal, gel) whose composition and structures were determined by salinity, temperature and other variable (Bourrel and Schechter 1988). Those high viscosity phases may plug the core (Irani and Solomon 1986). Therefore, from point view of foamability,

6

ACCEPTED MANUSCRIPT

preferential presence of foaming agents at water/oil interface will impose adverse effect on bubble generation and stability.

RI PT

Experimental Procedure Eight solution salinity test points were chosen for every surfactant, from 1 wt% to 15 wt%, in 2 wt% increment. Put 2 cc surfactant solutions (0.2wt%) and 1 cc oil into a 5cc capacity pipette, and then seal the opening. Mix the fluids by rotating the pipette frequently. The pipettes were then placed in an oven with constant temperature of 35 °C for 10 days to equilibrate.

SC

2.2.4 Bulk foam test It was generally recognized that oil tended to destabilize foam, which depended on oil composition, surfactant properties and aqueous phase composition (Li et al. 2010). Even though no reliable correlation has been found to exist between bulk foam stability and performance in porous media (Osei-Bonsu et al. 2017), we are still able to gain some sights of oil tolerance with different surfactants from foam stability test, since oil displacement efficiency indeed was strongly influenced by surfactant formula. Therefore, bulk foam stability with presence of oil was examined for three employed surfactants at 35 OC. Same amounts of surfactant solution with 3wt% salinity and oil sample were shaken in a graduated tube evenly and slightly. Foam height was observed with time, which gave oil tolerance qualitatively.

M AN U

2.2.5 CO2 flooding in 1 ft core (with and without fracture) Pure CO2 flooding in 1ft unfractured and fractured cores were conducted respectively, to demonstrate the dramatically adverse impact of high heterogeneity on sweep efficiency and oil recovery even though minimum miscible conditions have been achieved. The virgin core was fully saturated with brine initially after proper core preparation described in the following section. Then flood the core with crude oil until the water saturation decreases around 0.5. Pure CO2 flooding was conducted with 0.8 cc/min (~8.122*10-6 m/S or 2.333 ft/D) at 1500 psi and 35 oC until oil cut in produced effluent was close to 0.

TE D

Apparatus A schematic of core flood setup was shown in Fig.3. It was comprised of three main modules: fluids injection system, coreholder and pressure transducers, and backpressure and effluent collection system. Fluid injection system. A TELEDYNE ISCO Model 500D syringe was used for brine or surfactant solution injection. CO2 was displaced into the core by DI water delivered by Quizix pump through a high pressure accumulator in which two phases were separated by a piston.

EP

Core holder and pressure transducers. A Phoenix Hassler-type core holder was mounted in the vertical direction and fluids were injected from the top. Hydraulic oil was used as an overburden fluid, which compressed and sealed the 0.25 inch thick rubber sleeve to assure the axial flow of the injection fluids, and to prevent leak. The core holder has two end caps, and the top one has an adjustable end plug length to accommodate different core lengths. There were five pressure taps along side of the core holder in the vertical direction, which connected two absolute pressure transducers (channel 1 and 5) and three differential transducers (channel 2, 3 and 4). The differential transducers detected the pressure drops over sections along the core from the top, whose lengths are 2, 4, and 4 inches and denoted as section 1, 2, and 3 respectively.

AC C

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 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56

Backpressure regulator (BPR) and effluent fractional collector. Two BPRs were used in series to maintain a constant backpressure of 1500 psig during corefooding. The first BPR placed immediately at the outlet of the core holder was set to 1500 psig, and the second BPR was set at 1100 psig. The purpose of the second BPR was to prevent an undesired abrupt flash of CO2 and stabilize first BPR better. An ISCO fractional collector that had 48 test tubes with 10 ml capacity was employed to collect the produced effluents. Oil produced in each tube was read by naked eyes. Experimental Procedure Core preparation. Core was fractured into two halves along the longitudinal axis. Smooth fracture surfaces with minimal surface roughness are obtained when using band saw (Haugen et al. 2012). Then, it was rinsed and dried in a convection oven at 110 °C for 24 hours. We taped the three precut Teflon film strips with 75 micron (7.5*10-5 m)

7

ACCEPTED MANUSCRIPT

thickness onto the core’s cut surface with even spacing to build an artificial fracture whose volume was around 0.1% of whole core volume. Then the split core was fixed with tape, as shown in Fig.4. Since high-pressure CO2 could diffuse through the rubber sleeve and cause leaking, the core was wrapped in three layers of aluminum foil and a thin Teflon heat shrink tube. An Enerpac model P-391 hydraulic pump was used to exert confining pressure around the rubber sleeve. The wrapped core was then placed in the core holder whose outlet was connected to the vacuum pump with closed inlet. Both axial and radial confining pressures were maintained to keep the Teflon tubing, aluminum foil, and the sleeve in good contact. The core was vacuumed for 10 hours before measuring core porosity and saturating the core with formation brine.

26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54

=

RI PT

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

Where w was the fracture aperture in cm and

M AN U

SC

Porosity measurement. With ISCO syringe pump, the brine was injected at 0.5 cc/min while the outlet remained closed until stopping pressure (200 psi or 1.38*106 Pa). The total volume of brine injected into the core was recorded and referred as the core pore volume, with deduction of dead volume. Then, porosity was calculated accordingly. Core Saturation. Thereafter, brine solution was injected at 2 cc/min while the outlet was switched to the BPRs. Adjust hydraulic pump to increase the confining pressure correspondingly until 2000 psi (1.38*107 Pa) when brine began to flow through outlet. Then switch the injection rate to 0.5 cc/min. The brine injection stopped after 10 PV. Permeability measurement. The core permeability was measured with brine solution based on Darcy’s law. The consistence of fractured core permeability and pore volume among cases were monitored to ensure the fair comparison. With known permeability of matrix (~150 md) and dimensions, fracture permeability could be reversely estimated from measured overall permeability of the fractured system. Meanwhile, on the basis of proper assumptions, such as parallel plates, it also could be calculated with following equation, (1)

k f was the permeability of fracture represented by the open space.

EP

TE D

Thus, for our case here, 75 micron corresponded to 475 Darcy (4.69*1010 m2) approximately. Note that the Teflon strips occupied around 25% fracture area. It was maintained that the deviations of the estimated and calculated fracture permeability were lower than 15% with accounting for the heterogeneity of outcrops, since cores were not reused among cases. Oil injection. 2 PV crude oil was injected into the core at a pressure drop of 400 psi (2.76*106 Pa). The volume of oil in the core was calculated with mass balance with accounting for system dead volume. Oil saturation (So) was calculated with the following equation:

=





(2)

AC C

The initial oil saturations before each tested strategy and other composition core properties were listed in the Table.3 below, which demonstrated the close values among cases. 2.2.6 Co-injection of CO2 & Water in fracture system To better demonstrate the foam effect, co-injection of CO2 with brine was also conducted as baseline. Both 75% and 50%, gas volumetric fractions were tested. For 75% quality, the liquid and gas injection rates were 0.2 and 0.6 cc/min (3.33*10-9 m3/S and 1*10-8 m3/S) respectively at reservoir conditions (1500 psi and 35oC), while for 50% quality, gas rate was reduced to 0.2 cc/min without varying the liquid rate. Pressure drops across the core were recorded as well as the oil productions in fractional collector. 2.2.7 Foamflood in fracture system Liquid solution with variable surfactants (0.2wt%), CD 1045, S and 15S, were co-injected with CO2 respectively, without pre-saturation of surfactant solution. S and 15S were also delivered in aqueous phase even though they are CO2 soluble surfactants, which served the purpose here. It was expected that delivering by different media may give absolute performance difference, such as incubation time, and foam transportation, but not final

8

ACCEPTED MANUSCRIPT

RI PT

liquid desaturation. However, the priority goal here was to compare the performance of different surfactants but not individual or absolute one. as shown in Appendix A, by thermodynamic equilibrium, owing to constant rates injection and constant partition coefficient between phases, relative magnitudes of surfactants propagation contrast verse time in aqueous phase or in “Global” as same, which eliminated the dilemma of delivering media. In-situ foam generation was employed instead of pre-generated bubbles to probe the more realistic scenario in the field even though prior results suggested the former may be harder to achieve (Kovscek et al. 1995, Yan et al. 2009). Injection foam qualities that were volumetric fraction of gas phase, 75% and 50%, and phase injection rates were consistent with those used in above baselines without employing surfactants. For 50% quality, solo reduction of gas injection rate was aimed to keep same amount of surfactant injection rate. Pressure drops across the core were recorded as well as the oil productions in fractional collector. Outcrop cores were not reused and consistence of initial conditions was ensured by examining matrix permeability, composite system permeability, porosity, and initial oil saturation.

3. Results and discussion

TE D

M AN U

SC

3.1 Slim Tube We employ the definition of MMP as the pressure where at least 90% of the oil was recovered after 1.2 PV of gas injection. As shown in Fig.5, the inflection point on the recovery curve indicated the MMP of the tested Wasson crude oil was around 980 psi (6.76*106 Pa) at 35 oC. Accordingly, oil compositions were tuned roughly on the basis of known composition of oil sample from the field close to Wasson with CMG/WINPROP until the similar multiple contact miscible (MCM) pressure was achieved. It showed that first contact miscible (FCM) pressure of tested oil here was as high as 3000 psi (2.07*107 Pa). However, it is worth mentioning that developed miscibility in conventional one dimensional system was dependent on fluid properties alone, which may not be the case for heterogeneous and fractured reservoir. Uleberg and Høier (2002) showed that the minimum miscibility pressure/enrichment (MMP/MME) level in a fractured reservoir was significantly higher than for a conventional one dimensional non-fractured system, which was mainly due to multi-dimensional flow and molecular diffusion. This is important for current process since either the foam or emulsion discussed here required the presence of CO2 free phase. In other words, too high miscibility between CO2 and oil may impose adverse effect on foam stability.

EP

3.2 Wettability test The performances of oil and brine on two different rock samples from same block were shown in Fig.6. In Pic 1, oil droplet was absorbed much quicker than water droplet and contact angles also indicted the virgin employed outcrop may be slightly oil wet. This has been confirmed by putting water or oil droplet onto fully absorbed oil or water droplet at same locations respectively, as shown in Pic 2. Test was repeated with another outcrop on same block (Pic 3 and 4), which indicated intermediate wet. It was observed that the wettability may vary with location due to rock heterogeneity. However, in general, the employed outcrops were non-water wet.

