water foams over a wide range in salinity and temperature

Journal of Colloid and Interface Science 522 (2018) 151–162

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science journal homepage: www.elsevier.com/locate/jcis

Regular Article

Viscoelastic diamine surfactant for stable carbon dioxide/water foams over a wide range in salinity and temperature Amro S. Elhag a,b, Chang Da a, Yunshen Chen a, Nayan Mukherjee a, Jose A. Noguera a, Shehab Alzobaidi a, Prathima P. Reddy a, Ali M. AlSumaiti b, George J. Hirasaki c, Sibani L. Biswal c, Quoc P. Nguyen d, Keith P. Johnston a,⇑ a

McKetta Department of Chemical Engineering, The University of Texas at Austin, United States Petroleum Engineering Department, Khalifa University of Science and Technology, Petroleum Institute, United Arab Emirates c Department of Chemical and Biomolecular Engineering, Rice University, United States d Hildebrand Department of Petroleum and Geosystems Engineering, The University of Texas at Austin, United States b

g r a p h i c a l a b s t r a c t

a r t i c l e

i n f o

Article history: Received 2 January 2018 Revised 8 March 2018 Accepted 12 March 2018 Available online 13 March 2018 Keywords: CO2-in-water foam High Temperature High internal phase foam Viscoelastic surfactant Wormlike micelles

a b s t r a c t Hypothesis: The viscosity and stability of CO2/water foams at elevated temperature can be increased significantly with highly viscoelastic aqueous lamellae. The slow thinning of these viscoelastic lamellae leads to greater foam stability upon slowing down Ostwald ripening and coalescence. In the aqueous phase, the viscoelasticity may be increased by increasing the surfactant tail length to form more entangled micelles even at high temperatures and salinity. Experiments: Systematic measurements of the steady state shear viscosity of aqueous solutions of the diamine surfactant (C16-18N(CH3)C3N(CH3)2) were conducted at varying surfactant concentrations and salinity to determine the parameters for formation of entangled wormlike micelles. The apparent viscosity and stability of CO2/water foams were compared for systems with viscoelastic entangled micellar aqueous phases relative to those with much less viscous spherical micelles. Findings: We demonstrated for the first time stable CO2/water foams at temperatures up to 120 °C and

⇑ Corresponding author at: Department of Chemical Engineering, The University of Texas at Austin, Austin, TX 78712, United States. E-mail address: [email protected] (K.P. Johnston).

https://doi.org/10.1016/j.jcis.2018.03.037 0021-9797/Ó 2018 Published by Elsevier Inc.

152

A.S. Elhag et al. / Journal of Colloid and Interface Science 522 (2018) 151–162

CO2 volumetric fractions up to 0.98 with a single diamine surfactant, C16-18N(CH3)C3N(CH3)2. The foam stability was increased by increasing the packing parameter of the surfactant with a long tail and methyl substitution on the amine to form entangled viscoelastic wormlike micelles in the aqueous phase. The foam was more viscous and stable compared to foams with spherical micelles in the aqueous lamellae as seen with C12-14N(EO)2 and C16-18N(EO)C3N(EO)2. Ó 2018 Published by Elsevier Inc.

1. Introduction A fundamental understanding of the formation, stabilization and rheology of bulk (CO2 in water) C/W foams [1,2] (often referred to as emulsions for dense supercritical CO2) and C/W foams in porous media as a function of foam texture is needed to advance subsurface applications including CO2 sequestration [3], enhanced oil recovery (EOR) [1,4] fracturing with energized CO2-based fluids [5], and other applications including materials synthesis [6] and pharmaceuticals [7]. Designing nonionic and cationic surfactants for C/W foams at temperatures >100 °C (373 K) is challenging given chemical instability and/or limited solubilities in the aqueous phase. As the temperature is increased, nonionic surfactants with hydroxyethyl (EO) groups precipitate at the cloud point as hydrogen bonding with water becomes weaker [8–10]. Relatively few cationic surfactants are chemically stable and soluble in aqueous phases above 100 °C as the quaternary ammonium group is prone to Hoffman elimination of alkyl groups [11,12]. Our group found that the (EO) substituted amine C12-14N(EO)2 and diamine C16-18N(EO)C3N(EO)2 in the protonated cationic form at low pH are soluble at temperatures up 120 °C due to the strong ion dipole interactions between the cationic amine head group and water [13–15]. Although anionic surfactants such as sulfates and sulfonates are soluble and stabilize C/W foams at high temperatures, they are limited in certain cases, for example as they adsorb strongly on positively charged surfaces such as carbonate at low pH [16]. At high temperatures, the generation and stabilization of aqueous lamellae in bulk C/W foams and C/W foams in porous media becomes more challenging given limitations in interfacial and phase behavior [13,17]. For ionic surfactants increasing temperature at constant pressure usually increases the C-W interfacial tension (c), as the tails become less solvated with the decrease in CO2 density [2,13,18]. With the higherc, the minimum amount of work (Wmin) required to generate foam increases as given by W min ¼ cDA where DA is the change in the area of the liquid gas interface in the foam [17]. Furthermore, at high temperature, the foam is destabilized by faster drainage of the lamellae between the CO2 bubbles given the lower viscosity of the continuous aqueous phase (me) [19,20]. As the films thin, destabilization by coalescence and/or Ostwald ripening becomes more prevalent [19]. Also, as, the volume fraction of CO2 (foam quality) increases, the capillary pressure increases until it reaches the limiting capillary pressure and overcomes the disjoining pressure between the flat CO2 surfaces in the lamellae [21]. At this limiting capillary pressure, which increases at high T with an increase in c, the lamellae become instable [21]. Given all of these challenges, examples of C/W foam at higher temperatures >100 °C are extremely rare [5,13–15,22–24]. For example at 120 °C and qualities up to 0.90, C12-14N(EO)2 and C16-18N (EO)C3N(EO)2 stabilize bulk C/W foam formed at high shear rates (superficial velocities) in a sand pack with permeabilities in the range of 1.2 (1.2 mm2) to 76 Darcy (75 mm2) [13,15,24]. The same surfactant (C12-14N(EO)2) was used by Cui et al. to successfully generate foam at 120 °C in a Silurian dolomite core with qualities from 0.3 to 0.8 [24]. Given the scarcity of currently known surfactants that are soluble in brine at high temperature and that stabilize C/

W foams, it would be beneficial to discover additional surfactant candidates guided by a better understanding of interfacial properties and foam formation and stabilization mechanisms. To aid stabilization of a gas/water (G/W) foam at low pressure (below 7 MPa), a viscoelastic aqueous phase may be utilized to retard lamellae drainage and subsequent coalescence and Ostwald ripening [25–29]. This viscoelastic aqueous phase can be formed with entangled elongated cylinders with hemispherical caps called wormlike micelles (WLMs) [30]. The transition from spherical to wormlike micelles may be achieved by increasing the packing parameter p = v/alc, where v is the volume of surfactant tail, lc is the extended length of the tail and a0 is the head group area [30,31]. In particular, the head group area may be reduced by mixing a cationic and anionic surfactant to lower electrostatic repulsion [30,32,33]. Furthermore, different stimuli such as pH [34–37] or CO2 [35,36,38,39] were used to switch the micellar morphology and achieve reversible viscoelasticity of the aqueous phase. For example, Chu et al. found that spherical micelles of a mixture of Nerucamidopropyl-N,N-dimethylamine and maleic acid at high pH can be switched reversibly to entangled wormlike micelles by lowering the pH to quaternize the amines and form a ‘‘gemini – like” surfactant [34]. At ambient pressures, Monteux et al. found that viscoelastic lamellae formed by a complex of dodecyltrimethylammonium bromide and poly(styrenesulfonate) inhibited air/water (A/W) foam drainage and bubble coalescence [25]. Varade et. al. utilized a mixture of anionic myristic acid (C13COOH) and cationic cetyl trimethylammonium chloride (C16TA+Cl) to form lamellae with viscoelastic wormlike micelles that were stable against coalescence and Ostwald ripening in a nitrogen/W foam [26]. At high pressures, C/W foams stabilized with viscoelastic surfactants have been studied for applications such as hydraulic fracturing [5,28,29,40,41] and enhanced oil recovery [42]. Miscible injection of CO2 for enhanced oil recovery (EOR) is limited by an unfavorable mobility ratio of CO2 compared to the targeted crude oil that leads to viscous fingering, gas channeling and gravity override [1,4]. To overcome these challenges, (C/W) foams [2] are used to raise the viscosity of the CO2 phase by 2–4 orders of magnitude [1,43,44]. Recently, there has been an increasing demand for designing foams for CO2 EOR for carbonate reservoirs in the Middle East at high temperatures [45,46] Sun et. al. found that C/W foams formed with 3.8% w/w of sulfosalicylic acid triethanolamine ester were shear thinning and the viscosities decreased with increasing temperature up to 80 °C and qualities up to 0.75 [41]. In these studies, the morphology of the aqueous phase and the texture of the foam were not reported. Xue et al. formed viscous C/W foam up to extremely high qualities (0.98) utilizing mixtures of anionic and cationic (catanionic) surfactant [29] or a combination of surfactants, polymers and nanoparticles [28]. Despite the high qualities, the foam was stable up to several hours with small bubble sizes of 20 nm. For the catanionic wormlike micelles, the foam stability was increased as a consequence of slower drainage of the lamellae that retarded Ostwald ripening [29]. However, these studies were limited to temperatures less than 50 °C and required concentrations of surfactants above 4% w/w [29] or the addition of polymers [28,29]. Recently, Feng et al. used an amine surfactant, (N-erucamidopropyl-N,N-dimethylamine) that formed wormlike

