Langmuir 2007, 23, 12071-12078
12071
Biocompatible, Lactide-Based Surfactants for the CO2-Water Interface: High-Pressure Contact Angle Goniometry, Tensiometry, and Emulsion Formation Balaji Bharatwaj, Libo Wu, and Sandro R. P. da Rocha* Department of Chemical Engineering and Materials Science, Wayne State UniVersity, 5050 Anthony Wayne Dr., Detroit, Michigan 48202 ReceiVed June 20, 2007. In Final Form: August 29, 2007 The unique properties of compressed CO2, including its low cost, nontoxicity, easily tunable solvent strength, and favorable transport properties, make it an environmentally attractive alternative to volatile organic solvents. Suitable surface-active species can be utilized to realize the full potential of clean, CO2-based technologies, by helping to overcome the low solubility typically associated with many solutes of interest in CO2. In this work we synthesize and investigate the interfacial activity of a series of nonionic amphiphiles with a biocompatible and biodegradable CO2-phile at both the CO2-water (C|W) and CO2-water-solid (C|W|S) interfaces. We developed a high-pressure pendant drop tensiometer and contact angle goniometer that allows us to measure both tension and contact angle in tandem. The tension of the C|W interface was measured in the presence of the lactide (LA)-based surface active agents with varying molecular weight and hydrophilic-to-CO2-philic ratios. Emulsion studies with an optimum balanced surfactant were performed. The contact angle of water droplets against a silane-modified (hydrophobic) substrate under CO2 atmosphere was also measured in presence of a selected LA-based amphiphile. The results demonstrate that the nonionic copolymers with the biodegradable and biocompatible LA-based group can significantly reduce the tension of the C|W interface. The LA-based surface active species are also capable of forming stable emulsions of water and CO2 and reducing the angle of the three-phase C|W|S contact line.
Introduction Supercritical or near-critical fluids are environmentally benign candidates for the replacement of volatile organic solvents.1 They possess solvent strength that can be tuned with not only temperature but also pressure.2 The ability to achieve wide variations in density with small changes in pressure has tremendous practical relevance, as for example by facilitating downstream separation processes.3 CO2 is a particularly attractive compressible solvent because of its low cost, nontoxicity, mild critical point, and nonflammability.4,5 CO2-based systems have shown promise in a range of chemical, separation, and materials processing technologies including extraction,6,7 polymerization,8-10 organic reactions,11 biocatalysis,12,13 nanoparticle synthesis and separations,14-16 and cleaning in the microelectronic industry.17,18 * To whom correspondence should be addressed. Tel.: +1-313-5774669. Fax: +1-313-577-3810. E-mail:
[email protected]. (1) Dickson, J. L.; Gupta, G.; Hozorov, T. S.; Binks, B. P.; Johnston, K. P. Langmuir 2006, 22, 2161-2170. (2) Bharatwaj, B.; da Rocha, S. R. P. Braz. J. Chem. Eng. 2006, 23, 183-190. (3) Behles, J. A.; DeSimone, J. M. Pure Appl. Chem. 2001, 73, 1281-1285. (4) Dickson, J. L.; Smith, G. P., Jr.; Dhanuka, V. V.; Srinivasan, V.; Stone, M. T.; Rossky, P. J.; Behles, J. A.; Keiper, J. S.; Xu, B.; Johnson, C.; DeSimone, J. M.; Johnston, K. P. Ind. Chem. Eng. Res. 2005, 44, 1370-1380. (5) Yang, Y.; Liu, D.; Xie, Y.; Lee, L. J.; Tomasko, D. L. AdV. Mater. 2007, 19, 251-254. (6) Beckman, E. Ind. Chem. Eng. Res. 2003, 42, 1598-1602. (7) Mellein, B. R.; Brennecke, J. F. J. Phys. Chem. B 2007, 111, 4837-4843. (8) DeSimone, J. M.; Guan, Z.; Elsbernd, C. S. Science 1992, 257, 945. (9) Clarke, M. J.; Harrison, K. L.; Johnston, K. P.; Howdle, S. M. J. Am. Chem. Soc. 1997, 119, 6399-6406. (10) Yang, J.; Wang, W.; Sazio, P. J. A.; Howdle, S. Eur. Polym. J. 2007, 43, 663-667 (11) Jessop, P. G.; Ikariya, T.; Noyori, R. Nature 1994, 368, 231. (12) Holmes, J. D.; Steytler, D. C.; Rees, G. D.; Robinson, B. H. Langmuir 1998, 14, 6371-6376. (13) Villaroya, S.; Thurecht, K. J.; Heise, A.; Howdle, S. Chem. Commun. 2007. (14) Holmes, J. D.; Bhargava, P. A.; Korgel, B. A.; Johnston, K. P. Langmuir 1999, 15, 6613-6615. (15) Anand, M.; Odom, L. A.; Roberts, C. B. Langmuir 2007, 23, 7338-7343.