The Amott wettability indices of water (WIw) and oil (WIo) were calculated as follows with measured values, summarized in Table. 2,

AC C

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 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53



=

=











































(3)

Both WIw and WIo were close to 0, which indicted the used Silurian dolomite outcrops were intermediate wet. The limitations of Amott Index measurement are it may not be very sensitive at intermediate wet condition and it may be affected by core pore size distribution. From this point of view, contact angle test is less affected by core geometry characterizations. Therefore, it is important to have consistent observation between those two methods. 3.3 Phase behavior

9

ACCEPTED MANUSCRIPT

SC

RI PT

Figs.7 demonstrated the pictures for three tested surfactants at variant brine salinities, in which no typical Winsor type phase behavior variation was observed. CD 1045 produced small amount of “middle” phase at each tested brine salinity (Fig.7a), which was slightly more viscous than both oil and water phases. We believed it was caused by the precipitated insoluble solids such as asphaltene. This “junk” phase may consume small amount of CD 1045, but not severe. The viscosity of produced phase was not high by naked eye observation through titling test. Whether it will reduce the injectivity could be examined through coreflooding experiment. CD 1045 was normally employed as foaming agent in the past studies with proven good performance (Liu et al. 2005) and rarely few publications have been found to probe its phase behavior property with oil. On the contrary, this concern was eliminated for two novel CO2 soluble surfactants, which displayed the much less affinity to oil sample (Fig.7 b & c). In the literatures, similar concerns were raised by other researchers. For example, no stable viscous emulsion is observed by Chen et al. (2012) with a switchable CO2-soluble cationic surfactant, and by Mukherjee et al. (2014) with a water soluble surfactant. Chen et al.(2015) measured anther nonionic CO2-soluble foaming surfactant partition coefficient between oil/water, which was very small. Hence, the concern regarding of strong preferential presence of employed foaming surfactants on oil/water interface could be eliminated and they mainly served for bubble generation and stability.

M AN U

3.4 Bulk foam test Fig.8 displayed the measured foam column height with time for three tested surfactants. It seemed that both CD 1045 and 15S had good and similar foam stability, while oil tolerance of S was weaker. The quantitative study for the discrepancy owing to surfactant molecular structure distinction may be required and is in the further scope. Meanwhile, for CD1045, it was worth noting that current test may be not necessary to account in the effect of junk phase as shown in above phase behavior test, since the generation of such thermodynamically unstable phase require more server mixing and longer time in bulk. This bulk stability test with air under ambient conditions may not reflect the foam generation and coalescence in the porous media under reservoir condition with CO2, but it may give qualitative insight on oil tolerance capacity.

EP

TE D

3.5 Pure CO2 injection with/without fracture The oil recoveries with pure CO2 flooding in the systems with and without fracture were shown in the Fig.9. With presence of fracture, oil production was relative high in a short period initially, but leveled off quickly, which was caused by the preferential flow of low viscosity gas through high permeability fracture. In turn, only the region near the fracture has been swept and low recovery (24%) was achieved at limited period. Meanwhile, as mentioned above, it was possible the MMP measured in one dimensional non-fractured system was lower than that in fractured system due to multi-dimensional flow and molecular diffusion (Uleberg and Høier 2002). In contrast, without a fracture (Fig.9), recovery was improved dramatically to 61%. However, it was still much lower than 90% even though MCM has been satisfied as shown in the slim tube test. This was attributed to not only presence of water but also poor sweep efficiency due to heterogeneity of consolidated outcrop. Presence of water could result in a complex saturation pattern (water shielding or blocking effect) (Tiffin et al. 1991), which may reduce the CO2-oil contact, especially for this mixed wet rock. Meanwhile, the unfavorable mobility ratio between gas and liquid due to low viscosity of CO2 as well as the heterogeneity of carbonate core played an important role in the poor sweep efficiency.

AC C

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 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56

3.6 Co-injection of CO2 with brine in fracture system Analogous to the pure CO2 flooding shown above, co-injection of CO2 and brine at variable gas fractions were also employed as baseline for the following foam cases. As shown in Fig.10, all of those three scenarios hold low pressure drops (DP) corresponding to unfavorably low oil desaturation rates (R). Oil broke through at close injected PVs (IPVs) contributed by the presence of fracture. 20% oil recovery in the early time (0.1 IPV) was attributed to the oil produced near the inlet with negligible fracture volume. The simultaneous injection of gas and water was suggested to reduce capillary entrapment of oil, providing better mobility control of the gas than continuous water/gas injection (Samchez 1999). Gas trapping may be helpful for incremental oil recovery due to effect of threephase relative permeability. However, those enhancements are only valid in small scale reservoir heterogeneity. The presence of fracture dominated the process and directly resulted in the poor mobility control revealed by the low pressure drops. Higher gas fraction (75%) improved the recovery to a really limited extent.

10

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

3.7 Foam coreflooding with CD-1045 in fracture system Fig.11 demonstrated the pressure drops and recoveries of simultaneously injecting CO2 and CD 1045 surfactant solution with different foam qualities. Relative to pure gas flooding and co-injection of brine and CO2 (Fig.10), presence of foam accelerated the oil desaturation rate with improved mobility control, which has been proven by most of the lab experiments (Svorstol et al. 1996). Enhanced foam quality by increasing gas injection rate did improve the foam performance slightly. Foam in porous media is a dispersion of gas in liquid such that the liquid phase is continuous and at least some part of the gas phase is made discontinuous by lamellae (Falls et al. 1988). The thin films, called lamellae, that separate gas bubbles, are stabilized by the presence of surfactant molecules. Foam will reduce gas mobility through decreasing effective relative permeability by gas trapping because of the forces required to push lamellae through constricted pore throats (Bernard and Jacobs 1965) and increasing apparent viscosity by viscous shear stresses in thin films between the pore walls and the gas-liquid interface (Falls et al. 1988). The aqueous soluble only anionic surfactant employed here, CD 1045, has proven to be a good CO2 foaming agent with respect to variable salinities and pH values (Liu et al. 2005). It has been observed by many researchers that total injection rate could enhance the foam strength even with fixed foam quality (Osterloh and Jante 1992). With higher foam quality, foam strength could be enhanced as long as water saturation is far from critical water saturation which corresponds to the critical capillary pressure for foam coalescence (Friedmann et al. 1991). We also observed small amount of oily phase (emulsion) with lighter color and slightly higher viscosity in effluents. Previously, French et al. (1986) employed the in-situ generated emulsions during steamflooding as blocking and diverting agent and recently Chen and Zhao (2015) partially attributed additional oil production during foaming flooding to the emulsions generated by AOS and middle chain alcohols. However, for the scenario here, from pressure drop data, it seemed this higher viscous separate phase did not affect the injectivity severely, in turn which indicated its contribution to oil recovery from point view of mobility control may be limited.

EP

TE D

3.8 Foam coreflooding with CO2 soluble surfactants in fracture system When the surfactant could partition between aqueous and gas phase, i.e., the CO2 soluble surfactants were employed, foam performance could be improved further. As seen in Fig.12, the oil recoveries of both novel surfactants, 15S and S, were higher than all of the previous cases (Fig.10 and 11) with much higher pressure drop (Figs.13 to 16). As demonstrated in previous study in oil-free and non-fracture system (Ren et al. 2013), the faster foam propagation indicated by pressure drop and liquid desaturation rate were mainly attributed to the partition ability of novel surfactants in CO2 and partially to higher adsorption of CD1045 (1.1 mg/g) than S (0.25 mg/g) & 15S (0.65 mg/g). Higher mobility of gas than water phase really enhanced the novel surfactants propagation. Here, the much lower pressure drop shown by CD1045 indicated CO2 soluble surfactant foam may hold much better stability under reservoir conditions owing to improved phase dispersion caused by affinity of CO2 soluble surfactant to both aqueous and CO2 phases. Partially but to less extent, it may be also caused by certain surfactant loss to junk phase as shown in phase behavior test and by higher surfactant adsorption of CD1045. As mentioned above, the bulk foam test may not include the junk phase effect owing to less mixing conducted there and did not indicate the stability under reservoir conditions with presence of CO2. Meanwhile, higher foam quality (75%) outperformed 50% quality slightly, which was also attributed to the lower total injection rate of the latter scenario (M’barki et al. 2017). Before discussing the results of CO2 soluble surfactants in current fracture system, let’s briefly recall the observations in precious oil-free and non-fracture system (Ren et al. 2013). There, S outperformed 15S in faster foam propagation and liquid desaturation rate as well as higher pressure drop during pseudo steady state. Those were caused by surfactant spreading relevant to critical surfactant concentration (Csc) for foam generation and propagation (Appendix A), in conjunction with relatively lower adsorption of S relative to 15S on carbonates. The Csc was highly relevant to strong foam generation (Laurier et al. 1994) and indicated by the dramatically increasing of apparent foam viscosity or turning point from weak to strong foam. Similarly, based on those observations, Zeng et al. (2016) demonstrated surfactant partition effect and spreading effect through a 1-D foam simulation for different injection strategies. They concluded that surfactant with unit partition coefficient was advantageous to foam transport in regard to foam strength and propagation speed. However, we believed it may not be always true even without accounting for gravity force. Here, a concept model was built with CMG/STARS to demonstrate those effects, as shown in Appendix A in details. Theoretically, initially, by honoring mass conservation, S will display earlier foam responds (reaching Csc) than 15S since the concentration of the latter will be spread and lowered owing to its higher partition into high mobility CO2 phase. Thereafter, 15S may still show slower foam propagation even though its concentration passed the Csc (1st scenario), but its foam transportation front was approaching that of S. In

AC C

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 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56