A.S. Elhag et al. / Journal of Colloid and Interface Science 522 (2018) 151–162

micelles in the presence of CO2 to generate C/W foams at high temperatures up to 140 °C [42]. The foam was formed via injection of alternating slugs of surfactant dissolved in the aqueous phase followed by CO2 ‘‘gas” (SAG) such that the foam quality varied throughout the sand pack during the displacement process [42]. The objective of this study was to examine how a highly viscous and viscoelastic aqueous phase composed of wormlike micelles influences the properties of C/W foams for a single surfactant at a low concentration (1% w/w) over a wide range of temperatures from ambient up to 120 °C. Furthermore, the rheological properties of the foam and the stabilization mechanisms are described as a function of the bulk foam texture measured at high pressure and temperatures as high as 90 °C. For C16-18N(CH3)C3N(CH3)2 relative to previously reported switchable amine surfactants C12-14N(EO)2 [13,14] and C16-18N(EO)C3N(EO)2 [15], the greater hydration with methyl relative to hydroxyethyl substitution on the amine(s) at high temperature and the choice of a dual rather than a monoamine are shown to raise the brine solubility at high temperature. In addition, these structural changes also increase the surfactant adsorption at the CO2-water interface indicating a favorable hydrophilic-CO2philic balance. Interestingly, even with two polar amines in the head group, the surfactant is shown to still be soluble in the CO2 phase, because methyl groups are more CO2-philic compared to EO groups. To provide a basis for the design of C/W foams, we first studied the formation of WLMs in aqueous micellar phases at ambient pressure. We mapped the effect of the surfactant concentration, salinity and pH on the steady state viscosity and the elastic and viscous moduli using a cone and plate rheometer. For C16-18N(CH3) C3N(CH3)2, the longer tail compared to C12-14N(EO)2 or smaller head versus C16-18N(EO)C3N(EO)2, given methyl instead of hydroxyethyl substitution on the amine, increases the packing parameter and is shown to enable formation of a viscoelastic aqueous phase. The surfactant aggregates behave analogously to solutions comprised of entangled wormlike micelles. For C/W foams, the viscoelastic lamellae are shown to decrease the rate of capillary pressure drainage to stabilize foam at extremely high quality of 0.98 where capillary pressures become very large. The maintenance of thick viscoelastic films are shown to raise the stability of the C/W foams from coalescence and Ostwald ripening at all qualities as investigated by time dependent microscopy. The increase in C/W foam viscosity in the presence of viscous lamellae and the effect of shear rate are explained using previously developed models of foam viscosity in capillary tubes and porous media. Finally, foams were generated at lower superficial velocities with the more viscous aqueous phases, given the higher pressure gradient for a given velocity according to Darcy’s law. This fundamental study is of interest in various applications including enhanced oil recovery with CO2 at elevated temperatures to target enormous amounts of oil. Highly stable carbon dioxidein-water foams are of great interest for mobility control in CO2 EOR to raise the sweep efficiencies. Despite the need for these foams, the rheological behavior of CO2/water foams and the interfacial properties at temperatures beyond about 80–100 °C have very rarely been reported given limitations in surfactant solubility and chemical stability.

2. Experimental 2.1. Materials Bis(2-hydroxyethyl) cocoalkylamine (C12-14N(EO)2), Bis(2hydroxyethyl) tallowalkylamine (C16-18N(EO)2), Tris(2-hydroxye thyl)-N-tallowalkyl-1,3-diaminopropane (C16-18N(EO)C3N(EO)2) and N,N,N0 trimethyl-N’-tallow-1,3- diaminopropane (C16-18N

153

(CH3)C3N(CH3)2) were gifts from Akzo Nobel and used without further treatment. The commercial name, CAS# and activity are shown in Table 1. Carbon dioxide (99.99%) was purchased from Matheson and used as received. Sodium chloride (NaCl, certified ACS, Fisher), calcium chloride dihydrate (CaCl22H2O, 99+% Acros), magnesium chloride hexahydrate (MgCl26H2O, Fisher), hydrochloric acid (HCl, technical, Fisher) and isopropanol (certified ACS plus, Fisher) were used as received. Sodium chloride solutions were prepared by adding NaCl concentrations from 30 g/l to 182 g/l to deionized (DI) water ((Nanopure II, Barnstead, Dubuque, IA). Also, a brine (22% TDS or 225 g/l) composed of 182 g/l NaCl, 77 g/l CaCl2 2H2O, 26 g/l MgCl2 6H2O was used as a model reservoir brine. Crushed carbonate mineral (limestone) was obtained from Franklin Industrial Minerals and sieved with 20–40 mesh sieves to obtain average grain diameter from 420 to 840 lm. Spherical glass beads with 30–50 lm diameter were obtained from polysciences, Inc and washed with copious amounts of water and ethanol to remove any impurities. The beads were placed in an oven until dry. 2.2. Degree of protonation 40 mM solutions of switchable amine surfactants in DI water, 30 g/l NaCl or 22% TDS brine were prepared and titrated using 800 mM HCl solutions at room temperature or 90 °C. The volume of HCl added and the pH were recorded and the overall degree of protonation was obtained from Eq. (1)

h¼

Ca  Vtitrant þ ðCOH  CHþ Þ  Vtitrand Ctotal  Vtitrand0

ð1Þ

where C a is the concentration of HCl and CHþ and COH are the concentration of H+and OH determined from the pH of the solution. Ctotal is the total concentration of the unprotonated amine added initially (40 mM). V titrand0 is the volume of the amine solution before adding HCl, whereas V titrand is the total volume of titrand at any stage of titration. Further details are explained in a previous publication [13]. 2.3. Cloud point temperature The cloud point temperature of 1% w/w of switchable amine surfactants were measured at different pH values up to 120 °C (±1 °C) using a sealed pipette method as described in a previous publication [14]. 2.4. Aqueous phase rheology at room temperature Shear viscosities at steady state and room temperature (25 °C) for aqueous solution of switchable amine surfactants dissolved in different brines were characterized using AR G2 rotational rheometer (TA Instruments) equipped with a cone-and-plate geometry (40 mm, 2°). The pH of the aqueous phase solution was adjusted to pH 4 using HCl (to mimic the acidic conditions in the presence of high pressure CO2). The variation for multiple measurements was within 10%. Aqueous phase viscosity measurements were carried out with a shear rate between 1 s1 and 200 s1. A frequency scan from 0.1 to 100 rad/s was performed with the rotational rheometer to measure the elastic (G0 ) and viscous (G00 ) moduli. For proper loading, a preshear of 10 s1 for 1 min and then a rest for 100 s on the samples were applied. Further description is provided by a recent publication [29]. The steady state viscosity was measured at high temperature with ARES-G2 (TA instrument) equipped with cone-shaped cylinder geometry. The zero shear viscosity was examined at 25 °C, 60 °C, 80 °C, and 100 °C, separately.

154

A.S. Elhag et al. / Journal of Colloid and Interface Science 522 (2018) 151–162

Table 1 Composition of the tertiary amines and diamines. Surfactant commercial name

composition

HLB davis scale

CAS #

Activity % by weight

Ethomeen C/12 Ethomeen T/12 Ethoduomeen T/13 Duomeen TTM

C12-14N(EO)2 C16-18N(EO)2 C16-18N(EO)C3N(EO)2 C16-18N(CH3)C3N(CH3)2

12.2 10.1 19.0 14.0

61791-14-8 61791-44-4 61790-85-0 68783-25-5

97–100 97–100 90–94 90–100

2.5. CO2 solubility The solubility of 0.2% w/w of C16-18N(CH3)C3N(CH3)2 in CO2 was carried out using a variable volume view cell equipped with a piston. A known mass of the surfactant was first loaded into the cell followed by loading of certain volume of CO2 to make 0.2% w/w mixture. The mixture was then pressurized to 5000 psia (34 MPa) and stirred using a stir bar for two hours to ensure equilibrium. The pressure was then reduced and observation of cloudiness in the cell indicates the cloud point pressure. The experiment was then repeated at different temperatures. A complete description of the experiment was provided in a previous publication [14]. 2.6. Interfacial tension measurements between CO2 and brine Interfacial tension measurements between CO2 and 22% TDS brine in the presence of C16-18N(CH3)C3N(CH3)2 over a wide range of concentrations were conducted using an axisymmetric drop shape analysis of a captive bubble as described in previous publications. The measurements were performed at 120 °C and 3400 psia. the standard deviation of 10 measurements was typically less than 2% of the mean. From a plot of the interfacial tension vs. surfactant concentration was used to determine other interfacial properties such as the critical micelle concentration (CMC) and the surfactant adsorption at the interface. Further details can be found in our previous publications. 2.7. C/W foam formation and apparent viscosity Foam apparent viscosity was measured using the apparatus depicted in Fig. S1 up to 120 °C and 3400 psia. The porous media was either a 1 Darcy pack (30–50 mm in diameter spherical glass bead pack) or a 76 Darcy carbonate pack (420–840 lm crushed limestone). The capillary tube was 660 mm ID by 1.5 m in length hastelloy and was used to measure the bulk foam viscosity. The procedure for measuring the foam viscosity in the porous media or the capillary is highlighted in a previous publication. The error in the measurements was less than 15%. A high pressure cell located after the capillary is placed in a microscope (Nikon Eclipse ME600) equipped with a Photometrics CoolSNAP CF CCD camera to take micrographs of the foams. Image J was used to analyze the bubble sizes as described previously [9]. 3. Results and discussion In the results and discussion section, we demonstrate how the addition of a second amine in the head and/or methyl substitution on the amine enhance the aqueous phase behavior, aqueous phase viscosity, C-W interfacial behavior, and foam viscosity and stability in bulk and porous media. 3.1. Aqueous phase behavior 3.1.1. Degree of protonation The degree of protonation vs. pH of 40 mM switchable methyl and ethoxy substituted diamine surfactants at different salinities