In spite of all the recognized advantages of CO2-based technologies, their applicability is often limited by the fact that CO2 is a poor solvent to many solutes of interest.19 Suitable amphiphiles can be used to overcome some of the limitations associated with processing species that are weakly solvated by CO2. Within that context, both the CO2-liquid (C|L)and the CO2-liquid-solid (C|L|S) interfaces are of relevance.1,20, Polymeric3,21 and small molecular weight amphiphiles22,23 can be used in the formation and stabilization of CO2-based dispersions.6 Copolymer surfactants have been successfully employed in stabilizing growing polymer nuclei in dispersion polymerizations in CO2.24 Large molecular weight amphiphiles have been also used to stabilize the CO2-water interface.25-27 Dispersions of water and CO2 have found several potential uses, including media for inorganic reactions,28 template for the (16) Sane, A.; Thies, M. C. J. Supercrit. Fluids 2007, 40, 134-143. (17) Denison, G. M.; Jones, C., III; DeYoung, J.; Gross, S.; McClain, J.; Zannoni, L.; Hicks, E.; Wood, C.; Boggiano, M. K.; Visintin, P.; Bessel, C.; Schauer, C.; DeSimone, J. M. PMSE Prepr. 2004, 90, 152-153. (18) King, J. W.; Williams, L. L. Curr. Opin. Solid State Mater. Sci. 2003, 7, 413-424. (19) da Rocha, S. R. P.; Harrison, K. L.; Johnston, K. P. Langmuir 1999, 15, 419-28. (20) da Rocha, S. R. P.; Johnston, K. P. Langmuir 2000, 16, 3690-3695. (21) Yazdi, A. V.; Lepilleur, C.; Singley, E. J.; Liu, W.; Adamsky, F. A.; Enick, R. M.; Beckman, E. J. Fluid Phase Equilib. 1996, 117, 297-303. (22) da Rocha, S. R. P.; Dickson, J.; Cho, D.; Rossky, P. J.; Johnston, K. P. Langmuir 2003, 19, 3114-3120. (23) Eastoe, J.; Gold, S. PCCP 2005, 7, 1352-1362. (24) McClain, J. B.; Betts, D. E.; Canelas, D. A.; Samulski, E. T.; DeSimone, J. M.; Londono, J. D.; Cochran, H. D.; Wignall, G. D.; Chillura-Matino, D.; Triolo, R. Science 1996, 274, 2049-2052. (25) Hutton, B. H.; Perera, J. M.; Grieser, F.; Stephens, G. W. Colloids Surf., A 1999, 146, 227-241. (26) Harrison, K. L.; Goveas, J.; Johnston, K. P.; O’Rear, E. A. I. Langmuir 1994, 10, 3536-3541. (27) Ryoo, W.; Webber, S. E.; Johnston, K. P. Ind. Chem. Eng. Res. 2003, 42, 6348-6358. (28) Poliakoff, M.; George, M. W.; Howdle, S. M. Chemistry under Extreme or Non-Classical Conditions; van Eldik, R., Hubbard, C. D., Eds.; Wiley: New York, 1997; pp 189-218.
10.1021/la701831v CCC: $37.00 © 2007 American Chemical Society Published on Web 10/18/2007
12072 Langmuir, Vol. 23, No. 24, 2007 Scheme 1. Synthesis of the Triblock Copolymers from
Bharatwaj et al. DL-Lactide
synthesis of porous materials,29 cleaning in the garment30 and microelectronic industry,18 and electroless deposition processes.31 CO2-based free meniscus coatings,32 photoresist drying,33 and chemical mechanical planarization (CMP)34 are other examples where knowledge of both the C|L and C|S|L interfaces is needed for process design and optimization. Central to the design of interfacially active amphiphiles is the selection of CO2-philic moieties that not only interact highly with CO2 but are also inexpensive, environmentally acceptable and, in certain applications, biocompatible.6 While fluorinated chemistries are accepted as highly CO2-philic moieties and have been shown to stabilize both aqueous35 and nonaqueous36 dispersions in CO2, they are expensive37 and potentially toxic.38 The ability of CO2 to solvate polymers that may serve as surfactant tail-groups (CO2-philies) may be qualitatively predicted on the basis of the surface tension (σ) of the pure polymer melt.39 For nonpolar or slightly polar polymers, σ is a measure of van der Waals forces and is related to the cohesive energy density (δ). Polymers with low δ and σ are most soluble in CO2.19 Computer simulations,40 spectroscopy,41 tensiometry,19,42 contact angle goniometry,1 and thermodynamic principles43 have been applied in an effort to understand the interactions between CO2 and potential CO2-philes. Guided by such experimental/computational efforts, along with heuristic principles, novel polymeric surfactants for CO2-based systems have been proposed.19,43,44 The selection of CO2-philes is clearly of great relevance in surfactant design. However, understanding the activity and balance of the amphiphiles at interfaces of interest, such as that of CO2-monomer (emulsion polymerization),42 CO2-water (emulsions and microemulsions),19,45 and CO2-water-substrate is equally as important.1,46 High-pressure pendant drop tensiometry and contact angle goniometry can be used to directly assess the activity of the amphiphiles at the C|L and C|S|L (29) Cooper, A. I. AdV. Mater. 2003, 15, 1049-1059. (30) DeSimone, J. Science 2002, 297, 799-803. (31) Fernandes, N. E.; Fisher, S. M.; Poshusta, J. C.; Vlachos, D. G.; Tsapatsis, M.; Watkins, J. J. Chem. Mater. 2001, 13, 2023-2031. (32) Novick, B. J.; deSimone, J. M.; Carbonell, R. G. Ind. Eng. Chem. Res. 2004, 43, 515-524. (33) GoldFarb, D. L.; de Pablo, J. J.; Nealey, P. F.; Simons, J. P.; Moreau, W. M.; Angelopoulos, M. J. Vac. Sci. Technol., B 2000, 18, 3313-3317. (34) Vistinin, P. M.; Carbonell, R. G.; Schauer, C. K.; DeSimone, J. M. Langmuir 2005, 21, 4816-4823. (35) Johnston, K. P.; Harrison, K. L.; Clarke, M. J.; Howdle, S.; Heitz, M. P.; Bright, F. V.; Carlier, C.; Randolph, T. W. Science 1996, 271, 624-626. (36) Shiho, H.; DeSimone, J. M. Macromolecules 2001, 34, 1198-1203. (37) Eastoe, J.; Dupont, A.; Steytler, D. M. Curr. Opin. Colloid Interface Sci. 2003, 8, 267-273. (38) Woods, H. M.; Silva, M.; Nouvel, C.; Shakesheff, K. M.; Howdle, S. M. J. Mater. Chem. 2004, 14, 1663-1678. (39) O’Neill, M. L.; Cao, Q.; Fang, M.; Johnston, K. P.; P, W. S.; Smith, C. D.; Kerschner, J. L.; Jureller, S. H. Ind. Chem. Eng. Res. 1998, 37, 3067-3079. (40) da Rocha, S. R. P.; Stone, M. T.; Rossky, P. J.; Johnston, K. P. J. Phys. Chem. B 2003, 107, 10185-10192. (41) Kazarian, S. G.; Vincent, M. F.; Bright, F. V.; Liotta, C. L.; Eckert, C. A. J. Am. Chem. Soc. 1996, 118, 1729-1736. (42) Harrison, K. L.; Sanchez, I. P.; Johnston, K. P. Langmuir 1996, 12, 26372644. (43) Sarbu, T.; Beckman, E.; Styranec, T. J. Ind. Chem. Eng. Res. 2000, 39, 4678-4683. (44) Raveendran, P.; Wallen, S. J. Am. Chem. Soc. 2002, 124, 7274-7275.