11

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

the later time, once the cross point of 15S concentration trajectory with Csc passed over that of S, it will demonstrate faster foam propagation (2nd scenario), even though it holds higher surfactant adsorption. Those two scenarios will show up in series for a foam free system. However, 2nd scenario may not present for certain cases. For example, low surfactant injection rate or short core or too high surfactant partition coefficient or high Csc, any of them may mask the presence of 2nd scenario. Therefore, the two scenarios could give opposite choice of “optimal partition coefficient” which was really function of system (reservoir), fluid properties, injection conditions and injection strategy. In Appendix B, we demonstrated this idea with field scale simulation in details as continuous clarification of previous publication (Ren et al. 2013). Now, relative to prior results in oil-free and non-fracture system, the opposite phenomena were observed here that the oil recoveries of 15S were always higher than those of S (Fig.12) with higher pressure drops (Fig.13 ~16). Some hypothetic possibilities were listed below which were attributed to presence of fracture and oil as well as surfactant natures. The first, several uncertainties were promoted by surfactant natures. In above analysis with concept model simulation, we assumed same Csc was applied to both CO2 soluble surfactants. However, it was known that Csc was function of surfactant nature and structure, which may be a few times of critical micelle concentration (CMC) under corresponding conditions (Osman and Anthony 2000). In turn, this will affect the bubble generation, and coalescence (film stability). It was possible the 15S holds lower Csc than S, which promoted the foam propagation macroscopically. On one hand, with less EO group, 15S displayed less hydrophilic than S, which may provide more proper affinity at CO2/water interface since the higher hydrophilicity may enhance interface tension, in turn, deteriorate foamability. On the other hand, these relationships may not be able to simply derive from surfactant hydrophilicity but also depend on other surfactant natures. For example, microscopically, the foamability of surfactant was partially affected by micelle-monomer equilibrium (Seong-Geun and Dinesh et al. 1991). In other words, surfactant micelle disintegration speed or micelle stability indicated how fast the micelle could release the monomer, which explicitly affected foamability. Furthermore, critical capillary pressure for coalescence, indicated by highest disjoining pressure in bulk and named as limiting capillary pressure in porous media, played key role on foam stability which was strong function of surfactant nature (Khatib et al. 1988, Jimenez and Radke 1989, Kovscek and Radke 1994). The bubble stability was determined by disjoining pressure statically and film viscoelasity dynamically. Therefore, less micelle stability and higher limiting capillary could promote the foam stability and foamability. The second, the matrix permeability here (150 md) was half of that in previous publication fracture free system (300 md) (Ren et al. 2013), which may increase critical capillary pressure (Khatib et al. 1988). In turn, bubble would be more stable. The third, the bulk foam stability test provided some clues qualitatively. The 15S may display higher oil tolerance than S, which was relevant to how the oil drop overcame electrostatic and steric interactions in the aqueous pseudo-emulsion film to destabilize the lamella by spreading along the gas/water interface (Lau et al. 1988) or by bridging (Garrett 1993). The fourth, the presence of fracture promoted fluids transportation and decreases the flow frontage discrepancy between gas and water in fracture. Meanwhile, relative to previous case (Ren et al. 2013), doubled injection rates were employed here, which increased amount of surfactant injection. Flow frontage discrepancy between gas and water in matrix may be enlarged with lower matrix permeability, which will promote presence of 2nd scenario with sufficient surfactant. Therefore, the late 15S foam show-up due to spreading effect (1st scenario) may be impaired. State alternatively, the transition to 2nd scenario may be shortened. The fifth, the in-situ gas velocities in fracture could be higher than injection ones, which will promote foam coalescence by shear thinning effect (Alkan et al. 1991 ) or foam rupture by decreased critical capillary pressure (Khatib et al. 1988, Jimenez and Radke 1989). 15S foam could have different rheology and be more stable under in-situ gas velocity if it provided less shear thinning behavior or higher critical capillary pressure to velocity variation. All of above hypotheses deserve further studies in details to disclose the true mechanisms behind the observations. It may be the combination effects of above hypotheses and they can not be derived explicitly through current macroscopic scale corefloodings. Meanwhile, the results observed here reinforced our previous conclusion that the so-called optimal surfactant partition coefficient of CO2 soluble surfactant was case dependent and was the function of injection strategy, reservoir environment, and operation pressure or rates as well as other specific conditions, but was not simply indicated by its absolute value relative to unit.

AC C

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 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55

The oil recoveries of all the tested cases in fractured system were summarized in Fig.17 as well as Table.4, in which the pore volume and initial oil saturation were also listed. It was obvious that pure CO2 injection held the lowest oil recovery with least mobility control. With presence of water, gas mobility was inhibited slightly with respect to channeling. However, it also lost efficiency at less than 1 IPV. With surfactant involving, foam highly improved the mobility control. Piston-like gas fronts appeared and 15S promoted the foam performance with

12

ACCEPTED MANUSCRIPT

quicker surfactant propagation. In turn, foam performances were in the order of 15S, S and CD1045. 75% injection quality indeed outperformed corresponding 50% quality, which may reflect the role of total injection rates.

4. Further discussion and future work

EP

TE D

M AN U

SC

RI PT

The observations here were encouraging and inspiring. However, there were indeed several unclear aspects with this novel technology, which partially invoked the mechanism interpretations on the basis of hypotheses. Nevertheless, it was impossible to disclose those hypotheses mentioned above clearly under such macroscopic experiments, which deserved further studies in depth and may be probed as followings, 1. Microscopically, we need better understand how surfactant distributes at water/CO2 interface when the surfactant solubility in CO2 varies from near zero to non-zero. Surfactant adsorption may be disclosed y equilibrium and dynamic surface tension. The thermodynamic and kinetic properties of film rheology should be studied, with accounting for CO2 phase behavior. In turn, those will directly affect surfactant foamability and bubble stability. 2. CMC of employed surfactants could be derived from interfacial tension measurement between brine and CO2 at variable temperature, pressure, and salinity. Those could be used to estimate Csc indirectly. 3. Furthermore, direct pursue of Csc for stable foam generation could be conducted in bulk column test and unconsolidated/consolidated corefloodings, without presence of oil. It is possible Csc may vary among different CO2 soluble surfactants which have distinct partition coefficients between CO2 and aqueous phases, since the shape of disjoining pressure isotherm reflects the surfactant charge, size, structure and concetrtion as well as solvent ionic strength (Jimenez and Radke 1989). Hence, those values in conjunction with measured CMCs could facilitate us to better understand the spreading effect and interpret the coreflooding results here. 4. Regarding of oil detrimental effect on foam thermodynamically, we need to address the stability of pseudoemulsion film through measurement of its disjoining pressure with presence of oil (Bergeron et al. 1993). Kinetically, single film viscoelasity may be studied. Macroscopically, co-injection surfactant solution / CO2 with variable oil rate in outcrop or sandpack may give some ideas how foam performs in porous media at variable oil fractions. 5. As discussed above, non-water wet rock surface could be detrimental to surfactant foamability. However, here foam generation and propagation was not severely prohibited by this unfavorable rock condition, especially for employed CO2 soluble surfactants. It is desirable to probe the wettability alternation effect with employed surfactants. Also, high pressure micromodel experiments could be conducted to reveal whether snap-off still dominates bubble generation with those novel surfactants. 6. Critical capillary pressure for foam coalescence is strong function of surfactant nature and also affected by other factors, such as permeability and gas velocity (Khatib et al. 1988, Jimenez and Radke 1989). Therefore, it is interesting to probe those relationships for those employed surfactants under reservoir conditions with presence of CO2, which may be necessary to better analyze the corefloodig results. 7. It may be good to find another conventional aqueous soluble only foaming agent which does not produce any additional phase with oil to validate our comparison here. 8. Effluents could be analyzed through HPLC to precisely identify surfactant propagation with time. CT scan may be used to profile phase distribution in the core. In the future, surfactant may even be designed and manufactured with radioactive tracer to directly demonstrate surfactant distribution. 9. The foam rheology with different surfactants at distinct foam qualities could be tested in capillary tube viscometer and in the porous media.

AC C

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 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55

5. Conclusion

In summary, it was the first time the CO2 soluble surfactant foams performances were reported in an intermediate wet or weak oil wet carbonate fractured system. Relative performance comparisons disclosed not only the superiorities of CO2 soluble surfactant over conventional aqueous soluble only surfactant as foaming agent, but also the effect of surfactant partition coefficient. Faster foam propagation can dramatically improve oil production in limited time frame and meet the economical concern better. Higher foam strength facilitates fluid diversion and improve the CO2 utilization efficiency, which will significantly reduce the gas compressing cost. Also, this will be beneficial to mitigate gravity segregation due to essential thermodynamically unstable nature of foam, which is especially important in highly heterogeneous system. Meanwhile, the observations here reinforced our conclusion drawn in previous studies in non-fractured system that the so-called “optimum partition coefficient” was case

13

ACCEPTED MANUSCRIPT

Nomenclature

EP

TE D

M AN U

SC

RI PT

dependent. Our results suggested that one can tailor a surfactant with suitable solubility in the CO2 phase to optimize oil recovery according to reservoir conditions and injection strategy. The observations reported here extended our visions of CO2 soluble surfactant foam application to fractured system, which gave insights to this emerging research area. We believed the results were encouraging enough to warrant further R&D and eventual field piloting. The findings and observations of foam performance in fractured carbonate reservoir with those novel CO2 soluble surfactants as well as a conventional aqueous soluble foaming agent were summarized as followings, • The tested Silurian Carbonate was mixed wet or slightly preferential oil wet, varying with positions. However, it was observed that foam generation and propagation was not severely prohibited by this unfavorable rock condition, especially for employed CO2 soluble surfactants. • None of tested surfactant displayed typical Winsor type phase behavior. A clearly separate phase with lighter color and slightly higher viscosity presented at each tested brine salinity when CD 1045 was employed, which was almost negligible for both novel surfactants. • From bulk test, both CD 1045 and 15S have good and similar foam stability with presence of oil. 15S demonstrated weaker oil tolerance qualitatively. However, the oil tolerance result of CD1045 here may not include the effect of junk phase due to less mixing relative to phase behavior test. • In the 1ft core non-fractured matrix, pure CO2 flooding gave 61% oil recovery, which revealed that sweep efficiency was poor even though MCM condition has been satisfied. • With presence of water in the injection stream with variable gas volume fractions, pressure drops were still low and oil recoveries were only enhanced slightly. • With present of aqueous soluble surfactant, oil recovery was enhanced dramatically with improved mobility control indicated by higher pressure drop, which was attributed mainly by foam and partially by small amount of additional higher viscosity phase. • Novel CO2 soluble surfactants always outperformed aqueous soluble only surfactant, which may be attributed to promoted surfactant transportation, less surfactant adsorption and less surfactant consumption as absence of junk phase. Also, from pressure drop comparison, CO2 soluble surfactant foam may hold much better stability under reservoir conditions owing to improved phase dispersion caused by double affinity of CO2 soluble surfactant to aqueous and CO2 phases. The hypotheses here deserved further validations. • Opposite liquid desaturation rates of novel surfactants were observed here comparing with those from prior study, i.e., the higher partition coefficient surfactant foam gave faster liquid recovery rate and higher pressure drop. It may be combination effects of oil tolerance, impaired surfactant spreading, distinct foamability, stability and foam rheology. Those hypotheses can not be derived directly from current coreflooding and deserve separate fundamental studies. • Higher foam quality gave better performance with fixed surfactant solution injection rate, which may be attributed to combination effect of foam quality and increased total injection rate. • The results observed here reinforce our previous conclusion that the so-called optimal surfactant partition coefficient of CO2 soluble surfactant is case dependent and is the function of injection strategy, reservoir environment, and operation pressure or rates as well as other specific conditions.