and temperature is shown in Fig. 1. With the addition of HCl, the degree of protonation of the nonionic amine head group increased and this increased the surfactant solubility. In DI water, the titration curves were similar for C16-18N(EO)C3N(EO)2 and C16-18N (CH3)C3N(CH3)2 at each temperature, although the slopes were somewhat different. In comparison, similar degrees of protonation versus pH were seen for the single amine C12-14N(EO)2 reported previously by Chen et al. [13], but the concentration of the protonated amine is doubled for the diamine surfactants. The basicity of each N is about the same so that the two protonations appear to be continuous. For example, with surfactant concentration of 40 mM at 90 °C in DI water and pH 5, all surfactants had a degree of protonation of 0.7 which translates to protonated amine concentration of 28 mM and 56 mM for single and diamines, respectively. The higher ionic character of the diamine surfactants will prove to be advantageous in enhancing the aqueous solubility as discussed in the next section. The shift in the curves to the right towards greater protonation for C16-18N(CH3)C3N(CH3)2 with increasing salinity may be attributed to decrease in repulsion between the positively charged head groups in the micelles and the hydronium ions as these charges are screened [13,47]. For C16-18N(CH3)C3N (CH3)2, increasing temperature lowered the degree protonation from 0.75 at room temperature to only 0.1 at 90 °C at pH 6 in DI water. The effect of temperature can be explained by the exothermic nature of the protonation reaction as measured previously for tertiary amine surfactants [48–50]. Similar effects of salinity and temperature were observed for C12-14N(EO)2 [13]. 3.1.2. Aqueous solubility and cloud point temperature The aqueous solubility in terms of cloud point temperatures of 1% w/w switchable amine surfactants in 22% TDS solution are shown in Fig. 2. At pH 8 and all temperatures, all surfactant solutions were insoluble as the surfactants were not protonated. At low pH, the solubilities increase as the protonated nitrogen in the amine heads interacts with water [13,14]. As shown in Fig. 2, decreasing the tail length from tallow C16-18N(EO)2 to coco C1214N(EO)2 enhanced the solubility of the surfactant in brine, i.e. increased the pH, such that less protonation was required. This increase in solubility can be attributed to the decrease in the hydrophobicity of the shorter tail [51–53]. Despite the slightly longer tail, the addition of a second hydrophilic amine head as in C16-18N(EO)C3N(EO)2 raised the HLB (Table 1) and surfactant solubility in brine when compared to C12-14N(EO)2, as the pH required for solvation at room temperature increased from 6.6 to 7.6 (Fig. 2A and C) [15,17,48,51]. At room temperature, changing the type of substitution from EO to CH3 had little effect on the pH required to achieve aqueous solubility consistent with the protonation results in Fig. 1. Greater differences in behavior among the surfactants was observed at high temperature. At pH 7.6, C16-18N(EO)C3N(EO)2 exhibited a cloud point of 70 °C. With a decrease in pH to 7.0 and an increase in amine protonation, the cloud point was not reached even at the highest temperature measured [51] (Fig. 2D), suggesting that CH3 groups are solvated more effectively than EO groups at high temperature. The solvation of EO units in water diminishes at high temperature given the loss of hydrogen bonding with water, resulting in aggregation of surfactants at the cloud

A.S. Elhag et al. / Journal of Colloid and Interface Science 522 (2018) 151–162

A) Room temperature

155

B) 90 °C

Fig. 1. Degree of protonation of 40 mM solutions of switchable surfactants vs pH at 1 atm.

A) C12-14N(EO)2

B) C16-18N(EO)2

C) C16-18N(EO)C3N(EO)2

D) C16-18N(CH3)C3N(CH3)2

Fig. 2. Aqueous solubility of 1% w/w switchable surfactants in 22% TDS brine. Open circles indicate a clear solution and filled circles indicate cloudy solutions.

point temperature [8,13]. On the contrary, the aqueous solvation of the methyl groups in the head group increases with temperature, as the structure of water is more disordered at higher temperatures [54,55]. In summary, the combined effect of the second protonated amine in the head and methyl substitution on these amines enhanced the solubility in water and eliminated the cloud points up to 120 °C.

3.2. Aqueous solution morphology and rheology The steady state shear viscosity of an aqueous solution of 1% w/ C16-18N(CH3)C3N(CH3)2 at pH 4 (to mimic the pH of the surfactant solution in the presence of high pressure CO2) as a function of surfactant concentration and salinity is shown in Fig. 3 and Fig. S2 at atmospheric pressure. At a surfactant concentration of 1% w/w, and up to a salinity of 182 g/l NaCl, the zero shear viscosity was limited to 8 cP (mPa s) followed by a sharp increase to 700 cP in the presence of the model 22% TDS brine. The addition of electrolytes reduced the electrostatic repulsion between the charged heads

Fig. 3. Steady state viscosity vs. shear rate of C16-18N(CH3)C3N(CH3)2 at pH 4 and atmospheric pressure with brine salinity, temperature and surfactant concentration as variables. Both the shear rate and the viscosity are shown on a log scale.

156

A.S. Elhag et al. / Journal of Colloid and Interface Science 522 (2018) 151–162

and lowered their effective area (a0) to increase the packing parameter and produce a transition from spherical to entangled wormlike micelles [29,30]. Similar effects of simple salts such as NaCl and KCl on the aqueous solution viscosity were observed for anionic and cationic surfactants such as sodium dodecyl sulfate (SDS)and erucyl bis(hydroxyethyl)methylammonium chloride (EHAC), respectively [30,56]. To form viscous entangled wormlike micelles at lower salinities, the surfactant concentration must be increased. For example, at 182 g/L NaCl, the aqueous phase viscosity increased by over an order of magnitude upon increasing the surfactant concentration from 1% w/w to 2%w/w as shown in Fig. 3. Although the packing parameter is independent of surfactant concentration, an increase in surfactant concentration above the overlap concentration (c*), induces greater entanglement that raises the viscosity [56]. The steady state shear viscosities of an aqueous solution of 1% w/ C16-18N(CH3)C3N(CH3)2 at temperatures of 60 °C, 80 °C and 100 °C are shown in Fig. 3 and Fig. S4. At all shear rates, the viscosity of the aqueous phase decreases with an increase in temperature. Despite this decrease, the aqueous phase viscosity was three orders of magnitude higher than pure water at the same temperatures indicating the presence of viscous micellar assemblies. The continuous decrease of viscosity with temperature is typical for wormlike micelles and may be attributed to a decrease in the length of the wormlike micelles [57,58]. Similar behavior was reported for cationic amine surfactants (EHAC) and (ETAC) from room temperature to 90 °C [57]. At the highest tested salinity (22% TDS), the steady state shear viscosity of the aqueous solution of 1% w/w switchable amine surfactants at pH 4 is shown in Fig. 4. At pH 4, the zero shear viscosity (lo) of C16-18N(CH3)C3N(CH3)2 was found to be 700 cP indicating the presence of entangled wormlike micelles. Although this type of morphology has not been reported previously for this surfactant, it has been achieved in similar amine surfactants upon decreasing the pH to protonate nitrogen atoms [33,34,38,39,59]. For example, Chu and Feng found that micellar structure of a mixture of N-eruca midopropyl-N,N-dimethylamine and maleic acid can reversibly switch from spherical at high pH to entangled wormlike micelles at low pH values as the amine group is protonated [34]. The viscosity of the aqueous C16-18N(CH3)C3N(CH3)2 solution was two orders of magnitude higher than for the other switchable amine surfactants, namely, C16-18N(EO)C3N(EO)2 and C12-14N(EO)2 with packing parameters that are too small to form wormlike micelles. Relative to C12-14N(EO)2 the increase in the tail length for C16-18N(EO)C3N (EO)2 increases the aggregation number favoring the formation of wormlike micelles despite the independence of the packing parameter on tail length (v/alc is constant) [60]. Also, the smaller CH3 headgroup compared to EO in C16-18N(EO)C3N(EO)2 further raises p [30]. The shear thinning behavior for the C16-18N(CH3)C3N