and Poly(ethylene glycol) with Stannous Octoate as Catalyst
interfaces. Both the hydrophilic-CO2-philic balance (HCB) and area occupied/surfactant molecule at the C|L interface can be determined by tensiometry.19,47 The effect of surfactants at the three-phase contact line can be assessed by contact angle (θ) goniometry.48-50 A great deal of information about the C|L|S interface can be obtained by the combined knowledge of both interfacial tension (γ) and θ.1 The overall goal of this work was to investigate the activity of a series of biocompatible triblock copolymer surfactants containing biodegradable lactide (LA)-based tails at the CO2water (C|W) and CO2-water-solid (C|W|S) interfaces. Surfactants with a general structure LAnEOmLAn, where m and n are the number of LA and ethylene oxide (EO) repeat units, respectively, were synthesized for this work. The γ and θ was measured in situ, in a tandem high-pressure pendant drop tensiometer/contact angle goniometer recently developed in our laboratories. The effects of both the size of the hydrophilic block (EOm, with m ) 7, 14, and 23) and the HCB (EO/LA ratio) were investigated. Optimum surfactant balance (maximum activity) was determined for each series. We also studied the ability of a selected amphiphile in forming and stabilizing emulsions of water and CO2. To assess the degree of CO2-philicity of the LA moiety, the tensiometric results of the LAnEOmLAn series were compared and contrasted with those from methyl (CH2)- and propylene oxide (PO)-based surfactants. The effect of LA1EO14LA1 on the water contact angle on a hydrophobic substrate under compressed CO2 atmosphere was also investigated. Experimental Section Materials. CO2 of bone-dry quality (99.8% pure) was purchased from Airgas Inc. Ultrapure water (NANOpure DIamond UV; Barnstead International) with a resistivity of 17.7 MΩ·cm and surface tension of 72.9 mN‚m-1 at 298 K was used in all experiments. Chloroform (99.9%), acetone (99.8%), sulfuric acid (95.8%), hydrogen peroxide (30%), and hexane (99.9%) were purchased from Fisher Scientific. Tergitol, TMN-6 (R(OCH2CH2)8OH, R ) CH3(CH(CH3)CH2)2CH(CH2CH(CH3)2), and TMN-10 (R ) EO12) (90% assay) were purchased from Fluka Chemicals. The Pluronic L6 series surfactants with the general structure EOnPOmEOn (where PO is propylene oxide and EO is ethylene oxide and n and m are the average number of repeat units) were kindly donated by BASF. They were used as received. Polyethylene glycol with molecular weight 300 was purchased from Acros Organics, and PEG 1000 from GFS Chemicals. PEGs with molecular weights 600 and 2000 were purchased from Fluka Chemicals. All PEGs were dried by azeotropic distillation with toluene to remove traces amount of water. DL-Lactide was purchased from Frinton Laboratories, Inc., and (45) Eastoe, J.; Alison, P.; Nave, S.; Steytler, D. C.; Robinson, B. H.; Rumsey, E. H.; Thorpe, M.; Heenan, R. K. J. Am. Chem. Soc. 2001, 123, 988-989. (46) Carbonell, R. G.; Zweber, A. G.; DeSimone, J. M.; Jones, C.; DeYoung, J.; McClain, J. Crit. ReV. Solid State Mater. Sci. 2004, 29, 97-109. (47) Sagisaka, M.; Fujii, T.; Ozaki, Y.; Yoda, S.; Takebayashi, Y.; Kondo, Y.; Yoshino, N.; Sakai, H.; Abe, M.; Otake, K. Langmuir 2004, 20, 2560-2566. (48) Wagner, R.; Wu, Y.; Berlepsch, H. V.; Perepelittchenko, L. Appl. Organomet. Chem. 2000, 14, 177-188. (49) Binks, B. P.; Clint, J. H. Langmuir 2002, 18, 1270-1273. (50) Dutschk, V.; Sabbatovskiy, K. G.; Stolz, M.; Grundke, K.; Rudoy, V. M. J. Colloid Interface Sci. 2003, 267, 456-462.