AC C

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 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53

CMC critical micelle concentration Csc critical surfactant concentration for foam generation kf fracture permeability 15S S So WIw WIo w

14

CO2 soluble surfactant, 2EH-PO5-EO9 CO2 soluble surfactant, 2EH-PO5-EO15 Oil saturation Amott Index of water Amott Index of oil fracture aperture

ACCEPTED MANUSCRIPT

Acknowledgement The authors would like to thank Hang Zhang (University of Texas at Austin, currently at Schlumberger Houston) for his dedicative help on laboratory experiments.

Appendix A: Concept model for “optimal partition coefficient” in non-fractured system

M AN U

SC

RI PT

To explicitly interpret our laboratory observations in previous publication (Ren et al. 2013) in non-fractured system without presence of oil, a concept study was conducted here to reveal the spreading effect due to surfactant solubility in CO2 and foam performance with variable partition coefficients. With higher partition coefficient, which resulted to more surfactant extracted and delivered by CO2 corresponding to higher surfactant propagation rate, lower surfactant concentration was expected in the same location under constant rate injection. We demonstrated this distinct surfactant transportation owing to variant partition coefficients with a 1 foot cylindrical model built with CMG/STARS without foam effect, which was divided into 50 grids evenly. The k values of S and 15S were applied in two comparison simulations. Here, we applied 3 times higher adsorption of 15S than that of S, which was supported from experiment approximately. Other Parameters, including injection rates, surfactant concentration, and rock permeability were not necessary to be consistent with those in the core flooding because it was prediction and comparison, but not history matching, were the goal, as long as they were identical for both surfactants. Prior to address the details, we prove a premise of following discussion. Like mentioned in previous paper, theoretically, we should supervise the concentration in whole cell (global) instead of in water phase (water) because surfactant will act at the interface regardless of its partition in gas phase, which was another unique consideration for CO2 soluble surfactant. However, it was not necessary for the purpose here. Actually, the high consistence of variation trend between global and water concentrations were ensured by the injection fluids mass conservation and constant partition coefficients, as shown in Fig.A1a and b. Absolute performance of individual surfactant may be affected slightly but the compassion result between surfactants will not be altered. Therefore, it offered the escape from dilemma of model effect. The area below the curves were equivalent to amount of injected surfactants qualitatively before surfactant broke through, for constant rate injection. In turn, it could indicate the surfactant distribution in the core approximately.

EP

TE D

Fig.A1b displayed the variation of surfactant concentration in aqueous phase along the core at 1 PV and 20 PV for S and 15S. Initially, near the injector, with smaller partition coefficient and lower adsorption, S displayed the trend with much steeper slope and higher surfactant concentration. Correspondingly, lower surfactant concentration was expected for 15S at same location due to surfactant spreading and adsorption. As shown in plot, the critical surfactant concentration (Csc), at 1 IPV, crossing with surfactant distribution curve of S indicated the presence of strong foam, while there was no foam for 15S since its local concentration was lower than Csc. Then, with more injection, surfactant concentrations raised as well as the propagation of crossing points. At certain time and location, 15S will display foam effect but its apparent foam propagation was still behind that of S. This scenario will recommend S as optimal CO2 soluble surfactant. In later stage, with continue injection and surfactant propagation, the crossing point of 15S will be ahead of that of S, as shown in Fig.A1b at 20 IPV, which indicated 15S displayed faster foam propagation for that scenario. This resulted from more surfactant extraction by CO2 even with higher adsorption. Therefore, it was possible that surfactant partition ability could overwhelm the effect of adsorption to dominate the foam propagation. The schematic plot was shown in Fig.A2 to represent above interpretations further. As a consequence, two scenarios may exist during the process and must appear in sequence. Any direct or indirect manipulations to lower Csc will promote faster attainment of 2nd scenario. However, it is not necessary those two scenarios must show up for a specific case. For example, any of higher injection rate, short core, too high surfactant partition coefficient or higher surfactant critical value, could mask or delay the presence of 2nd scenario. Hence, the two scenarios gave opposite result with respect to foam propagation. Hence, the so-called “optimal partition coefficient” is function of system (reservoir), fluid properties, injection conditions and injection strategy.

AC C

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 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55

Appendix B: “Optimal surfactant partition coefficient” in field scale In our previous publication (Ren et al. 2013), with presence of gravity in homogenous 2D reservoir scale, we pointed out the so-called optimal surfactant partition coefficient was case dependent as the function of injection strategy, reservoir environment, and operation pressure or rates as well as other specific conditions. More simulations as shown in Fig.B1 and B2 were conducted to reinforce the statement. The details of model and

15

ACCEPTED MANUSCRIPT

SC

RI PT

injection strategy design could be found in original literature. The model setting and symbol designations were consistent, in which “partition coefficient value” represented the times of artificial CO2 soluble surfactant partition coefficient to S. CO2 storage was equivalent to foam sweep efficiency or liquid production efficiency. For alternating injection with 36.5 day slug size (SAG), S was still in the leading position and the satisfying candidates would be from 0.05S to 5S. With slug size increasing, this suitable selection window would narrow down owing to aggravating extraction effect. Similar observations were obtained for SISI (simultaneous injection through same interval). Then, partition coefficient effect was also tested for SIDI (simultaneous injection through different interval) and 4S was in the leading position. Good sweep efficiency could e achieved from 0.05S to 15S. Especially, the variance of sweep efficiency was as low as 0.0072 from 0.1S to 15S, which displayed the advantage of this injection strategy.

Reference

EP

TE D

M AN U

Aarra, M. G., Ormehaug, P. A., Skauge, A., & Masalmeh, S. K. (2011, January 1). Experimental Study of CO2- and Methane- Foam Using Carbonate Core Material at Reservoir Conditions. Society of Petroleum Engineers. doi:10.2118/141614-MS. Abbaszadeh, M., Rodriguez De La Garza, F., Yuan, C., & Pope, G. A. (2010, January 1). Single-Well Simulation Study of Foam EOR in GasCap Oil of the Naturally-Fractured Cantarell Field. Society of Petroleum Engineers. doi:10.2118/129867-MS. Abbaszadeh, M., & Ren, G. (2013, September 16). Simulation Techniques for Surfactant Phase Behavior and Foam Flow Modeling in Fractured Reservoirs. Society of Petroleum Engineers. doi:10.2118/166032-MS. Adkins, S.S., Chen, X., Chan, I., Torino, E., Nguyen, Q.P., Sanders, A.W., Johnston, P. Keith. Morphology and stability of CO2-in-water foams with nonionic hydrocarbon surfactants. Langmuir 26 (2010a) 5335–5348. Adkins, S.S., Chen, X., Nguyen, Q.P., Sanders, A.W. and Johnston, K.P. 2010b. Effect of Branching on the Interfacial Properties of Nonionic Hydrocarbon Surfactants at the Air–Water and Carbon Dioxide–Water Interfaces. J. Colloid and Interface Science 346 (2): 455–463. Akbar, M., Vissapragada, B., Alghamdi, A.H., and Allem, D. A Snapshot of Carbonate Reservoir Evaluation. Oilfoeld Review, 12, n4 (winter 2000-2001): 20-21. Alavian, S. A., & Whitson, C. H. (2010, January 1). Scale Dependence of Diffusion in Naturally Fractured Reservoirs for CO2 Injection. Society of Petroleum Engineers. doi:10.2118/129666-MS. Alkan, H., Goktekin, A., & Satman, A. (1991, January 1). A Laboratory Study of CO2-Foam Process for Bati Raman Field, Turkey. Society of Petroleum Engineers. doi:10.2118/21409-MS. Allan, J., & Sun, S. Q. (2003, January 1). Controls on Recovery Factor in Fractured Reservoirs: Lessons Learned from 100 Fractured Fields. Society of Petroleum Engineers. doi:10.2118/84590-MS. AlMaqbali, A., Agada, S., Geiger, S., Haugen, Å., & Fernø, M. A. (2015, November 9). Modelling Foam Displacement in Fractured Carbonate Reservoirs. Society of Petroleum Engineers. doi:10.2118/177600-MS. Almaqbali, A., Spooner, V. E., Geiger, S., Arnold, D., & Mackay, E. (2017, February 20). Uncertainty Quantification for Foam Flooding in Fractured Carbonate Reservoirs. Society of Petroleum Engineers. doi:10.2118/182669-MS. Alshehri, A. J., & Kovscek, A. R. (2014, December 10). Experimental and Numerical Study of Gravity Effects on Oil Recovery in Fractured Carbonates. International Petroleum Technology Conference. doi:10.2523/IPTC-18213-MS. Aroonsri, A., Worthen, A. J., Hariz, T., Johnston, K. P., Huh, C., & Bryant, S. L. (2013, September 30). Conditions for Generating \ Nanoparticle-Stabilized CO2 Foams in Fracture and Matrix Flow. Society of Petroleum Engineers. doi:10.2118/166319-MS. Baily, R.E.: “Report of the National Petroleum Council, Committee on EOR”, Enhanced Oil Recovery (June 1984). Benson, D.J., Mancini, E.A., Groshong, R.H., Fang, J.H., and Carlson, E.S. (1998). Alabama Carbonate Reservoir Characterization and Management: Key to Improved Oil Recovery at Appleton Oil Field, Escambia County, Alabama. Extended Abstracts Book A57 Vol. 1. AAPG Annual Convention, Salt Lake City, Utah, 17–20 May. Bernard, G. G., & Jacobs, W. L. (1965, December 1). Effect of Foam on Trapped Gas Saturation and on Permeability of Porous Media to Water. Society of Petroleum Engineers. doi:10.2118/1204-PA. Bertin, H. J., Apaydin, O. G., Castanier, L. M., & Kovscek, A. R. (1999, June 1). Foam Flow in Heterogeneous Porous Media: Effect of Cross Flow. Society of Petroleum Engineers. doi:10.2118/56009-PA. Boerrigter, P. M., Verlaan, M., & Yang, D. (2007, January 1). EOR Methods To Enhance Gas Oil Gravity Drainage. Society of Petroleum Engineers. doi:10.2118/111403-MS. Bourbiaux, B., Rosenberg, E., Robin, M., Chabert, M., Chevallier, E., & Gautier, S. (2017, June 1). Computed-Tomography-Scan Monitoring of Foam-Based Chemical-Enhanced-Oil-Recovery Processes in Fractured Carbonate Cores. Society of Petroleum Engineers. doi:10.2118/179811-PA. Bourrel M., Schechter R. S. Microemulsions and Related Systems, Marcel Dekker (1988). Buchgraber, M., Castanier, L. M., & Kovscek, A. R. (2012, January 1). Microvisual Investigation of Foam Flow in Ideal Fractures: Role of Fracture Aperture and Surface Roughness. Society of Petroleum Engineers. doi:10.2118/159430-MS. Casteel, J. F., & Djabbarah, N. F. (1988, November 1). Sweep Improvement in CO2 Flooding by Use of Foaming Agents. Society of Petroleum Engineers. doi:10.2118/14392-PA. Chabert, M., Nabzar, L., Beunat, V., Lacombe, E., & Cuenca, A. (2014, April 12). Impact of Surfactant Structure and Oil Saturation on the Behavior of Dense CO2 Foams in Porous Media. Society of Petroleum Engineers. doi:10.2118/169116-MS. Chahardowli, M., Zholdybayeva, A., Farajzadeh, R., & Bruining, H. (2013, June 10). Solvent-enhanced Spontaneous Imbibition in Fractured Reservoirs. Society of Petroleum Engineers. doi:10.2118/164908-MS.