(CH3)2 solution indicates breaking of the entanglements of the wormlike micelles [30,56,61]. As expected, the viscosity of spherical micellar solutions of other surfactants stayed constant with increasing shear rate indicating Newtonian behavior. The viscosity of the aqueous solution of 1% w/w C16-18N(CH3)C3N(CH3)2 in 22% TDS saturated with CO2 at high pressure of 3400 psia (similar conditions used in the C/W foam sections) was measured in the capillary tube shown in Fig. S1. At shear rate of 230 s1, the surfactant maintained high viscosity of 21 cP comparable to the viscosity measured with the cone and plate rheometer indicating the existence of wormlike micelles even in the presence of CO2. The viscoelastic properties of the viscous aqueous solution of 1% w/w C16-18N(CH3)C3N(CH3)2 in 22% TDS brine were characterized by oscillatory shear measurements of the elastic and viscous moduli, G’ and G’’ respectively, as shown in Fig. S4. As wormlike micelles become more entangled and more elastic, the intersection of G’ and G’’ shifts to lower frequencies [30,56]. Despite a concentration of only 1% w/w surfactant, the low intersection angular frequency of 5 rad/s indicates greater entanglement compared to the experiment with a concentration of 2% w/w at 182 g/l NaCl (Fig. S4) and previously reported highly concentrated solutions for a mixture of 3.6% SLES and 0.4%C10DMA [29]. The extremely high salinity appears to reduce the headgroup area via charge screening and raise the packing fraction to produce a highly viscous and elastic phase even with only a single surfactant and at very low concentration. 3.3. CO2 solubility The cloud point pressures of 0.2% w/w amine surfactant solutions in CO2 at temperature between 25 and 120 °C are shown in Fig. S6. Since these are tertiary amines they do not form carbamates [62]. In dry CO2, the switchable amine surfactants are unprotonated and thus nonionic, which allows CO2 solvation [63]. For ionic surfactants, the low solvation strength of CO2 is usually insufficient to overcome the attractive forces between the ionic head groups and counterions [63]. From Fig. S6, it was found that only C12-14N(EO)2 and C16-18N(CH3)C3N(CH3)2 are soluble conditions of 120 °C and 3400 psia, which are of interest for example in certain oil reservoirs. CO2 is a weak solvent for long alkanes and cations as it does not have a dipole moment [64]. Relative to C1214N(EO)2, the increase in the size of both the tail and the addition of a second cationic amine group decreased the solubility in CO2 that is, increased the cloud point pressure (or density) [63,64]. For a fixed C16-18 tail, changing the substitution on the amine head from EO to CH3 units enhanced the CO2 solubility by decreasing the polar interactions and hydrogen bonding between EO groups [63,64]. Furthermore, CH3 groups which have a lower cohesive energy density than even CH2 groups are known to be relatively CO2 philic, and more so than EO groups [64]. 3.4. Interfacial tension at the C-W interface

Fig. 4. Steady state viscosity vs. shear rate for 1% w/w switchable amines in 22% TDS, pH4 at room temperature and atmospheric pressure.

Interfacial tension measurements between 22% TDS and CO2 were conducted as a function of C16-18N(CH3)C3N(CH3)2 concentration in the aqueous phase at 120 °C and 3400 psia from 3.0  10t M to 3.0  102  M (Fig. 5). In the presence of high pressure CO2 the pH of the aqueous phase is 3 [65] and the surfactant is nearly completely protonated (cationic), as shown in Fig. 1. The discontinuity in the slope indicates the presence of a critical micelle concentration (CMC) at a value of 0.4 mM. Furthermore, above the CMC, the interfacial tension of only 5.8 mN/m was reduced markedly from the value of 35 mN/m in the absence of surfactant [66]. A comparison of the molar surface density and other interfacial properties of the three amines at the CO2-brine interface is shown in Table 2. The adsorption of C16-18N(EO)C3N(EO)2 was lower than

157

A.S. Elhag et al. / Journal of Colloid and Interface Science 522 (2018) 151–162

Fig. 5. Interfacial tension (IFT) for C16-18N(CH3) C3N(CH3)22 at the CO2-22% TDS brine interface versus the logarithm of surfactant concentration at 120 °C and 3400 psia.

that of C12-14N(EO)2 in part due to the larger area occupied by that larger tail and larger head group [15,17]. Another possibility is that the second cation pulled the surfactant towards water away from the interface, despite the longer tail. Interestingly, the adsorption was highest for C16-18N(CH3)C3N(CH3)2. Relative to C16-18N(EO) C3N(EO)2, the higher adsorption may be attributed to the smaller head group and the stronger CO2 solvation of the CH3 relative to the EO groups [64]. The variation in CMC among the surfactants was greater than that of the adsorption values. Compared to C12-14N(EO)2 and C1618N(EO)C3N(EO)2, the CMC of C16-18N(CH3)C3N(CH3)2 is an order of magnitude higher suggesting greater hydrophilicity for the smaller CH3 groups than EO groups at high temperature [23]. At high temperature, and even more so with added salt, the hydrogen bonding of EO by water is greatly weakened. This increase in hydrophilicity was also apparent above in the higher solubility in brine with CH3 relative to EO modification. A similar trend was reported for single amines where the CMC of C12-14N(CH3)3Cl was an order of magnitude higher than that of protonated C12-14N (EO)2 at 120 °C [13,23]. 3.5. C/W foam texture, viscosity and stability 3.5.1. Effect of salinity on apparent foam viscosity In the 76 Darcy (75 mm2) limestone pack, the apparent viscosity from the pressure drop across the pack indicated C/W foams were formed with C16-18N(CH3)C3N(CH3)2 at 120 °C in the absence of salt at 1%w/w, but not for C12-14N(EO)2, (Fig. 6). However, at 30 g/L NaCl the foam apparent viscosity was similar for both surfactants 8 cP and remained at this value for up to 180 g/L. At the experimental conditions, the calculated pH was 4.6 from a geochemical model of the relevant reaction equilibria, PHREEQC [13]. The measured effluent pH at the outlet of the setup (low pressure CO2) was 5.7 in the absence of salt. At this pH and 90 °C, the degree of protonation of both surfactants was 0.7 in DI water and may be expected to decrease with increasing temperature to 120 °C as discussed above. Both protonated surfactants were soluble as discussed above in Fig. 2 and in an earlier publication which is required for lamellae stabilization [13]. For the point at 22% TDS only, the brine consisted of 182 g/L NaCl, 77 g/L CaCl2 2H2O, 26 g/ L MgCl2 6H2O. Here, a sharp increase in the foam viscosity for

Fig. 6. Apparent viscosity of C/W foam stabilized by co-injecting 1% (w/w) C16-18N (CH3)C3N(CH3)2 or C12-14N(EO)2 pH 6.5 solution with CO2, at 120 °C, 3400 psia and total superficial velocity 938 ft/day in 76 Darcy limestone pack or a capillary tube with 90% foam quality. The salt is NaCl up to a TDS of 18%. For the 22% TDS, the salt is 182 g/L NaCl, 77 g/L CaCl2 2H2O, 26 g/L MgCl2 6H2O.

C16-18N(CH3)C3N(CH3)2 may be attributed to the high aqueous phase viscosity as a consequence of the formation of WLMs, consistent with the behavior in aqueous phases at atmospheric pressure in Fig. 3. This behavior will be further discussed after first examining the rheology at other qualities and for the other amine surfactants. Bulk foam behavior in the capillary tube was similar to that in the limestone pack except the apparent viscosities were lower. Stretching and contracting of lamellae through valves and tubes on the path to the capillary may cause partial destabilization of the foam [67]. However, the foam was still stable as depicted from the foam texture measured in a high pressure microscopic cell downstream of the capillary as discussed below (Fig. 7). 3.5.2. Foam texture The measurement of foam texture with high pressure microscopy is used to better understand foam generation, viscosity and stability. To our knowledge, we are not aware of previous studies to determine foam texture at elevated temperatures. As shown in Fig. 7, for a foam quality of 0.9 at both 25 and 90 °C, the distribution of bubble sizes is observed to range from 10 to 60 nm, as shown in the histograms in Fig. S7. Furthermore, the flattening of the lamellae may be seen at this high quality, but less than would be present if the polydispersity were lower since small bubbles fill the plateau borders of larger bubbles. The Sauter mean diameter, Dsm, was comparable for the three protonated amines Table 3 ranging from 18 to 28 nm, and smallest for C12-14N(EO)2 . Also, for C1618N(CH3)C3N(CH3)2 the initial Dsm increased only a small amount from 28 to 40 nm when temperature was raised from 25 to 90 °C. The bubble formation can be described by the Weber number (We) which defines the balance between the shear forces and the capillary forces

We ¼

Gls R

ð2Þ

c

where G is the shear stress, ls is the continuous phase viscosity and R is the bubble radius [68]. Thus, the lower interfacial tension and Laplace pressure for C12-14N(EO)2 , with the same applied energy, facilitated the formation of smaller initial bubble size when

Table 2 Interfacial properties of switchable amines at the CO2-22% TDS brine interface at 120 °C and 3400 psia. Surfactant

CO2 density (g/mL)

IFTcmc (mN/m)

CMC (mmol/L)

C * 106 (mol/m2)

Αm (Å2/molecule)

C16-18N(CH3) C3N(CH3)2 C16-18N(EO) C3N(EO)2 C12-14N(EO)2

0.476 0.476 0.476

5.8 6.2 5.1

0.400 0.015 0.038

0.88 0.70 0.80

188 238 207

158

A.S. Elhag et al. / Journal of Colloid and Interface Science 522 (2018) 151–162

a) C12-14N(EO)2 , 25 °C

b) C16-18N(EO)C3N(EO)2, 25 °C

c) C16-18N(CH3)C3N(CH3)2, 25 °C

d) C16-18N(CH3)C3N(CH3)2, 90 °C

Fig. 7. Micrographs of C/W foam stabilized by 1% w/w switchable amine surfactants in 22% TDS brine pH 4, 3400 psia and total superficial velocity 938 ft/day in a 76 Darcy limestone pack.