Surfactants for the CO2-Water Interface
Langmuir, Vol. 23, No. 24, 2007 12073
Table 1. Surfactant Structure and Tension at the CO2-Water Interface (γC|W) at 298 K, 0.85 g‚mL-1 CO2 Density, and 0.01 wt % Surfactant Concentration surfact
MW (g‚mol-1)
struct
% EO
γC|W (mN‚m-1)
Pluronic L61 Pluronic L62 Pluronic L64 Pluronic P65 Pluronic F68 PLEG300-20 PLEG300-50 PLEG300-67.5 PEG300 PLEG600-20 PLEG600-57.5 PLEG600-80 PEG600 PLEG1000-10 PLEG1000-50 PLEG1000-90 PEG1000 PLA2000 TMN-6 TMN-10
2000 2500 2900 3500 8400 1452 588 444 300 3048 1176 744 600 8632 2010 1127 1000 2000 552 694
EO2.5PO31EO2.5 EO6PO34EO6 EO13PO30EO13 EO20PO30EO20 EO76PO30EO76 LA8EO7LA8 LA2EO7LA2 LA1EO7LA1 EO7 LA17EO14LA17 LA4EO14LA4 LA1EO14LA1 EO14 LA53EO23LA53 LA7EO23LA7 LA1EO23LA1 EO23 LA28 5b-C12EO8 5b-C12EO12
10 20 40 50 80 20 50 67.5 100 20 57.5 80 100 10 50 90 100 0 64 73
11.1 5.9 7.9 8.8 10.5 11.1 10.6 13.5 NA 5.5 2.5 7.5 12.6 4.9 3.6 8.5 9.5 12.2 12 6.5
recrystallized from ethyl acetate. Stannous octoate (initiator) was purchased from Sigma. Octyltrichlorosilane (C8TS, 97%) was purchased from Aldrich Chemicals Ltd. Glass slides (22 mm2, No. 2) were purchased from Corning Labware & Equipment. Surfactant Synthesis. The synthesis for the triblock copolymer surfactants LAnEOmLAn is shown in Scheme 1.51 Appropriate amounts of PEG, DL-lactide, and stannous octoate were initially charged into a 25 mL round-bottomed flask. The reaction was carried out at 433 K under nitrogen atmosphere for 10 h and subsequently cooled to room temperature. The products were then dissolved in methylene chloride, precipitated from ethyl ether twice, filtered out, and dried under vacuum at 393 K for 12 h. 1H NMR was used to confirm the molecular structure of the resulting triblockcopolymersurfactants.Theintegralintensitiesof-CH2CH2O(at 3.6 ppm) from PEG and -COCH(CH3)O- (at 1.5 ppm) from lactide were used to determine the degree of polymerization (m).52 A list with the LA-surfactants synthesized in this work, their MW, % EO, and interfacial tension at the CO2-water interface (γC|W) is shown in Table 1. The structure and γC|W determined for the other amphiphiles (PO-based and CH2-based) are also listed in Table 1. Interfacial Activity. We have developed a tandem high-pressure pendant drop tensiometer and contact angle goniometer. The apparatus is schematically illustrated in Figure 1. The main components of the setup are as follows: a variable-volume pressure cell (to control pressure independently of temperature or composition) equipped with two side windows for extraction of the profiles of the droplets (hanging drop for tension or drop for contact angle) and a front window for visualization of the inner contents of the cell; a six port switching valve (VICI); a computer-controlled pressure pump (ISCO SERIES D). The view cell also has a translation stage with a stainless steel plank. The plank was used to support the silanemodified substrate (described below), where water droplets were placed for the contact angle measurements. The light source and video camera were mounted on an optical rail for alignment. A computer was utilized to digitize the droplet (pendant or contact angle) images. The temperature inside the pressure cell was controlled using a temperature controller to (0.4 K. To ensure accuracy of the reading, the thermocouple was placed inside the pressure cell, close to the capillary used to form the droplet. The temperature of the cell was maintained with a heating tape (36050-10, Barnstead/Thermolyne) wrapped around it. A pressure transducer (Sensotec FP2000) was used to monitor the pressure to (0.04 MPa. The transducer was placed in the front part of the pressure cell. Densities of pure water and CO2 were used to determine the tension.53 (51) Deng, X. M.; D, X. C.; Cheng, L. M. J. Appl. Polym. Sci., Part C: Polym. Lett. 1990, 28, 411-416. (52) Mohammadi-Rovshandeh, J.; Farnia, S. M. F.; Sarboulouki, M. N. J. Appl. Polym. Sci. 1999, 47, 2004-2009.
Interfacial Tension. The binary interfacial tension of the C|W interface (γC|W) was determined using the pendant drop analysis. The pressure cell was initially filled with CO2. CO2 was saturated with excess water present on the bottom of the pressure cell, and subsequently the cell was allowed to equilibrate at the desired temperature. Water droplets were then generated at the tip of a stainless steel capillary using an HPLC pump (Waters 501). Once a droplet was formed at the tip of a capillary with known OD, the water reservoir was isolated from the pressure cell using the six-port injection valve. Immediately after droplet generation, the pendant drop was digitized. Measurements were performed until the tension reached equilibrium.54 Automated software (KSV 2001) was used to fit the droplet profile to the Laplace equation. The whole profile was used in the fitting algorithm, thus providing very accurate values for γ. The γ reported here are averages of at least three different (completely independent) runs. For measurement of the γ in the presence of surfactant, known masses of the amphiphile and CO2 were initially added to the cell and then allowed to equilibrate prior to the formation of the water droplet. Subsequent steps were as discussed in the measurement of the C|W binary interfacial tension. All runs were above the surfactant cloud point in pure CO2 and CO2 saturated with water. Contact Angle. To measure the contact angle (θ), droplets of water generated at the tip of the capillary were transferred onto a silane-modified substrate. The transfer of the droplet was accomplished by translating the stage until it contacted the droplet and subsequently retracting the stage. The droplets were then digitized for the θ measurements (KSV 2001). Each run consisted of generating a single droplet onto the modified substrate, followed by measurement of the θ as a function of pressure. A scan at increasing and decreasing pressures consisted of one experimental run. The θ values reported here are averages (at each pressure) of at least three such runs. The largest variation in θ (hysteresis) upon increasing and decreasing pressure did not exceed 4 deg. These results indicate that the monolayer on the substrate is compact.55 The setup allows for the measurement of tension and contact angle in tandem during the same measurement. For the measurement of the θ in the presence of surfactant, water with a known amount of surfactant was injected into the capillary, instead of pure water. This procedure was used instead of adding surfactant to bulk CO2 since in this case we were interested in investigating the effect of an amphiphile with high water solubility (low solubility in CO2). (53) Lemmon, E. W.; Huber, M. L.; McLinden, M. O. NIST Standard Reference Database 23: Reference Fluid Thermodynamic and Transport PropertiesREFPROP, Version 8.0, National Institute of Standards and Technology, Standard Reference Data Program, Gaithersburg, 2007. (54) Hebach, A.; Olbrich, A.; Dahmen, N.; Kogel, A.; Ederer, H.; Dinjus, E. J. Chem. Eng. Data 2002, 47, 1540-1546. (55) Clear, S. C.; Nealey, P. F. J. Colloid Interface Sci. 1999, 213, 238-250.