AC C

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 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64

16

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

Chambers, K.T. and Radke, C.J. (1991). Capillary Phenomena in Foam Flow through Porous Media. In: N.R Morrow (Editor), Interfacial Phenomena in Petroleum Recovery. Marcel Dekker Inc., New York, N.Y., pp. 191-255. Chen, X., Adkins, S., Sanders. A., Nguyen, Q. and Johnston, K. 2010. Interfacial Tension and the Behavior of Microemulsions and Macroemulsions of Water and Carbon Dioxide with a Branched Hydrocarbon Nonionic Surfactant. J. of Supercritical Fluids 55(2): 712723. Chen, Y., Elhag, A. S., Poon, B. M., Cui, L., Ma, K., Liao, S. Y., Johnston, K. P. (2012, January 1). Ethoxylated Cationic Surfactants For CO2 EOR In High Temperature, High Salinity Reservoirs. Society of Petroleum Engineers. doi:10.2118/154222-MS. Chen Y., Elhag Amro S., Cui Leyu, Worthen Andrew J., Reddy Prathima P., Noguera Jose A., Anne Marie Ou, Ma Kun, Puerto Maura, Hirasaki George J., Nguyen P. Quoc, Biswal Sibani L., and Keith P. Johnston Industrial & Engineering Chemistry Research 2015 54 (16), 42524263 DOI: 10.1021/ie503674m. Chen, Zehua , Zhao, Xiutai . Enhancing Heavy-Oil Recovery by Using Middle Carbon Alcohol-Enhanced Waterflooding, Surfactant Flooding, and Foam Flooding. Energy Fuels, 2015, 29 (4), pp 2153–2161. Chen, P., & Mohanty, K. K. (2014, April 12). Wettability Alteration in High Temperature Carbonate Reservoirs. Society of Petroleum Engineers. doi:10.2118/169125-MS. Chen, P., & Mohanty, K. K. (2015, April 13). Surfactant-Enhanced Oil Recovery from Fractured Oil-wet Carbonates: Effects of Low IFT and Wettability Alteration. Society of Petroleum Engineers. doi:10.2118/173797-MS. Chevallier, E., Tchamba, O., Chabert, M., Bekri, S., Martin, F., & Gautier, S. (2015, August 11). Foams with Ultra-Low Interfacial Tensions for an Efficient EOR Process in Fractured Reservoirs. Society of Petroleum Engineers. doi:10.2118/174658-MS. Chevallier, E., Chabert, M., Gautier, S., Ghafram, H., Khaburi, S., & Alkindi, A. (2018, March 26). Design of a Combined Foam EOR Process for a Naturally Fractured Reservoir. Society of Petroleum Engineers. doi:10.2118/190363-MS. Chillenger, G.V. and Yen, T.F (1983). Some Notes on Wettability and Relative Permeability of Carbonate Rocks. II. Energy and Sources, 7, 6775. Conn, C.; Ma, K.; Hirasaki, G.J.; Biswal, S.L., Visualizing oil displacement with foam in a microfluidic device with permeability contrasts. Lab on a Chip 2014, 14, 3968-3977 Consani, K.A., Smith, R.D. Observations on the solubility of surfactants andrelated molecules in carbon dioxide at 50◦C. J. Supercritical Fluids 3 (1990)51–56. Denney, D. (2008, July 1). History Matching Thermally Assisted Gas/Oil Gravity Drainage In Fractured Reservoirs. Society of Petroleum Engineers. doi:10.2118/0708-0051-JPT. Do. H.D. and Pinczewski, W.V. (1991). Diffusion Controlled Swelling of Reservoir Oil by Direct Contact with Injection Gas. Chemical Engineering Science, 46 (5-6): 1259-1270. Dong, P., Puerto, M., Ma, K., Mateen, K., Ren, G., Bourdarot, Hirasaki, G. (2017, April 3). Low-Interfacial-Tension Foaming System for Enhanced Oil Recovery in Highly Heterogeneous/Fractured Carbonate Reservoirs. Society of Petroleum Engineers. doi:10.2118/184569MS. Dong, P., Puerto, M., Ma, K., Mateen, K., Ren, G., Bourdarot, Hirasaki, G. (2018, April 14). Ultralow-Interfacial-Tension Foam Injection Strategy Investigation in High Temperature Ultra-High Salinity Fractured Carbonate Reservoirs. SPE 190259-MS, SPE Improved Oil Recovery Conference, Tulsa, Oklahoma, USA, 14-18 April 2018. Eide, O., Ersland, G., Brattekas, B., Haugen, A., Graue, A., & Ferno, M. A. (2015, February 1). CO2 EOR by Diffusive Mixing in Fractured Reservoirs. Society of Petrophysicists and Well-Log Analysts. Elhag, A. S., Chen, Y., Chen, H., Reddy, P. P., Cui, L., Worthen, A. J., Johnston, K. P. (2014, April 12). Switchable Amine Surfactants for Stable CO2/Brine Foams in High Temperature, High Salinity Reservoirs. Society of Petroleum Engineers. doi:10.2118/169041-MS. Elsharkawy, A. M., Poettmann, F. H., & Christiansen, R. L. (1992, January 1). Measuring Minimum Miscibility Pressure: Slim-Tube or RisingBubble Method? Society of Petroleum Engineers. doi:10.2118/24114-MS. Falls, A. H., Hirasaki, G. J., Patzek, T. W., Gauglitz, D. A., Miller, D. D., & Ratulowski, T. (1988, August 1). Development of a Mechanistic Foam Simulator: The Population Balance and Generation by Snap-Off. Society of Petroleum Engineers. doi:10.2118/14961-PA. Falls, A. H., Musters, J. J., & Ratulowski, J. (1989, May 1). The Apparent Viscosity of Foams in Homogeneous Bead Packs. Society of Petroleum Engineers. doi:10.2118/16048-PA. Fan, X., Potluri Vijay K., M. Chandler McLeod, Y. Wang. Oxygenated hydrocarbon ionic surfactants exhibit CO2 solubility. J. American Chemical Society 127(2005) 11754–11762. Farajzadeh, R.B., Barati, A., Delil, H.A., Burning, J., Zitha, P.L.J. Mass Transfer of CO2 into Water and Surfactant Solutions. Journal of Petreoleum Science and Technology, Vol.25, Iss.12, 2007. Farajzadeh, R., Wassing, B.M., Boerrigter, P.M.. Foam assisted gas–oil gravity drainage in naturally-fractured reservoirs. Journal of Petroleum Science and Engineering 94-95 (2012) 112-122. Fernø, M. A., Gauteplass, J., Pancharoen, M., Haugen, A., Graue, A., Kovscek, A. R., & Hirasaki, G. J. (2014, October 27). Experimental Study of Foam Generation, Sweep Efficiency and Flow in a Fracture Network. Society of Petroleum Engineers. doi:10.2118/170840-MS. Fernø, M.A. , Eide, Ø., Steinsbø, M. , Langlo, S.A.W., Christophersen, A., Skibenes, A., Ydstebø, T., Graue, A., Mobility control during CO2 EOR in fractured carbonates using foam: Laboratory evaluation and numerical simulations. Journal of Petroleum Science and Engineering 135 (2015) 442-451. Fjelde, I., Zuta, J., & Duyilemi, O. V. (2008, January 1). Oil Recovery from Matrix during CO2-Foam Flooding of Fractured Carbonate Oil Reservoirs. Society of Petroleum Engineers. doi:10.2118/113880-MS. Flock, D. L., & Nouar, A. (1984, September 1). Parametric Analysis On The Determination Of The Minimum Miscibility Pressure In Slim Tube Displacements. Petroleum Society of Canada. doi:10.2118/84-05-12. Fredriksen, S.B., Alcorn, Z.P., Froland A., Viken, A., Rognmo, A.U., Seland, J.G., Ersland, G., Ferno, M.A., Graue, A. Surfactant Pre-floods during CO2 Foam for integrated Enhanced Oil Recovery in Fractured Oil-wet Carbonates. SPE 190168-MS, SPE Improved Oil Recovery Conference, Tulsa, Oklahoma, USA, 14-18 April 2018. French, T. R., Broz, J. S., Lorenz, P. B., & Bertus, K. M. (1986, January 1). Use of Emulsions for Mobility Control During Steamflooding. Society of Petroleum Engineers. doi:10.2118/15052-MS. Friedmann, F., Chen, W. H., & Gauglitz, P. A. (1991, February 1). Experimental and Simulation Study of High-Temperature Foam Displacement in Porous Media. Society of Petroleum Engineers. doi:10.2118/17357-PA. Garrett , P.R. (ed.). 1993. Defoaming, Marcel Dekker, New York. Gauteplass Jarand, Chaudhary Kuldeep, Kovscek R. Anthony, Fernø A. Martin. Pore-level foam generation and flow for mobility control in