Table 3 Bubble sizes and Ostwald ripening rate for foams of C16-18 N(CH3)C3N(CH3)2, C12-14N (EO)2 and C16-18 N(EO)C3N(EO)2 with concentrations of 1 wt% in 22% TDS brine at 0.9 CO2 volume fraction, 25 °C and 3400 psia. Surfactant

Dsm (t = 0 min) nm

Dsm (t = 1200 min) nm

X3 mm3/min

C16-18N(CH3)C3N(CH3)2 C16-18N(EO)C3N(EO)2 C12-14N(EO)2

28 ± 10.1 27 ± 10.9 18 ± 5.8

57 109 75

135 925 325

compared to diamine surfactants with higher interfacial tensions (Table 2) [17,69,70]. Similar reasoning can be used to explain the increase in initial bubble size with temperature where for cationic surfactants increasing temperature generally results in an increase in interfacial tension [17,70]. Despite the high temperature of 90 °C, the bubble size generated by C16-18N(CH3)C3N(CH3)2 was still comparable to those of a previously reported stable and viscoelastic C/W foams at room temperature generated with a mixture of high concentration (3.6%w/w) of sodium lauryl ethoxylated sulfate (SLES) and protonated decyldimethyl amine (C10DMA) at similar shear rates [29].

3.5.3. Foam apparent viscosity The apparent viscosity(lfoam) of C/W foams generated by C1618N(CH3)C3N(CH3)2 at different foam qualities is shown in Fig. 8A along with previous results [13,15] for the two other amines. For all surfactants, lfoam increased with foam quality up to 0.9 except for C16-18N(CH3)C3N(CH3)2 where it increased up to 0.95. Assuming a constant bubble size, increasing the internal phase fraction (quality) increases the number of lamellae and provides more resistance

to flow to increase the foam viscosity as described by Eq. (3) developed by Princen et al for bulk foam [20,71].




s l c_ R lfoam ¼ _0 þ 32ði  0:73Þle e c c

1=2 ð3Þ

where s0 is the yield stress, c_ is the shear rate, c is the interfacial tension, and R is the radius of a spherical droplet, le is the viscosity of external aqueous phase and £i is internal phase volume fraction. The foam destabilization observed at extremely high qualities will be discussed later. A key result of this study is that the lfoam of C/W foams generated by C16-18N(CH3)C3N(CH3)2 were higher those of the other two amine surfactants at all qualities (Fig. 8A). Given that all other parameters such as interfacial tension and shear rate in Eq. (2) are similar, the larger lfoam for C16-18N(CH3)C3N(CH3)2 is caused primarily by the viscous aqueous phase as shown in Fig. 3 [67,71,72]. This behavior is further illustrated in Fig. 6 where a sharp increase in lfoam from 10 cP to 27 cP was observed upon forming a viscous WLM phase as the salinity increased from 182 g/L NaCl to 22% TDS. As le increases, it will increase lfoam as shown in Eq. (3). On the other hand, as depicted from Fig. (S8), the apparent viscosity of foam decreased from 13 cP to 5 cP as the temperature increased from 25 to 120 °C partially due to the expected decrease in the viscosity of the aqueous phase (Fig. 3). Higher apparent viscosities were observed for C16-18N(CH3)C3N (CH3)2 relative to C12-14N(EO)2 when injecting the surfactant from the CO2 phase (Fig. 8B and Table 4) at all qualities tested. Interestingly, at 90% foam quality – were the mass concentration of surfactant in total injected fluid is similar - lfoam was about the same regardless of whether the surfactant was injected with the aqueous or the CO2 phase. Thus, the surfactant underwent transport to the interface in both cases and the mixing of CO2 surfactant solution

A.S. Elhag et al. / Journal of Colloid and Interface Science 522 (2018) 151–162

159

Fig. 8. Apparent viscosity of C/W foam by coinjection at 120 °C, 3400 psia and total superficial velocity 938 ft/day in 76 Darcy limestone pack (A) 1% w/w surfactant dissolved in 22% TDS brine pH 6.5 solution injected with CO2. (B) 0.2% w/w surfactant dissolved in CO2 injected with 22% TDS brine.

Table 4 Apparent viscosity of C/W foam stabilized by co-injecting 1% (w/w) C12-14N(EO)2 22% TDS brine pH 6.5 solution with CO2 or by co-injecting 0.2% (w/w) C12-14N(EO)2 CO2 solution with 22% TDS brine, at 120 °C, 3400 psia and total superficial velocity 938 ft/day in 76 Darcy limestone pack with 90% foam quality. Surfactant

From aqueous phase

From CO2 phase

C12-14N(EO)2 C16-18 N(CH3)C3N(CH3)2

14.5 cP 25.1 cP

13.2 cP 22.4 cP

and brine was sufficient to enable the formation of viscous aqueous lamellae. C16-18N(EO)C3N(EO)2 was not tested at these conditions because it was not soluble in CO2 (Fig. S6). In CO2 EOR studies, lab scale experiments have found that injection of surfactants from the CO2 phase can potentially enhance in situ foam formation, lower the adsorption of surfactant on minerals and ultimately increase oil recovery [13,73–75].

3.5.4. Foam stability In this section various interrelated destabilization mechanism will be discussed to emphasize the benefits of enhancing the stability with viscoelastic lamellae. In addition to raising the apparent viscosity, the presence of viscoelastic WLM phases will now be shown to enhance foam stability. The C16-18N(CH3)C3N(CH3)2C/W foam was stable for 20 h in that the Dsm increased by only 2 folds (28–57 nm) compared to a 4 folds for C16-18N(EO)C3N(EO)2 and C1214N(EO)2 with spherical micelles as shown in Fig. 9 and Table 3. The Ostwald ripening rate (X3 ), given by Lifshitz and Slyozov followed by Wagner (LSW) in Eq. (4) may be calculated from the slope of the linear fit of the time change in bubble diameter as depicted in Fig. 9 and Table 3.

dDsm 64cDdiff SV m ¼ F dt 9RT 3

X3 ¼

ð4Þ

where Ddiff is the diffusion coefficient, F is a correction factor, S and Vm are the bulk solubility and the molar volume of the dispersed phase, respectively. Clearly, the thicker and more viscous lamellae slowed down the diffusion of gas bubbles and lowered (Ddiff) as other parameters in Eq. (4) can be assumed constant. Moreover, the entangled WLMs are expected to provide a barrier for gas diffusion to further decrease (Ddiff) [25–29]. In Table 3, the rate of change in bubble volume with time (referred to as Ostwald ripening rate, X3 ) was reduced by half by the viscoelastic

aqueous phase of C16-18N(CH3)C3N(CH3)2. Similarly, Xue et al. measured a reduction in Ostwald ripening rate (6 folds) when WLMs where utilized relative to spherical micelles [29]. As shown in Fig. 8A, C16-18 N(CH3)C3N(CH3)2 with a viscous aqueous phase – analogous to wormlike micellar solutions - generated viscous foam up to a foam quality of 0.95 versus only 0.9 for the other surfactants. An increase in foam quality (u) decreases the radius at the plateau border and increases the capillary pressure (Pc) as given by Pc = c/Rf(1  u)0.5 [76], where Rf is the film radius. Consequently, at u = 0.95, Pc is extremely high. To stabilize lamellae against coalescence, the surfactant must provide a sufficient disjoining pressure (pd ) to oppose the capillary pressure. Given the comparable structure and interfacial behavior of C16-18N(EO) C3N(EO)2 and C16-18N(CH3)C3N(CH3)2, it is reasonable to assume that the contributions of electrostatic repulsion, which can be small due to the high ionic strength and van der Waals attraction to pd are similar. Therefore, some additional component of pd must partially explain why C16-18N(CH3)C3N(CH3)2 can stabilize C/W foam at higher qualities. Fameau et al. argued that large entangled WLMs provide additional steric repulsion that raises foam stability [77]. Also, Espert et al. and Klitzing et al. measured an increase in the thickness of the foam lamellae and the number of steps required for stepwise thinning of lamella formed by entangled WLMs [78] and polymers [79] indicating a higher disjoining pressure [78,79]. In the present study, the higher disjoining pressure combined with the high viscosity of the aqueous phase can further stabilize the foam by decreasing the drainage velocity (V) to keep a thicker lamella [26,77,19] according to Eq. (5) 2

V ¼

hf dhf DPfilm ¼ dt 3le R2f

ð5Þ

where DPfilm ¼ 2ðPc  Pd Þ and hf is the film thickness. 3.5.5. Effect of superficial velocity on C/W foam generation and rheology in a 1 Darcy glass bead pack A glass bead pack with a lower permeability of 1.2 Darcy and low superficial velocities were used to mimic foam generation mechanisms and rheology in consolidated porous media. Generation of strong foam requires lamellae mobilization where a minimum pressure gradient (MPG) overcomes the Laplace pressure (Dpmax ) for (n) number of lamellae over the length of the porous medium (L) as shown in Eq. (6) [80,81].