12074 Langmuir, Vol. 23, No. 24, 2007
Figure 1. Schematic diagram of a tandem high-pressure pendant drop tensiometer and goniometer.
Bharatwaj et al.
Figure 3. Interfacial tension of the CO2-water interface (γC|W) vs density of CO2 (F) at 298, 308, and 318 K. Error bars represent the deviation from three different runs. For those measurements, error bars are approximately the size of the symbols. steel capillary. The emulsion was constantly stirred and recirculated for 20 min. After this, the recirculation was stopped, and the emulsion stability was visually monitored through the front window of the pressure cell.
Results and Discussion
Figure 2. Schematic diagram of the experimental setup for emulsion studies. Substrate Preparation and Modification. The glass microscope slides were first degreased under ultrasonic stirring in chloroform for 10 min and then placed in a freshly prepared piranha solution (a 7:3 v/v mixture of 95.8% H2SO4 and 30% H2O2) at 373 K for 20 min. This step serves to remove organic contaminants and produce a glass surface with high concentration of hydroxyl groups, which act as attachment points for the trichlorosilane molecules. Subsequently, the substrates were rinsed with deionized water and blowndried in a light stream of N2. C8TS was deposited on the pretreated substrates by incubating the slides in a 2.5 mM hexane solution for 3 h.55,56 The quality of the deposited monolayer was tested by contact angle (in air). Emulsion Formation and Stability. Figure 2 is a schematic representation of the experimental apparatus employed in the emulsion studies. The setup is comprised of a variable-volume view cell and a high-pressure reciprocating pump (Milton Roy miniPump, 46-460 mL‚h-1). The view cell was fitted with a six-port switching valve (Valco), a temperature controller (Cole-Parmer EW-8900010), and a pressure transducer (Sensotec FP2000). In a typical experiment, a known mass of deionized water was loaded into the view cell along with the desired amount of surfactant. A known mass of CO2 was then added to the front part of the cell. The system was subsequently pressurized and equilibrated at the desired temperature. A magnetic stir bar was used to help the system reach equilibration. The water-CO2-surfactant mixture was then recirculated and sheared through a 127 µm i.d. × 50 mm long stainless (56) Fadeev, A. Y.; Mccarthy, T. J. Langmuir 2000, 16, 7268-7274.
CO2-Water Interface. The tension of the bare CO2-water interface was determined at 298, 308, and 318 K and densities up to 0.92 g‚mL-1. These results are complementary to and extend the range of previously reported tension values.19,54,57,58 The isotherms are plotted in Figure 3 as a function of the density of pure CO2 (F). The values were measured for a duration of 15-20 min at each condition, which was enough time for droplet equilibration, as also seen in previous works.54,57 To eliminate any possible effect due to droplet aging,54,57 a new droplet (of similar volume) was generated for each measurement (at each density). It can be seen that all isotherms fall into approximately the same curve when plotted as a function of F. This behavior can be explained by carefully analyzing the effect of temperature and density on the interaction between water-CO2, CO2-CO2, and water-water and with the help of eq 1:59
γcw ) γc + γw - 2 xγcdγwd - 2 xγcpγwp
(1)
Here γC and γW are the surface tension of CO2 and water and the superscripts d and p represent the dispersive and polar components of the surface tension, respectively. While CO2 has a zero dipole moment, it has a considerable quadrupole.60 An increase in temperature results in a decrease in both the dispersive and the dipole-quadrupole interaction between water and CO2, which have been estimated to be approximately of the same magnitude.61 From the -2(γcdγwd)1/2 and -2(γcpγwp)1/2 terms of eq 1 that represent the CO2-water interactions, it can be seen that an increase in temperature will favor an increase in γC|W.62 However, there is a corresponding decrease in γW with temperature, of about 4 mN‚m-1 within the range shown in Figure (57) Tewes, F.; Boury, F. J. Phys. Chem. B 2004, 108, 2405-2512. (58) Chun, B.-S.; Wilkinson, G. T. Ind. Eng. Chem. Res. 1995, 34, 43714377. (59) Good, R. J.; Girifalco, L. A. J. Phys. Chem. 1960, 64, 561-565. (60) Raveendran, P.; Ikushima, Y.; Wallen, S. Acc. Chem. Res. 2005, 38, 478-485. (61) Tassaing, T.; Oparin, R.; Danten, Y.; Besnard, M. J. Supercrit. Fluids 2005, 33, 85-92. (62) da Rocha, S. R. P.; Johnston, K. P.; Westacott, R. E.; Rossky, P. J. J. Phys. Chem. B 2001, 105, 12092-12104.
Surfactants for the CO2-Water Interface
Langmuir, Vol. 23, No. 24, 2007 12075
Figure 4. Interfacial tension of the CO2-water interface (γC|W) and solubility parameter of CO2 (δ) as a function of pressure at 318 K.