AC C

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 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70

17

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

fractured systems. Colloids and Surfaces A: Physicochem. Eng. Aspects 468 (2015) 184-192. Graue, A., Moe, R.W., and Bogno, T. (2000). Impacts of Wettability on Oil Recovery in Fractured Carbonate Reservoirs. Paper presented at the Bergen University Society Core Annual International Symposium, Abu Dhabi, UAE, 18–22 October. SCA 2000-25. Grigg, R. B., Tsau, J.-S., & Martin, F. D. (2002, January 1). Cost Reduction and Injectivity Improvements for CO2 Foams for Mobility Control. Society of Petroleum Engineers. doi:10.2118/75178-MS. Grogan, A. T., & Pinczewski, W. V. (1987, May 1). The Role of Molecular Diffusion Processes in Tertiary CO2 Flooding. Society of Petroleum Engineers. doi:10.2118/12706-PA. Gupta, R., & Mohanty, K. K. (2008, January 1). Wettability Alteration of Fractured Carbonate Reservoirs. Society of Petroleum Engineers. doi:10.2118/113407-MS. Haugen, Å., Fernø, M. A., Graue, A., & Bertin, H. J. (2012, April 1). Experimental Study of Foam Flow in Fractured Oil-Wet Limestone for Enhanced Oil Recovery. Society of Petroleum Engineers. doi:10.2118/129763-PA. Haugen Ãsmund, Mani Nima, Svenningsen Sondre, Brattekas, Bergit, Graue Arne, Ersland Geir, Fernø Martin A. Miscible and Immiscible Foam Injection for Mobility Control and EOR in Fractured Oil-Wet Carbonate Rocks. Transport in Porous Media 2014, 104 (1) 109-131. Hirasaki, G. J., & Lawson, J. B. (1985, April 1). Mechanisms of Foam Flow in Porous Media: Apparent Viscosity in Smooth Capillaries. Society of Petroleum Engineers. doi:10.2118/12129-PA. Irani, C. A., & Solomon, C. (1986, January 1). Slim-Tube Investigation Of CO2 Foams. Society of Petroleum Engineers. doi:10.2118/14962-MS. Jimenez. A.I. and Radke C.J. Dynamic Stability of Foam Lamellae Flowing Through a Periodically Constricted Pore. Chapter 25, pp 460-479, ACS Symposium Series, Vol. 396, DOI: 10.1021/bk-1989-0396.ch025. Johnston, K.P., Randolph, T.W., Bright, F.V., Howdle, S.M., Toxicology of a PFPE surfactant. Science 272 (1996) 1726. Kahrobaei, S., Farajzadeh, R., Suicmez, V. S., & Bruining, J. (2012, January 1). Gravity-Enhanced Transfer between Fracture and Matrix in Solvent-Based Enhanced Oil Recovery. Society of Petroleum Engineers. doi:10.2118/154171-MS. Kamali, F., Khosravi, M., Vahidi, A., & Roayaei, E. (2011, January 1). Miscible GOGD Using CO2 in One of the Iranian Fractured Reservoirs – A Case Study. Society of Petroleum Engineers. doi:10.2118/144935-MS. Karimaie, H., Torsater, O. Low IFT gas-oil gravity drainage in fractured carbonate porous media. Journal of Petroleum Science and Engineering 70 (2010b) 67-73. Karimaie, H., & Torsaeter, O. (2010a, January 1). CO2 and C1 Gas Injection for Enhanced Oil Recovery in Fractured Reservoirs. Society of Petroleum Engineers. doi:10.2118/139703-MS. Khatib, Z. I., Hirasaki, G. J., & Falls, A. H. (1988, August 1). Effects of Capillary Pressure on Coalescence and Phase Mobilities in Foams Flowing Through Porous Media. Society of Petroleum Engineers. doi:10.2118/15442-PA. Kovscek, A., Radke, C. 1994. Fundamentals of foam transport in porous media. In Foams: Fundamentals and Applications in the Petroleum Industry, 115-163. Washington DC, American Chemical Society. Kovscek, A., Tretheway, D., Radke, C., and Persoff, P. (1995). Foam Flow Through a Transparent Rough-Walled Rock Fracture. presented at the Bartlesville Project Office, U.S. Department of Energy, Bartlesville, Oklahoma, DOE/BC/95000168. Latta, T. M., Seiersten, M. E., & Bufton, S. A. (2013, May 6). Flow Assurance Impacts on Lean/Rich MEG Circuit Chemistry and MEG Regenerator/Reclaimer Design. Offshore Technology Conference. doi:10.4043/24177-MS. Lau, H. C., & O'Brien, S. M. (1988, August 1). Effects of Spreading and Nonspreading Oils on Foam Propagation Through Porous Media. Society of Petroleum Engineers. doi:10.2118/15668-PA. Laurier L. Schramm. Foams: Fundamentals and Applications in the Petroleum Industry. 1994 American Chemical Society DOI: 10.1021/ba1994-0242. Le, V. Q., Nguyen, Q. P., & Sanders, A. (2008, January 1). A Novel Foam Concept With CO2 Dissolved Surfactants. Society of Petroleum Engineers. doi:10.2118/113370-MS. Le Gallo, Y., Le Romancer, J. F., Bourbiaux, B., & Fernandes, G. (1997, January 1). Mass Transfer in Fractured Reservoirs during Gas Injection: Experimental and Numerical Modeling. Society of Petroleum Engineers. doi:10.2118/38924-MS. Lescure, B. M., & Claridge, E. L. (1986, January 1). CO2 Foam Flooding Performance vs. Rock Wettability. Society of Petroleum Engineers. doi:10.2118/15445-MS. Li, R. F., Yan, W., Liu, S., Hirasaki, G., & Miller, C. A. (2010, December 1). Foam Mobility Control for Surfactant Enhanced Oil Recovery. Society of Petroleum Engineers. doi:10.2118/113910-PA. Li, R. F., Hirasaki, G., Miller, C. A., & Masalmeh, S. K. (2012, December 1). Wettability Alteration and Foam Mobility Control in a Layered, 2D Heterogeneous Sandpack. Society of Petroleum Engineers. doi:10.2118/141462-PA. Liu, J., Han, B., Li, G., Zhang, X. Investigation of nonionic surfactant Dynol-604based reverse microemulsions formed in supercritical carbon dioxide. Langmuir 17 (2001) 8040–8043. Liu, Y., Grigg, R. B., & Bai, B. (2005, January 1). Salinity, pH , and Surfactant Concentration Effects on CO2-Foam. Society of Petroleum Engineers. doi:10.2118/93095-MS. Liu Pengcheng, Zhang Xiaokun, Wu Yongbin, Li Xiuluan. Enhanced oil recovery by air foam flooding system in tight oil reservoirs: Study on the profile-controlling mechanisms. Journal of Petroleum Science and Engineering 150 (2017) 208-216. Lopera Castro, S. H., Restrepo, A., & Ocampo, A. (2009, January 1). Use of Divergent Fluids as an Alternative For Enhanced Recovery in Naturally Fractured Cores. Society of Petroleum Engineers. doi:10.2118/122061-MS. Ma Kun, Marten Khalid, Ren Guangwei (2018). Modeling Foam Flow at Achievable Flow Rates in the Subterranean Formation Using the Population-balance Approach and Implications for Experimental Design. Journal of Non-Newtonian Fluid Mechanics (in press). https://doi.org/10.1016/j.jnnfm.2018.02.007. Mannhardt, K. (1999, December 31). Core Flood Evaluation of Solvent Compositional And Wettability Effects On Hydrocarbon Solvent Foam Performance. Petroleum Society of Canada. doi:10.2118/99-13-21. McLendon W.J., Koronaios P., Enick R.M., Biesmans G., Salazar L., Miller A., Soong Y., McLendon T., Romanov V., Crandall D., Assessment of CO2-soluble non-ionic surfactants for mobility reduction using mobility measurements and CT imaging. Journal of Petroleum Science and Engineering, 119 (2014) 196-209. Morel, D., Bourbiaux, B., Latil, M., & Thiebot, B. (1993, July 1). Diffusion Effects in Gasflooded Light-Oil Fractured Reservoirs. Society of Petroleum Engineers. doi:10.2118/20516-PA. Mukherjee, J., Norris, S. O., Nguyen, Q. P., Scherlin, J. M., Vanderwal, P. G., & Abbas, S. (2014, April 12). CO2 Foam Pilot in Salt Creek Field, Natrona County, WY: Phase I: Laboratory Work, Reservoir Simulation, and Initial Design. Society of Petroleum Engineers. doi:10.2118/169166-MS.