160

A.S. Elhag et al. / Journal of Colloid and Interface Science 522 (2018) 151–162

Fig. 9. Evolution of D3sm for foams of C12-14N(EO)2, C16-18N(EO)C3N(EO)2 or C16-18N(CH3)C3N(CH3)2with concentrations of 1 wt% in 22% TDS brine at 0.9 CO2 volume fraction, 25 °C and 3400 psia.

MPG ¼ Dpmax  

n 4c n ¼  L r min L

ð6Þ 

where r min is the minimum value of r  2=ð1=r1 þ 1=r2 Þ, and r1 and r2 are the principal radii of curvature of the lamella. The effect of superficial velocity (U) on C/W foam generation and viscosity for C16-18N(CH3)C3N(CH3)2 and C12-14N(EO)2 is shown in Fig. 10. Here, C16-18N(CH3)C3N(CH3)2 generated extremely viscous foam (700 cP) at a superficial velocity as low as 3 ft/day (11 mm/s). Assuming a similar MPG for both surfactants (similar interfacial tensions), the lower superficial velocity requirement may be attributed to an increase in the pressure gradient produced by the higher viscosity of the aqueous phase from the WLMs as shown in Fig. 4. From Darcy’s law (Eq. (7), the minimum superficial velocity required for foam generation is inversely proportional to the viscosity of the aqueous phase to achieve the same pressure gradient.

q=A ¼

kDP lL

ð7Þ

where q is the volumetric flow rate, L is the length of the sand pack and A is the cross sectional area. Similarly, in a 76 Darcy limestone pack, C16-18N(CH3)C3N(CH3)2 generated foam at lower superficial velocity (Fig. S9). Hence, C16-18N(CH3)C3N(CH3)2 is a more

promising candidate for CO2 mobility control due to its ability to generate more viscous foam even at extremely low flow rates encountered away from the well bore. The maximum foam apparent viscosity achieved in the 1 Darcy bead pack was 5 times higher for C16-18N(CH3)C3N(CH3)2 compared to C12-14N(EO)2. Beyond this maximum viscosity, shear thinning behavior was observed whereby apparent viscosity decreased with the increase in superficial velocity (Fig. 10). As in the case of a capillary tube, the displacement of more viscous lamellae separating the gas bubbles in the bead pack causes an increase in foam apparent viscosity according to model developed by Hirasaki and Lawson (Eq. (8) [71].


1 i lfoam ðnL Rc Þ 3le U 3  ¼ Ls nL þ 0:85 ½ r c =Rc Þ2 þ 1 le ðr c =Rc Þ c
1 3le U 3 pffiffiffiffiffiffi ð1  eNL Þ þ ðnL Rc Þ Ns ð1 þ eNL Þ c

ð8Þ

where Ls is the length of liquid slugs, nL is lamellae density, rc is the radius of curvature of gas-liquid interface, Rc is the capillary radius, U is the velocity of bubbles, Ns is a dimensionless number for the interfacial tension gradient effect, and NL is a dimensionless bubble length [71]. A plot of foam viscosity as a function of the gas superficial velocity (Ug) for C16-18N(CH3)C3N(CH3)2 in the shear thinning region is shown in Fig. S10 and the following relationship was found (lfoam / U g 0:73 ). The model by Hirasaki and Lawson (Eq. (8) describes the different components that contribute to the viscosity of foam in porous media, namely, deformation of bubbles and the interfacial tension gradient between the front and the back of the bubbles flowing through each channel in the porous media [71]. Both factors scale with U1/3. However, Falls et al. added an additional term to account for the constriction in porous media and found that the foam viscosity is proportional to the gas superficial velocity to the power 2/3 [82]. Therefore, the experimental exponent of 0.73 was quite close to the model prediction. 4. Conclusions

Fig. 10. Apparent viscosity of C/W foam stabilized by co-injecting 1% (w/w) C12-14N (EO)2, or C16-18N(CH3)C3N(CH3)2 22% TDS brine pH 4 solution with CO2, at 120 °C, 3400 psia and with 80% foam quality in 1.2 Darcy glass bead pack.

This paper systematically tests the hypothesis that modification of the structure of switchable amine surfactants by the addition of a second amine in the head group, methyl substitution on the protonated nitrogens and longer hydrocarbon tails can enhance the

A.S. Elhag et al. / Journal of Colloid and Interface Science 522 (2018) 151–162

solubility in brine, aqueous phase viscoelasticity, and C-W interfacial properties. As a result, C/W foam may be generated at lower superficial velocity with higher apparent viscosity and increased stability. The use of amine surfactants to generate highly stable and viscous C/W foams is beneficial for CO2 sequestration and EOR particularly in regions (ex. Middle East) where anionic and zwitterionic surfactants can become economically infeasible due to high adsorption on carbonates. Compared to previously reported switchable (pH- responsive) amine surfactants [13–15], the zero shear viscosity of aqueous solution of 1% w/w C16-18N(CH3)C3N(CH3)2 was 700 cP, is two orders of magnitude higher than for C12-14N(EO)2 and C16-18N(EO) C3N(EO)2. The longer tail combined with the small size of the CH3 (relative to EO) groups and high salinity increased the packing parameter of C16-18N(CH3)C3N(CH3)2 whereby it formed entangled wormlike micelles that increased the viscosity markedly. The effect of salinity and concentration of the aqueous phase on the zero shear viscosity was consistent with previous papers [30,56]. At higher salinities, the electrostatic repulsion between the protonated nitrogen heads was screened to effectively decrease the size of the head and increase the packing parameter and aid the formation of WLMs. Surfactant concentrations above 1% w/w further increased the entanglements of the micelles to increase the viscosity. The increase in headgroup hydrophilicity with a methyl substituted dual amine and greater basicity (ease of protonation) raised the solubility of the surfactant at higher pH values, especially at high temperature, relative to previously studied monoamines [13,24]. The comparison of the behavior of three amines provided novel insight into the effect of head type, substitution on the head and the length of the tail on the C-W interfacial properties. The small size of the amine head group compared to previously reported nonionic surfactants with similar tails resulted in higher adsorption at the C-W interface and lower interfacial tensions [18]. Remarkably, despite the larger head group of C16-18N(CH3)C3N (CH3)2 compared to previously studied cationic surfactants [13,15], the adsorption at the C-W interface (188 Å2/molecule) at high temperature was higher as a consequence of the long tail length that drove the surfactant from water to the interface and the relatively small head group area, given the small size of the CH3 substituents. As hypothesized in this paper, relative to spherical micelles formed with C12-14N(EO)2 and C16-18N(EO)C3N(EO)2 the highly viscous aqueous phase for C16-18N(CH3)C3N(CH3)2 led to higher apparent viscosities for bulk C/W foams [13,15]. As a consequence of slower drainage of the viscoelastic aqueous lamellae, the lamellae remained thick and provided more resistance to coalescence and Ostwald ripening resulting in long term stability up to 20 h, as shown by microscopy. The disjoining pressure was also increased as a result of the higher steric repulsion of the bulkier WLMs. Consequently, C/W foams could be stabilized even at very high foam qualities up to 0.98, where the capillary pressure is extremely large. The viscoelastic behavior also influenced foam formation in porous media to overcome the practical limitation of an undesirably large superficial velocity associated with a previously reported surfactant, C12-14N(EO)2 [24]. In a 1 Darcy bead pack, with the highly viscous aqueous phase, the minimum pressure gradient (MPG) for C/W foam generation was achieved at a superficial velocity as low as 3 ft/day, an order of magnitude lower than for the non-viscoelastic lamellae in the case of C12-14N(EO)2. Further characterization of this surfactant is being conducted to determine the effect of these different structural modifications on the thermal stability of these surfactants at temperatures up to 120 C. Furthermore, the adsorption of this surfactant on different minerals is being evaluated for successful implementation in field trials.