1,63 which can be attributed to a decrease in the dipole-dipole interactions between water molecules. The same can be expected for CO2-CO2 interactions; however, this is a secondary effect, given the much lower γC.62 These competing effects balance out, at least within the range investigated, causing the isotherms to more or less collapse into the same curve. A greater reduction in γC|W with density is observed at lower CO2 densities, where its compressibility is high. The γC|W levels off at intermediate pressures and then starts decreasingsbut with less pronounced changesswhen the density of CO2 is liquidlike. This behavior can be explained in terms of the CO2 solvent power, which in turn can be correlated to its solubility parameter (δ).64 A plot of γC |W and δ as a function of pressure at 318 K is shown in Figure 4. The solubility parameter was calculated using the following65 equation:
δ)
(
) (
)
uig - u hig - RT - h + PV ) V V
(2)
Here δ is in (J‚m-3)1/2, hig is the ideal gas enthalpy, h is the enthalpy of CO2, R is the ideal gas constant (J‚mol-1‚K-1), and V is the molar volume of CO2 (m3‚mol-1). The values of the h and V were sourced from REFPROP.53 Figure 4 clearly indicates that, for a given temperature, the solvent power of CO2 governs the tension of the C|W interface. It is also worth noticing that, contrary to previous reports, no cusp in γC|W exists within the pressure range investigated. The same is true for the isotherms from Figure 3 (not shown as a function of pressure). This is in agreement with more recent tensiometric results for the C|W interface.54 The results shown here also corroborate and are in excellent agreement with molecular dynamics (MD) simulations of the C|W interface,62 validating the potential models and methodology used in that work. Surfactant-Modified CO2-Water Interface. The activity of several amphiphiles with nonfluorinated tails, including vinyl acetate,66 ether carbonate,43 and methylated branched hydrocarbons,27 have been previously reported at the C|W interface. However, surfactants containing the biodegradable and biocompatible lactic acid as the CO2-phile have not yet been explored. Ab initio results indicate that the interactions between CO2 and (63) Vargaftik, N. B.; Volkov, B. N.; Volyak, L. D. J. Phys. Chem. Ref. Data 1983, 12, 817-820. (64) Allada, S. R. Ind. Eng. Chem. Process Des. DeV. 1984, 23, 344-348. (65) Johnston, K. P.; Peck, D. G.; Kim, S. Ind. Chem. Eng. Res. 1989, 28, 1115-1125. (66) Tan, B.; Cooper, A. I. J. Am. Chem. Soc. 2005, 127, 8938-8939.
Figure 5. Interfacial tension (γ) of the LAmEOnLAm-modified CO2water interface as a function of the hydrophilic-CO2-philic balance (HCB). In this case, the HCB is represented as the % EO in the molecule. Conditions are 0.01 wt % surfactant, 298 K, and 0.85 g‚mL-1 CO2 density.
acetate-based (vinyl and sugar acetate) moieties are quite strong, with energies half of that of the hydrogen-bonded water dimer.44,67 Spectroscopic evidence confirms such findings.41 Lactic acid (LA) is similar to the above-mentioned acetate-based moieties in that it contains carbonyl (ester) groups. Even though the solubility of poly(lactic acid) (PLA) and poly(glycolic acid) (PGA) and their copolymers is relatively small in CO2,68 CO2 is highly soluble in PLGA-based polymers.69,70 An increase in the LA/ GA (glycolic acid) ratio in PLGA copolymers has also shown to increase the swelling of the copolymer, indicating the role of the LA in enhancing the interaction between polymer and CO2.71 As discussed above, the CO2-philicity of polymers can be qualitatively estimated from their surface tension (σ). For example, PDMS (13 kg‚mol-1), a highly CO2-philic polymer, has a surface tension of about 20 mN‚m-1 at 293 K.39 The surface tension of 137 K g‚mol-1 PLA at 295 K is reported to be approximately 43 mN‚m-1.72 This value is comparable to the surface tension of other CO2-philic polymers such as poly(vinyl acetate) (PVA) (36.5 mN‚m-1 at 293 K, MW not provided) and poly(propylene oxide) (PO) (31.5 mN‚m-1 for 2 K g‚mol-1 at 293 K).39 These results suggest that LA is a viable CO2-phile. Moreover, because the MW of the moiety can greatly influence its surface tension (even at high MW),39 it can be used to fine-tune the interaction with CO2. Further enhancement of the interfacial activity might be achieved by careful control of the hydrophilic-CO2-philic balance (HCB). The results for the activity of the LA-based surfactants at the CO2-water interface as a function of the HCB are shown in Figure 5. The γC|W for three surfactants series with varying lengths of the hydrophile (EO7, EO14, EO23) and HCB are shown. The results were obtained at 0.01 wt % surfactant concentration, 298 K, and constant CO2 density of 0.85 g‚mL-1. It is interesting to note that both the LA and EO homopolymers are somewhat active at the C|W interface. The γC|W is reduced from about 28 mN‚m-1, down to approximately 12 mN‚m-1 for both LA28 and (67) Baradie, B.; Shoichet, M. S.; Shen, Z.; McHugh, M. A.; Hong, L.; Wang, Y.; Johnson, K. J.; Beckman, E.; Enick, R. Macromolecules 2004, 37, 77997807. (68) Shen, Z.; McHugh, M. A.; Xu, J.; Belardi, J.; Kilic, S.; Mesiano, A.; Bane, S.; Karnikas, C.; Beckman, E.; Enick, R. Polymer 2003, 44, 1491-1498. (69) Bodmeier, R.; Wang, H.; Dixon, D. J.; Mawson, S.; Johnston, K. P. Pharm. Res. 1995, 12, 1211-1217. (70) Sparacio, D.; Beckman, E. J. ACS Symp. Ser. 1998, No. 713, 181-193. (71) Liu, D.; Tomasko, D. J. Supercrit. Fluids 2007, 39, 416-425. (72) Gajria, A. M.; Dave, V.; Gross, R. A.; McCarthy, S. A. Polymer 1996, 37, 437-444.