AC C

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 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70

18

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

Nguyen, Q. P., Alexandrov, A. V., Zitha, P. L., & Currie, P. K. (2000, January 1). Experimental and Modeling Studies on Foam in Porous Media: A Review. Society of Petroleum Engineers. doi:10.2118/58799-MS . Nguyen, N., Ren, G., Mateen, K., Cordelier, P. R., Morel, D. C., & Nguyen, Q. P. (2015, August 11). Low-Tension Gas (LTG) Injection Strategy in High Salinity and High Temperature Sandstone Reservoirs. Society of Petroleum Engineers. doi:10.2118/174690-MS. Norris, S. O., Scherlin, J. M., Mukherjee, J., Vanderwaal, P., Abbas, S., & Nguyen, Q. P. (2014, October 27). CO2 Foam Pilot in Salt Creek Field, Natrona County, WY: Phase II: Diagnostic Testing and Initial Results. Society of Petroleum Engineers. doi:10.2118/170729-MS. Ocampo-Florez, A., Restrepo, A., Rendon, N., Coronado, J., Correa, J. A., Ramirez, D. A., Lopera, S. H. (2014, December 10). Foams Prove Effectiveness for Gas Injection Conformance and Sweep Efficiency Improvement in a Low porosity Fractured Reservoir – Field Pilots. International Petroleum Technology Conference. doi:10.2523/17950-MS. Osei-Bonsu Kofi, Grassia Paul, Shokri Nima. Relationship between bulk foam stability, surfactant formulation and oil displacement efficiency in porous media. Fuel 203 (2017) 403-410. Osman G. Apaydin, Anthony R. Kovscek. (2000, October). Surfactant concentration and end effects on foam flow in porous media. DE-FC2600BC15311. Osterloh, W. T., & Jante, M. J. (1992, January 1). Effects of Gas and Liquid Velocity on Steady-State Foam Flow at High Temperature. Society of Petroleum Engineers. doi:10.2118/24179-MS. Panahi, H. (2004, January 1). Improving the Recovery Factor of Heavy Crude Reservoirs by Co-injecting CO2 and Other Conventional Gaseous Injecting Materials at Immiscibility Condition with Foam. Society of Petroleum Engineers. doi:10.2118/92011-MS. Pancharoen, M., Fernø, M. A., Kovscek, A. R., Modeling foam displacement in fractures, Journal of Petroleum Science and Engineering, Vol 100, December 2012, Pages 50-58, ISSN 0920-4105. Perkins, T. K., & Johnston, O. C. (1963, March 1). A Review of Diffusion and Dispersion in Porous Media. Society of Petroleum Engineers. doi:10.2118/480-PA . Randall, T. E., & Bennion, D. B. (1988, November 1). Recent Developments In Slim Tube Testing For Hydrocarbon-Miscible Flood (Hcmf) Solvent Design. Petroleum Society of Canada. doi:10.2118/88-06-02. Prieditis, J., & Paulett, G. S. (1992, January 1). CO2-Foam Mobility Tests at Reservoir Conditions in San Andres Cores. Society of Petroleum Engineers. doi:10.2118/24178-MS. Princen, H. M. and Goddard, E.D. The Effect of Mineral Oil on the Surface Properties of Binary Surfactant Systems. Journal of Colloid and Interface Science, Vol.38, No.2, February 1972. Qu, M., Hou, J., Wang, Q., Su, W., Ma, S., Yang, T., … Bai, Y. (2017, November 13). Modified Starch Gel Foam: Research and Application in Naturally Fractured-Vuggy Carbonate Reservoir. Society of Petroleum Engineers. doi:10.2118/188812-MS. Ren, G., Zhang, H., & Nguyen, P. Q. (2013, May 19). Effect of Surfactant Partitioning on Mobility Control During Carbon-Dioxide Flooding. Society of Petroleum Engineers. doi:10.2118/145102-PA. Ren, G., Nguyen, Quoc P. Understanding of Aqueous Foam with Novel CO2 Soluble Surfactants for Controlling CO2 Vertical Sweep in Sandstone Reservoirs. Pet. Sci. (2017) 14:330-361. DOI 10.1007/s12182-0170149-2. Ren, G., Aaron W., and Nguyen, Q. P., (2014). New Method for the determination of surfactant solubility and partitioning between CO2 and Brine. J. of Supercritical Fluids 91 (214) 77-83. Rossen, W.R.: Foams in Enhanced Oil Recovery. Chapter 11 in Foams: Theory, Measurements and Applications, R.K. Prud’homme and S. Khan(eds.), Marcel Dekker, New York City (1996). Rutherford, W. M. (1962, December 1). Miscibility Relationships in the Displacement of Oil By Light Hydrocarbons. Society of Petroleum Engineers. doi:10.2118/449-PA. Sagir, M. , Tan, Isa M., Mushtaq, M., Ismail, L., Nadeem, M. & Azam, M. R. (2014a). Synthesis of a New CO2 Philic Surfactant for Enhanced Oil Recovery Applications. Journal of Dispersion Science and Technology, 35:5, 647-654, DOI:10.1080/01932691.2013.803253. Sagir, M., Tan, I. M., Mushtaq, M., & Talebian, S. H. (2014b, April 21). FAWAG Using CO2 Philic Surfactants for CO2 Mobility Control for Enhanced Oil Recovery Applications. Society of Petroleum Engineers. doi:10.2118/172189-MS. Sagir, M.,Tan, Isa M., Mushtaq, M., Ismail, L.,Nadeem, M., Azam, M. R.& Hashmet, M.R. (2014c). Novel Surfactant for the Reduction of CO2/Brine Interfacial Tension. Journal of Dispersion Science and Technology, 35:3, 463-470, DOI: 10.1080/01932691.2013.794111. Sagir, M.,Tan, Isa M., Mushtaq, M. & Nadeem, M. (2014d) CO2 Mobility and CO2/Brine Interfacial Tension Reduction by Using a New Surfactant for EOR Applications. Journal of Dispersion Science and Technology, 35:11, 1512-1519, DOI:10.1080/01932691.2013.859087. Sanchez, N. L. (1999, January 1). Management of Water Alternating Gas (WAG) Injection Projects. Society of Petroleum Engineers. doi:10.2118/53714-MS. Sanchez, J. M., & Hazlett, R. D. (1992, February 1). Foam Flow Through an Oil-Wet Porous Medium: A Laboratory Study. Society of Petroleum Engineers. doi:10.2118/19687-PA. Sanders, A., Nguyen, Q. P., Nguyen, N., Adkins, S., & Johnston, K. P. (2010, January 1). Twin-Tailed Surfactants for Creating CO2-in-Water Macroemulsions for Sweep Enhancement in CO2-EOR. Society of Petroleum Engineers. doi:10.2118/137689-MS. Sanders, A., Jones, R. M., Rabie, A., Putra, E., Linroth, M. A., & Nguyen, Q. P. (2012, January 1). Implementation of a CO2 Foam Pilot Study in the SACROC Field: Performance Evaluation. Society of Petroleum Engineers. doi:10.2118/160016-MS. Sarbu, T., Styranec, T., Beckman, E.J. Non-flourous polymers with very high solubility in supercritical CO2 down to low pressures. Nature 405 (2000) 165–168. Schramm, L. L., Mannhardt, K.. The effect of wettability on foam sensitivity to crude oil in porous media. Journal of Petroleum Science and Engineering 15 (1996) 101-113. Seong-Geun oh and Dinesh O. Shah. Relationship between Micelle Lifetime and foamability of Sodium Dodecyl Sulfate and Sodium Dodecyl Sulfate/1-Hexanol mixtures. Langmuir 1991, 7, 1316-1318. Seyed Alizadeh, S. M., Alizadeh, N., & Maini, B. B. (2007, January 1). A full-field Simulation Study of the Effect of Foam Injection on Recovery Factor of an Iranian Oil Reservoir. International Petroleum Technology Conference. doi:10.2523/IPTC-11181-MS. Skoreyko, F. A., Villavicencio, A. P., Rodriguez Prada, H., & Nguyen, Q. P. (2011, January 1). Development of a New Foam EOR Model From Laboratory and Field Data of the Naturally Fractured Cantarell Field. Society of Petroleum Engineers. doi:10.2118/145718-MS. Sloan, R. (2003, January 1). Quantification of Uncertainty in Recovery Efficiency Predictions: Lessons Learned from 250 Mature Carbonate Fields. Society of Petroleum Engineers. doi:10.2118/84459-MS Srivastava Mayank, Ph.D dissertation, 2010, University of Texas-at Austin. Steinsbø, M. , Brattekås, B. , Ersland, G. , K. Bø, Opdal, I.E. , Tunli, R. , Graue, A. and Fernø, M.A. (2015) . Foam as Mobility Control For Integrated CO2-EOR in Fractured Carbonates. 18th European Symposium on Improved Oil Recovery. April 14-2015. DOI: 10.3997/2214-

AC C

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 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70