161

Acknowledgement This work was supported by the Abu Dhabi National Oil Company, the U. S. Department of Energy Center for Frontiers of Subsurface Energy Security, and the Welch Foundation F-1319. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.jcis.2018.03.037. References [1] R. Enick, D. Olsen, J. Ammer, W. Schuller, Mobility and conformance control for CO2 EOR via thickeners, foams, and gels- a literature review of 40 years of research and pilot tests, in: The Eighteenth SPE Improved Oil Recovery Symposium, Tulsa, Oklahoma, 2012. [2] K.P. Johnston, S.R.P. Da Rocha, Colloids in supercritical fluids over the last 20 years and future directions, J. Supercritical Fluids 47 (2009) 523–530. [3] A. Forooghi, A.A. Hamouda, T. Eilertsen, Co-optimization of CO2 EOR and sequestration process in a North Sea Chalk reservoir, in: SPE/EAGE Reservoir Characterization & Simulation Conference, 2009. [4] L.L. Schramm, Foams: Fundamentals and Applications in the Petroleum Industry, An American Chemical Society Publication, 1994. [5] A. Phillips, D. Couchman, J. Wilke, Successful field application of hightemperature rheology of CO2 foam fracturing fluids, in: Low Permeability Reservoirs Symposium, Society of Petroleum Engineers, 1987. [6] F.A. Adamsky, E.J. Beckman, Inverse emulsion polymerization of acrylamide in supercritical carbon dioxide, Macromolecules 27 (1994) 312–314. [7] E. Reverchon, R. Adami, S. Cardea, G. Della, Porta supercritical fluids processing of polymers for pharmaceutical and medical applications, J. Supercritical Fluids 47 (3) (2009) 484–492. [8] P. Mukherjee, S.K. Padhan, S. Dash, S. Patel, B.K. Mishra, Clouding behavior in surfactant systems, Adv. Colloid Interface Sci. 162 (1–2) (2011) 50–79. [9] S.S. Adkins, X. Chen, I. Chan, E. Torino, Q.P. Nguyen, A.W. Sanders, K.P. Johnston, Morphology and stability of CO2-in-water foams with nonionic hydrocarbon surfactants, Langmuir 26 (8) (2010) 5335–5348. [10] Y. Chen, A.S. Elhag, L. Cui, A.J. Worthen, P.P. Reddy, J.A. Noguera, A.M. Ou, K. Ma, M. Puerto, G.J. Hirasaki, CO2-in-water foam at elevated temperature and salinity stabilized with a nonionic surfactant with a high degree of ethoxylation, Indust. Eng. Chem. Res. 54 (16) (2015) 4252–4263. [11] M. Galimberti, M. Martino, M. Guenzi, G. Leonardi, A. Citerio, Thermal stability of ammonium salts as compatibilizers in polymer/layered silicate nanocomposites, e-Polymers 9 (1) (2009) 686–699. [12] L. Cui, D.M. Khramov, C.W. Bielawski, D.L. Hunter, P.J. Yoon, D.R. Paul, Effect of organoclay purity and degradation on nanocomposite performance, part 1: surfactant degradation, Polymer 49 (17) (2008) 3751–3761. [13] Y. Chen, A.S. Elhag, P.P. Reddy, H. Chen, L. Cui, A.J. Worthen, K. Ma, H. Quintanilla, J.A. Noguera, G.J. Hirasaki, Phase behavior and interfacial properties of a switchable ethoxylated amine surfactant at high temperature and effects on CO2-in-water foams, J. Colloid Interface Sci. 470 (2016) 80–91. [14] Y. Chen, A.S. Elhag, B.M. Poon, L. Cui, K. Ma, S.Y. Liao, P.P. Reddy, A.J. Worthen, G.J. Hirasaki, Q.P. Nguyen, S.L. Biswal, K.P. Johnston, Switchable nonionic to cationic ethoxylated amine surfactants for CO2 enhanced oil recovery in hightemperature, High-Salinity Carbonate Reserv. SPE J. 19 (2) (2014) 249–259. [15] A.S. Elhag, Y. Chen, P.P. Reddy, J.A. Noguera, A.M. Ou, G.J. Hirasaki, Q.P. Nguyen, S.L. Biswal, K.P. Johnston, Switchable diamine surfactants for CO2 mobility control in enhanced oil recovery and sequestration, Energy Procedia 63 (2014) 7709–7716. [16] J.B. Lawson, The adsorption of non-ionic and anionic surfactants on sandstone and carbonate, in: The Fifth Symposium on Improved Methods for Oil Recovery of the Society of Petroleum Engineerings of AIME, Tulsa, Oklahoma, USA, 1978, pp. 159–170. [17] M.J. Rosen, J.T. Kunjappu, Surfactants and Interfacial Phenomena, fourth ed., John Wiley & Sons Inc, Hoboken, New Jersey, 2012. [18] S.S. Adkins, X. Chen, A. Sanders, Q.P. Nguyen, K.P. Johnston, Effect of branching on the interfacial properties of nonionic hydrocarbon surfactants at the airwater and carbon dioxide-water interfaces, J. Colloid Interface Sci. 346 (2) (2010) 455–463. [19] I.B. Ivanov, P.A. Kralchevsky, Stability of emulsions under equilibrium and dynamic conditions, Colloids Surf. A: Physicochem. Eng. Aspects 128 (1997) 155–175. [20] H.M. Princen, A.D. Kiss, Rheology of foams and highly concentrated emulsions. IV. An experimental study of the shear viscosity and yield stress of concentrated emulsions, J. Colloid Interface Sci. 128 (1) (1989) 176–187. [21] Z.I. Khatib, G.J. Hirasaki, A.H. Falls, Effects of capillary pressure on coalescence and phase mobilites in foams flowing through porous media, SPE Reserv. Eng. 3 (3) (1988) 919–926. [22] Z. Xue, K. Panthi, Y. Fei, K.P. Johnston, K.K. Mohanty, CO2-soluble ionic surfactants and CO2 foams for high-temperature and high-salinity sandstone reservoirs, Energy Fuels 29 (9) (2015) 5750–5760.

162

A.S. Elhag et al. / Journal of Colloid and Interface Science 522 (2018) 151–162

[23] Y. Chen, A.S. Elhag, A.J. Worthen, P.P. Reddy, A.M. Ou, G.J. Hirasaki, Q.P. Nguyen, S.L. Biswal, K.P. Johnston, High temperature CO2-in-water foams stabilized with cationic quaternary ammonium surfactants, J. Chem. Eng. Data, 2016. [24] L. Cui, K. Ma, M. Puerto, A.A. Abdala, I. Tanakov, L.J. Lu, Y. Chen, A. Elhag, K.P. Johnston, S.L. Biswal, Mobility of ethomeen C12 and carbon dioxide (CO2) foam at high temperature/high salinity and in carbonate cores, SPE J. (2016). [25] C. Monteux, G.G. Fuller, V. Bergeron, Shear and dilational surface rheology of oppositely charged polyelectrolyte/surfactant microgels adsorbed at the airwater interface. Influence on foam stability, J. Phys. Chem. B 108 (42) (2004) 16473–16482. [26] D. Varade, D. Carriere, L. Arriaga, A.-L. Fameau, E. Rio, D. Langevin, W. Drenckhan, On the origin of the stability of foams made from catanionic surfactant mixtures, Soft Matter 7 (14) (2011) 6557–6570. [27] A.L. Fameau, A. Saint-Jalmes, F. Cousin, B. Houinsou Houssou, B. Novales, L. Navailles, F. Nallet, C. Gaillard, F. Boué, J.P. Douliez, Smart foams: switching reversibly between ultrastable and unstable foams, Angew. Chem. Int. Ed. 50 (36) (2011) 8264–8269. [28] Z. Xue, A. Worthen, A. Qajar, I. Robert, S.L. Bryant, C. Huh, M. Prodanovic´, K.P. Johnston, Viscosity and stability of ultra-high internal phase CO2-in-water foams stabilized with surfactants and nanoparticles with or without polyelectrolytes, J. Colloid Interface Sci. 461 (2016) 383–395. [29] Z. Xue, A.J. Worthen, C. Da, A. Qajar, I.R. Ketchum, S. Alzobaidi, C. Huh, M. Prodanovic´, K.P. Johnston, Ultradry carbon dioxide-in-water foams with viscoelastic aqueous phases, Langmuir 32 (1) (2016) 28–37. [30] R. Zana, E.W. Kaler, Giant Micelles: Properties and Applications, CRC Press, 2007. [31] J. Yang, Viscoelastic wormlike micelles and their applications, Curr. Opin. Colloid Interface Sci. 7 (5) (2002) 276–281. [32] L.M. Walker, Rheology and structure of worm-like micelles, Curr. Opin. Colloid Interface Sci. 6 (5) (2001) 451–456. [33] Z. Chu, C.A. Dreiss, Y. Feng, Smart wormlike micelles, Chem. Soc. Rev. 42 (17) (2013) 7174–7203. [34] Z. Chu, Y. Feng, pH-switchable wormlike micelles, Chem. Commun. 46 (47) (2010) 9028–9030. [35] X. Su, M.F. Cunningham, P.G. Jessop, Switchable viscosity triggered by CO2 using smart worm-like micelles, Chem. Commun. 49 (26) (2013) 2655–2657. [36] Y. Liu, P.G. Jessop, M. Cunningham, C.A. Eckert, C.L. Liotta, Switchable surfactants, Science 313 (5789) (2006) 958–960. [37] P. Brown, C.P. Butts, J. Eastoe, Stimuli-responsive surfactants, Soft Matter 9 (8) (2013) 2365–2374. [38] Y. Zhang, Y. Feng, Y. Wang, X. Li, CO2-switchable viscoelastic fluids based on a pseudogemini surfactant, Langmuir 29 (13) (2013) 4187–4192. [39] Y. Zhang, Y. Feng, J. Wang, S. He, Z. Guo, Z. Chu, C.A. Dreiss, CO 2-switchable wormlike micelles, Chem. Commun. 49 (43) (2013) 4902–4904. [40] Y. Chen, T.L. Pope, J.C. Lee, Novel CO2-emulsified viscoelastic surfactant fracturing fluid system, in: SPE European Formation Damage Conference, Society of Petroleum Engineers, 2005. [41] X. Sun, X. Liang, S. Wang, Y. Lu, Experimental study on the rheology of CO2 viscoelastic surfactant foam fracturing fluid, J. Petrol. Sci. Eng. 119 (2014) 104– 111. [42] D. Li, B. Ren, L. Zhang, J. Ezekiel, S. Ren, Y. Feng, CO 2-sensitive foams for mobility control and channeling blocking in enhanced WAG process, Chem. Eng. Res. Des. 102 (2015) 234–243. [43] W.R. Rossen, Foams in enhanced oil recovery, in: R.K. Prudhomme, S.A. Khan (Eds.), Foams: Theory, Measurements, and Applications, Marcel Dekker, New York, 1996. [44] J.P. Heller, CO2 foams in enhanced oil recovery, Foams: Fundamentals and Applications in the Petroleum Industry 242(Foams: Fundamentals and Applications in the Petroleum Industry) (1994) 201–234. [45] M. Algharaib, Potential applications of CO2-EOR in the middle east, in: The SPE Middle East Oil & Gas Show and Conference, Kingdom of Bahrain, 2009. [46] A. Nairn, A. Alsharhan, Sedimentary Basins and Petroleum Geology of the Middle East, Elsevier, 1997. [47] G.J. Hirasaki, J.B. Lawson, An electrostatic approach to the association of sodium and calcium with surfactant micelles, SPE Reserv. Eng. 1 (2) (1986) 119–130. [48] K. Golec, E.C. Hill, P. Kazemi, R.O. Skold, Oil/water partition for some hydroxyalkylamines and antimicrobial efficacy in metalworking coolants, Tribol. Int. 22 (6) (1989) 375–382. [49] D. Matulis, V.A. Bloomfield, Thermodynamics of the hydrophobic effect. I. coupling of aggregation and pKa shift in solutions of aliphatic amines, Biophys. Chem. 93 (1) (2001) 37–51. [50] D. Matulis, Thermodynamics of the hydrophobic effect. III. Condensation and aggregation of alkanes, alcohols, and alkylamines, Biophys. Chem. 93 (2001) 67–82. [51] D. Rubingh, Cationic Surfactants: Physical Chemistry, CRC Press, 1990. [52] C. Tanford, Solubility of hydrocarbons in water, in: C. Tanford (Ed.), Hydrophobic Effect, John Wiley & Sons Inc, USA, 1980, pp. 5–13. [53] R. Antón, N. Garcés, A. Yajure, A correlation for three-phase behavior of cation1C surfactant-oil-water systems, J. Dispers. Sci. Technol. 18 (5) (1997) 539–555.