12076 Langmuir, Vol. 23, No. 24, 2007
EO23. This is an indication of the favorable interactions of both water and CO2 with the ether19 and ester73 moieties. The value for EO14 is in reasonable agreement with the previously reported tension reduction of about 20 mN‚m-1 for EO14 at the C|W interface at 318 K and 0.88 g‚mL-1.19 The measured γ of the surfactant-modified C|W interface may be explained in part by the solubility of the amphiphiles in CO2 and in water. At the phase inversion point, where the surfactant is said to be balanced with respect to its partitioning (solubility) to both phases, the tension is the lowest since the surfactant will prefer to reside at the interface.74 As the HCB changes from the optimum balance, the surfactant will preferentially partition to one of the phase, resulting in a decrease in interfacial activity or an increase in interfacial tension. This is the origin of the V-shaped plots typically seen for tension vs HCB.19 Such behavior is also observed for the LAnEOmLAn series investigated here, as shown in Figure 5. Pure LA partitions more preferentially to the CO2 phase. As the % or the hydrophile increases, the tension decreases, goes through a minimum at the balance point (surfactant prefers the interface the most), and then starts increasing again. The balanced state is found to be at approximately 50% EO for all the three series. This result is somewhat expected since the homopolymers have similar interfacial activities. It is also clear from Figure 5 that the overall number of repeat units in the hydrophile dramatically impacts the activity of the surfactant at the C|W interface. The LAmEO7LAm series is observed to be the least interfacially active, with maximum tension reduction of about 18 mN‚m-1 (from the ∼28 mN‚m-1 for the bare C|W interface). This can be attributed to the relatively short size of the EO group and, consequently, the small number of LA repeat units (short LA groups) that are required to sweep the HCB space. Due to its small size, the molecules are less effective in separating the unfavorable contacts between water and CO2 since relatively fewer CO2-LA and water-EO contacts can be made, resulting in a higher tension.40 A significant reduction in tension compared to the LAmEO7LAm series is achieved by increasing the size of the EO group (n ) 14) and consequently also the size of the LA moiety across the HCB range. Tension values as low as 2.5 mN‚m-1 were observed for surfactants of the LAmEO14LAm series. However, further increase in the size of the molecule (to LAmEO23LAm) has little effect on the tension. While microscopic information on the interfacial structure of this system is not available (e.g., absorbed amount and conformation of the amphiphile), a couple of arguments can be offered to justify the observed trend. It is likely that an increase in the size of the triblock from LAmEO14LAm to LAmEO23LAm will be accompanied by an increase in area occupied by surfactant molecule at the interface.22 Therefore, the effective ability of the copolymer to screen CO2water interactions might not change significantly. Even if the lowering on the surface coverage ends up being not so significant, the interaction between LA segments (intra- and intermolecular) would have to be small compared to that of the CO2-LA interaction so that the LA moiety could be extended into bulk CO2 and, thus, reduce the interfacial tension. The relatively low solubility of LA compared to other hydrogenated CO2-philes68 indicates that the solvent-solute interactions might not be as favorable as solute-solute interactions, and thus, tail extension might be somewhat limited. The interfacial activity of some commercially available hydrocarbon-based surfactants was also determined to assess (73) Nelson, M. R.; Borkman, R. F. J. Phys. Chem. A 1998, 102, 7860-7863. (74) Aveyard, R.; Binks, B. P.; Clark, S.; Fletcher, P. D. I. J. Chem. Technol. Biotechnol. 1990, 48, 161-171.
Bharatwaj et al.
Figure 6. Comparison of the interfacial tension values of the LAmEO14LAm, CmEOn, and EOnPO31EOn surfactant series. Conditions are 0.01 wt % surfactant concentration, 298 K, and 0.85 g‚mL-1 CO2 density.
the relative CO2-philicity of the LA moiety. The surfactants chosen for this purpose were those from the EOmPO31EOm copolymer series and two branched hydrocarbons (TMN, b-CH2). TMN surfactants were chosen because they have been shown to form microemulsions of water-in-CO2.27 The γ of the aforementioned surfactants as a function of % EO was measured at 298 K and 0.85 g‚mL-1 CO2 density. The results, along with those for the LAmEO14LAm series, are shown in Figure 6. The interfacial activity of the LAnEO14LAn series is seen to be greater than that of the b-CH2 and PO series. While care should be exercised when comparing such different surfactant structures, some interesting parallels can be safely drawn. The results are quite surprising considering that PO is significantly more soluble in CO2 than LA.68 We can compare some of our results for the EOmPO31EOm series with the literature.19 To do so, however, we need to define the surface pressure Π ) γ0 γ, where γ0 is the tension of the bare C|W interface and γ is that of the surfactant-modified interface. This is necessary since measurements were done at different conditions (temperature and pressure). At 318 K and 0.88 g‚mL-1 CO2 density the EOmPO31EOm surfactant with 10% EO has a Π of ∼17 mN‚m-1. This is in excellent agreement with the tension reduction observed in this work at 298 K and 0.85 g‚mL-1 CO2 density. The lower tension values reported in the literature for this amphiphile are due to a lower binary C|W tension observed in that work.27 It is also interesting to note that similar Π’s were observed in spite of the fact that the surfactant concentration in this study is much lowers0.01 compared to 0.1 wt %.19 A similar trend is observed for the EOmPO31EOm surfactant with 20% EO. Another interesting observation from the results shown in Figure 6 is that the tension for the LA-based surfactants seen here is lower than that for a surfactant shown to form reverse aqueous aggregates in CO25b-C12EOx.27 This is very encouraging, suggesting that indeed the LA-based moiety is a good CO2-phile and a good balance can be reached with EO as the hydrophile. However, despite these advantages, the relatively low solubility of the LA surfactants cannot be overlooked. TMN surfactants were shown to form aggregates at a concentration of ∼1 wt %, which is well above the solubility limit of some of the LA-based surfactants. For instance, the solubility limit for the LA-based surfactant with a molecular weight of 1176 g‚mol-1 was found to be 0.2 wt %. Emulsion Formation and Stability. Emulsification is achieved by mechanical deformation of an interface so that droplets of one phase can be dispersed into a second phase. Emulsions with droplets in the micrometer range are turbid and thermodynamically unstable but may be kinetically stable.62 Surfactants are
Surfactants for the CO2-Water Interface
Figure 7. Schematic representation of the emulsification occurring in the presence of the triblock LA4EO14LA4 surfactant depicting the gradual upward motion of the CO2 droplets as a function of time and their eventual coalescence.