19

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

4609.201412125. Su, W., Hou, J., Zhao, F., Xi, Y., Zhao, T., & Ding, B. (2017, November 13). Feasibility and Influencing Factors of Oil Tolerant Nitrogen Foam on EOR Effect in the Fractured-Vuggy Carbonate Reservoir. Society of Petroleum Engineers. doi:10.2118/188798-MS. Suffridge, F. E., Raterman, K. T., & Russell, G. C. (1989, January 1). Foam Performance Under Reservoir Conditions. Society of Petroleum Engineers. doi:10.2118/19691-MS. Suicmez, V. S., Karpan, V., & Dindoruk, B. (2011, January 1). Impact of Miscibility on Gas-Oil Gravity Drainage in Fractured Reservoirs. Society of Petroleum Engineers. doi:10.2118/147839-MS. Svorstol, I., Vassenden, F., & Mannhardt, K. (1996, January 1). Laboratory Studies for Design of a Foam Pilot in the Snorre Field. Society of Petroleum Engineers. doi:10.2118/35400-MS. Tabary, R., Fornari, A., Bazin, B., Bourbiaux, B. J., & Dalmazzone, C. S. H. (2009, January 1). Improved Oil Recovery With Chemicals in Fractured Carbonate Formations. Society of Petroleum Engineers. doi:10.2118/121668-MS. Telmadarreie, A., & Trivedi, J. (2015, June 9). New Insight on Carbonate Heavy Oil Recovery: Pore Scale Mechanisms of Solvent Alternating CO2 Foam/Polymer Enhanced Foam Flooding. Society of Petroleum Engineers. doi:10.2118/174510-MS. Thiebot, B. M., & Sakthikumar, S. S. (1991, January 1). Cycling Fractured Reservoirs Containing Volatile Oil: Laboratory Investigation of the Performance of Lean Gas or Nitrogen Injection. Society of Petroleum Engineers. doi:10.2118/21427-MS. Tiffin, D. L., Sebastian, H. M., & Bergman, D. F. (1991, May 1). Displacement Mechanism and Water Shielding Phenomena for a RichGas/Crude-Oil System. Society of Petroleum Engineers. doi:10.2118/17374-PA. Uleberg, K., & Høier, L. (2002, January 1). Miscible Gas Injection in Fractured Reservoirs. Society of Petroleum Engineers. doi:10.2118/75136MS. Wu, R. S., & Batycky, J. P. (1990, November 1). Evaluation Of Miscibility From Slim Tube Tests. Petroleum Society of Canada. doi:10.2118/90-06-06. Xiao Siyang, Zeng Yongchao, Vavra Eric D., He Peng, Puerto Maura, Hirasaki J.George, and Biswal L. Sibani. Destabilization, Propagation, and Generation of Surfactant-Stabilized Foam during Crude Oil Displacement in Heterogeneous Model Porous Media. Langmuir 2018 34 (3), 739-749. Xing, D., Wei, B., McLendon, W. J., Enick, R. M., McNulty, S., Trickett, K., Soong, Y. (2012, December 1). CO2-Soluble, Nonionic, WaterSoluble Surfactants That Stabilize CO2-in-Brine Foams. Society of Petroleum Engineers. doi:10.2118/129907-PA. Yan, W., Miller, C., and Hirasaki, G. 2006. Foam Sweep in Fractures for Enhanced Oil Recovery. Colloids and Surfaces A: Physicochem. Eng. Aspects: 282-283. Yellig, W. F., & Metcalfe, R. S. (1980, January 1). Determination and Prediction of CO2 Minimum Miscibility Pressures (includes associated paper 8876 ). Society of Petroleum Engineers. doi:10.2118/7477-PA. Za. Donaldson, E.C. and Tiab, D. 2003. Petrophysics, Second Edition: Theory and Practice of Measuring Reservoir Rock and Fluid transport Properties. Burlington, Massachusetts: Gulf Professional. Zeng Yongchao, Ma Kun, Farajzadeh Rouhi, Puerto Maura, Biswal Sibani L., Hirasaki George J.. Effect of Surfactant Partitioning Between Gaseous Phase and Aquoues Phase on CO2 Foam Transport for Enhanced Oil Recovery. Transp Porous Med. 2016 July, doi: 10.1007/s11242-016-0743-6. Zuta, J., Fjelde, I., and Berenblyum R. 2009. Oil Recovery During CO2-Foam in Fractured Chalk Rock at Reservoir Conditions. Proc., International Symposium of the Society of Core Analysts, Noordwijk, The Netherlands, 27-30 September. Zuta, J., & Fjelde, I. (2010, August 1). Transport of CO2-Foaming Agents During CO2-Foam Processes in Fractured Chalk Rock. Society of Petroleum Engineers. doi:10.2118/121253-PA. Zuta, J., Fjelde, I., Berenblyum, R., Vartdal, L. H., & Ovesen, H. (2010a, January 1). Modeling of Transport of a CO2-Foaming Agent During CO2-Foam Processes in a Fractured Chalk Rock. Society of Petroleum Engineers. doi:10.2118/129601-MS. Zuta, J., Fjelde, I., & Berenblyum, R. (2010b, January 1). Experimental And Simulation Of CO2-Foam Flooding In Fractured Chalk Rock At Reservoir Conditions: Effect Of Mode Of Injection On Oil Recovery. Society of Petroleum Engineers. doi:10.2118/129575-MS. Zuta, J., & Fjelde, I. (2011, January 1). Mechanistic Modeling of CO2-Foam Processes in Fractured Chalk Rock: Effect of Foam Strength and Gravity Forces on Oil recovery. Society of Petroleum Engineers. doi:10.2118/144807-MS.

AC C

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 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51

20

ACCEPTED MANUSCRIPT

Table.1 – Slim tube setups Here

Internal Diameter (inch)

0.12-0.63

0.3125

Length (feet)

5-120

60

Packing Material (Glass beads, sand, Mesh)

50-270

100

Porosity (%)

32-45

43

Permeability (Darcy)

2.5-250

22

Displacement velocity (ft/day)

30-650

53

RI PT

Literatures

6.5

0.3

6.9

M AN U

0.2

SC

Table.2 - The volume of fluid recorded during Amott test. Volume of brine Volume of oil Volume of brine displaced by forced displaced by displaced by forced displacement (cc) imbibition (cc) displacement (cc)

Volume of brine displaced by imbibition (cc)

WIw

WIo

0.042

0.03

Table.3 - Summary of employed core properties.

Surfactant

15S-75% S-75% 1045-75% W&G-75% 15S-50% S-50% 1045-50% W&G-50% Pure CO2

15S S 1045 N/A 15S S 1045 N/A N/A

0.75 0.75 0.75 0.75 0.5 0.5 0.5 0.5 1

Measured Matrix permeability (md) 159 160 143 167 157 158 163 161 132

Calculated Composite core permeability (md)

Measured Composite core permeability (md)

Porosity (fraction)

PV (cc)

Initial So (fraction)

944 945 928 952 942 943 948 946 917

1018 1023 967 1098 877 1008 1021 1017 899

0.212 0.165 0.165 0.17 0.196 0.197 0.165 0.169 0.17

105 82 82 84 82 97 82 84 84

0.48 0.49 0.46 0.52 0.47 0.48 0.47 0.5 0.51

EP

TE D

EXP

Injection gas fraction

Table.4 - Summary of initial oil saturation, final saturation and R factor. PV Initial So Final So R factor (cc) (fraction) (fraction) (fraction)

AC C

EXP

15S-75%

105

0.48

0.04

0.92

S-75%

82

0.49

0.14

0.71

1045-75%

82

0.46

0.21

0.54

W&G-75%

84

0.52

0.34

0.35

15S-50%

82

0.47

0.07

0.85

S-50%

97

0.48

0.17

0.64

1045-50%

82

0.47

0.23

0.52

W&G-50%

84

0.5

0.34

0.32

Pure CO2

84

0.51

0.39

0.24

RI PT

ACCEPTED MANUSCRIPT

TE D

M AN U

SC

Fig.1-Schematic diagram of slim tube experiment setup

Fig.2 – Imbibition cell

Pressure

EP

Surfactant /brine

Channel 1

Accumulator

AC C

ISCO Pump

Quizix Pump

Channel 2 Channel 3

Oil

Channel 4

CO2 Hydraulic Pump Quizix Pump

Channel 5

BPR Effluent

Heating Tape

Fractional Collector

ACCEPTED MANUSCRIPT

RI PT

Fig.3 – Schematic plot of coreflooding setup.

(a) (b) Fig.4 – Core preparation: (a) Split core with Teflon; (b) Fixed core with tape.

SC

0.8 0.7 0.6 0.5 0.4

M AN U

Recovery Factor

1 0.9

MMP

0.3 0.2 0.1 0 0

500

1000

1500

2000

2500

Pressure (psi)

EP

TE D

Fig.5 – Slim tube results

2

AC C

1

3 4 Fig.6 – Qualitative demonstration of fluids contact angle: (1) Wasson oil and brine on a virgin core; (2) Brine and oil at same spots of picture 1; (3) Wasson oil and brine on another virgin core; (4) Brine and oil at same spots of picture 3.

RI PT

ACCEPTED MANUSCRIPT

M AN U

(b)

SC

(a)

(c)

AC C

EP

TE D

Fig.7 - Phase behavior results of oil and surfactant solution at variable salinities (@35oC), from left to right as 1wt% to 15wt% in 2wt% increment: (a) CD-1045, (b) S (2EH-PO5-EO15), (c) 15S (2EH-PO5-EO9).

Fig.8 – Foam stability with presence of oil.

RI PT

ACCEPTED MANUSCRIPT

TE D

M AN U

SC

Fig.9 – Recoveries of pure CO2 flooding in the systems with and without fracture.

AC C

EP

Fig.10 – Pressure drops and recoveries of brine/CO2 co-injection (75% & 50%) and pure CO2 flooding.

Fig.11 - Pressure drops and recoveries of co-injection CO2 and CD1045 surfactant solution (75%&50%).

RI PT

ACCEPTED MANUSCRIPT

TE D

M AN U

SC

Fig.12 – Oil recoveries of 15S and S with variable foam qualities (75% and 50%).

AC C

EP

Fig.13 – Pressure drop of co-injection CO2 and surfactant solution with 15S at 75% (L, 0.2 cc/min and G, 0.6 cc/min).

Fig.14 – Pressure drop of co-injection CO2 and surfactant solution with S at 75% (L, 0.2 cc/min and G, 0.6 cc/min).

RI PT

ACCEPTED MANUSCRIPT

TE D

M AN U

SC

Fig.15 – Pressure drop of co-injection CO2 and surfactant solution with 15S at 50% (L, 0.2 cc/min and G, 0.2 cc/min).

AC C

EP

Fig.16 – Pressure drop of co-injection CO2 and surfactant solution with S at 50% (L, 0.2 cc/min and G, 0.2 cc/min).

Fig.17 – Summary of oil recoveries for all experiments conducted in fractured system.

RI PT

ACCEPTED MANUSCRIPT

TE D

M AN U

SC

Fig.A1 – Surfactant concentrations at 1IPV and 20 IPV, (a) Global, (b) Water

AC C

EP

Fig.A2 – Schematic demonstration of two scenarios

Fig.B1 - CO2 storages for novel surfactants with imaginary partition coefficients during alternating injection of CO2 and surfactant solution with 36.5 day slug

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

Fig.B2 - CO2 storages for novel surfactants with imaginary partition coefficients during SIDI

ACCEPTED MANUSCRIPT

RI PT

SC M AN U

•

TE D

• •

EP

• •

Foam was able to generate and propagate in tested intermediate-wet analogue core, especially for employed CO2 soluble surfactants The tested CO2 soluble surfactants did not present affinity with used Wasson crude oil In fractured cores, oil recoveries were in ascending order of pure CO2 flooding, COinjection of CO2 and water, water-soluble surfactant foam and CO2 soluble surfactant foams Oil recovery was further enhanced with high surfactant partitioning into CO2 phase The partitioning of surfactant into the CO2 phase leads faster foam propagation and stronger foam in current scenarios The optimal CO2 soluble surfactant is case dependent an done could tailor a surfactant to optimize the oil recovery in corresponding scenarios

AC C

•