[54] C. Tanford, The effect of temperature: anomalous entropy and heat capacity, in: C. Tanford (Ed.), Hydrophobic Effect, John Wiley & Sons Inc, USA, 1980, pp. 21–28. [55] H.S. Frank, M.W. Evans, Free volume and entropy in condensed systems III. Entropy in binary liquid mixtures; partial molal entropy in dilute solutions; structure and thermodynamics in aqueous electrolytes, J. Chem. Phys. 13 (1945) 507–532. [56] C.A. Dreiss, Wormlike micelles: where do we stand? recent developments, linear rheology and scattering techniques, Soft Matter 3 (8) (2007) 956–970. [57] S.R. Raghavan, E.W. Kaler, Highly viscoelastic wormlike micellar solutions formed by cationic surfactants with long unsaturated tails, Langmuir 17 (2) (2001) 300–306. [58] G.C. Kalur, S.R. Raghavan, Anionic wormlike micellar fluids that display cloud points: rheology and phase behavior, J. Phys. Chem. B 109 (18) (2005) 8599– 8604. [59] Y. Zhang, Z. Chu, C.A. Dreiss, Y. Wang, C. Fei, Y. Feng, Smart wormlike micelles switched by CO2 and air, Soft Matter 9 (27) (2013) 6217–6221. [60] R. Nagarajan, Molecular packing parameter and surfactant self-assembly: the neglected role of the surfactant tail, Langmuir 18 (1) (2002) 31–38. [61] A. Khatory, F. Lequeux, F. Kern, S. Candau, Linear and nonlinear viscoelasticity of semidilute solutions of wormlike micelles at high salt content, Langmuir 9 (6) (1993) 1456–1464. [62] G. Sartori, D.W. Savage, Sterically hindered amines for carbon dioxide removal from gases, Ind. Eng. Chem. Fundam. 22 (2) (1983) 239–249. [63] K.A. Consani, R.D. Smith, Observations on the solubility of surfactants and related molecules in carbon dioxide at 50 °C, J. Supercritical Fluids 3 (2) (1990) 51–65. [64] M.L. O’Neill, Q. Cao, M. Fang, K.P. Johnston, S.P. Wilkinson, C.D. Smith, J.L. Kerschner, S.H. Jureller, Solubility of homopolymers and copolymers in carbon dioxide, Ind. Eng. Chem. Res. 37 (1998) 3067–3079. [65] J.D. Holmes, K.J. Ziegler, M. Audriani, C.T. Lee Jr., P.A. Bhargava, D.C. Steytler, K. P. Johnston, Buffering the aqueous phase pH in water-in-CO2 microemulsions, J. Phys. Chem. B 103 (1999) 5703–5711. [66] S. Bachu, D.B. Bennion, Interfacial tension between CO2, freshwater, and brine in the range of pressure from (2 to 27) MPa, temperature from (20 to 125) °C, and water salnity from (0 to 334 000) mgL1, J. Chem. Eng. Data 54 (3) (2009) 765–775. [67] J.P. Heller, M.S. Kuntamukkula, Critical review of the foam rheology literature, Ind. Eng. Chem. Res. 25 (1987) 318–325. [68] P. Walstra, P.E.A. Smulders, Emulsion formation, in: B.P. Binks (Ed.), Modern Aspects of Emulsion Technology, The Royal Society of Chemistry, Cambridge, 1998, pp. 56–99. [69] S. Oh, D. Shah, Effect of counterions on the interfacial tension and emulsion droplet size in the oil/water/dodecyl sulfate system, J. Phys. Chem. 97 (2) (1993) 284–286. [70] J. Floury, A. Desrumaux, J. Lardieres, Effect of high-pressure homogenization on droplet size distributions and rheological properties of model oil-in-water emulsions, Innovative Food Sci. Emergy Technol. 1 (2) (2000) 127–134. [71] G.J. Hirasaki, J.B. Lawson, Mechanisms of foam flow in porous media: apparent viscosity in smooth capillaries, Soc. Petrol. Eng. J. (1985) 176–190. [72] R.K. Prud’homme, S.A. Khan, Experimental Results on Foam Rheology, Surfactant Science Series 57(Foams) (1996) (217–4). [73] V.Q. Le, Q.P. Nguyen, A.W. Sanders, A novel foam concept with CO2 dissolved surfactants, in: SPE/DOE Improved Oil Recovery Symposium, Tulsa, Oklahoma, USA, 2008, pp. 1–15. [74] G. Ren, H. Zhang, Q.P. Nguyen, Effect of surfactant partitioning between CO2 and water on CO2 mobility control in hydrocarbon reserviors, in: The SPE Enhanced Oil Recovery Conference, Kuala Lumpur, Malaysia, 2011, pp. 1–15. [75] X. Chen, S.S. Adkins, Q.P. Nguyen, A.W. Sanders, K.P. Johnston, Interfacial tension and the behavior of microemulsions and macroemulsions of water and carbon dioxide with a branched hydrocarbon nonionic surfactant, J. Supercritical Fluids 55 (2) (2010) 712–723. [76] E. Rio, A.L. Biance, Thermodynamic and mechanical timescales involved in foam film rupture and liquid foam coalescence, ChemPhysChem 15 (17) (2014) 3692–3707. [77] A.-L. Fameau, A. Salonen, Effect of particles and aggregated structures on the foam stability and aging, Comptes Rendus Phys. 15 (8) (2014) 748–760. [78] A. Espert, R.V. Klitzing, P. Poulin, A. Colin, R. Zana, D. Langevin, Behavior of soap films stabilized by a cationic dimeric surfactant, Langmuir 14 (15) (1998) 4251–4260. [79] R.V. Klitzing, A. Espert, A. Colin, D. Langevin, Comparison of different polymerlike structures in the confined geometry of foam films, Colloids Surf. A: Physicochem. Eng. Aspects 176 (1) (2001) 109–116. [80] D. Tanzil, G.J. Hirasaki, C.A. Miller, Conditions for foam generation in homogeneous porous media, in: The SPE/DOE Improved Oil Recovery Symposium, Tulsa, Oklahoma, USA, 2002. [81] P.A. Gauglitz, F. Friedmann, S.I. Kam, W.R. Rossen, Foam generation in homogeneous porous media, Chem. Eng. Sci. 57 (2002) 4037–4052. [82] A.H. Falls, J.J. Musters, J. Ratulowskl, The apparent viscosity of foams in homogeneous bead packs, SPE Reserv. Eng. (1989) 155–164.