used to reduce the energy required to deform the interface and are necessary to stabilize the droplets against flocculation and coalescence. There is a strong relationship among the morphology of emulsions, the surfactant phase behavior, and the γ. For equal amounts of oil (in our case CO2) and water and above the critical micellar concentration (for systems forming aggregates), the continuous phase of the emulsion will be the one in which the surfactant is most soluble. This is called Bancroft’s rule. Emulsion morphology and stability can be characterized on the basis of visual observations. A system where a creaming front is formed (rising CO2 droplets) followed by the appearance of an upper clear phase (coalescence of CO2 droplets) is identified as a CO2in-water emulsion. The reverse is true for a water-in-CO2 emulsion.22 Emulsion stability can be determined by following the rate of creaming and coalescence. Here, we investigated the ability of LA4EO14LA4 in forming and stabilizing emulsions of CO2 and water. This surfactant was chosen as it has an optimum balance and, thus, the largest tension lowering compared to the others in the series. Emulsion formation and stability of a system composed of equal masses of CO2 and water, in the presence of 1 wt % of LA4EO14LA4 at 298 K and 13.6 MPa, is reported. The system was sheared for 20 min at 9 mL‚min-1 flow rate, resulting in complete emulsification of both phases. The stability of the emulsion, immediately after recirculation and stirring were stopped, is schematically illustrated in Figure 7. The different shades of the phases represent the degree of emulsification at different times. A milky white emulsion phase is represented in dark/black, a clearing phase is represented in gray, and a single phase (no emulsion) is represented in white/transparent. In this study, a creaming front could be observed 13 min after recirculation and stirring stopped. After 20 min, the lower line of the creaming front was above half of the variable volume cell leaving behind a transparent (aqueous) phase. At this point, a clear upper phase could be also identified. The results indicate the formation of a CO2-in-water emulsion. The emulsion stability with LA4EO14LA4 follows a similar trend to those formed with EOmPO31EOm surfactants.62 A similar comparison can be drawn between the emulsions of the surfactants of the R series and the LA-based surfactant used in this work. For instance, the surfactant 17R2 was found to reduce the γ to 5.1 mN‚m-1 at 318 K.19 However, despite the significant reduction in the interfacial tension, they could only form weak and unstable water-in-CO2 emulsions at 295 K and 1.6 wt % surfactant. This was attributed to the fact that the surfactant was on the CO2-phobic branch of the log γ vs formulation variable plot. This was against the natural curvature
Langmuir, Vol. 23, No. 24, 2007 12077
Figure 8. Contact angle (θ) of pure water (C|W|C8TS) and an aqueous solution containing 0.1 wt % LA1EO14LA1 (LA1EO14LA1|C|W|C8TS) on a C8TS-modified substrate under CO2 atmosphere. The results were determined at 308 K and are shown as a function of the density of pure CO2 (F). The inset is a schematic diagram of the experimental three-phase contact line.
of the surfactant, thereby resulting in the formation of a weak emulsion. High-Pressure Contact Angle Goniometry. To elucidate the effect of the LA-based amphiphiles on the wetting behavior of water on a hydrophobic (C8TS-modified glass substrate) surface in compressed CO2, we determined the contact angle (θ) of a 0.1 wt % aqueous solution of LA1EO14LA1 under pressure. The results were obtained at 308 K as a function of the CO2 density and are shown in Figure 8 along with the θ results for pure water under the same conditions. Prior to discussion of the variation of θ as a function of CO2 density and the effect of the surfactant, a brief outline about the variation in the water contact angle on hydrophobic surfaces against air is necessary. The contact angle of pure water (against air) on the hydrophobic substrate discussed here was found to be 103°. This value is in excellent agreement to literature values and represents a closed-packed monolayer.75 From Young’s equation (eq 3), a negative value for cos θ is thus found, implying that the surface energy of the air-solid interface (γsv) is lower than that of the liquid-solid interface (γsw)sγlv in that equation is the surface tension of water.
γlv cos θ ) γsv - γsw
(3)
While γsw cannot be directly determined, it has been evaluated for a similar system as 29.2 mN‚m-1.1 The γsv is 22 mN‚m-1,76 which is close to the surface tension of octane (21.62 mN‚m-1).77 As CO2 is charged into the high-pressure cell, the θ is seen to increase, and it continues increasing as the CO2 density goes up. At a density close to that for the saturation pressure, a change (decrease) in slope is observed. It can be attributed to smaller variations in solvent power. This dewetting behavior of pure water on a solid interface under compressed atmosphere is similar to that recently reported in the literature.1 Since the θ of acidic water on the C8TS-modified substrate against air and the surface tension of acidic water remain approximately constant within the conditions of pH expected in this study,1 we attribute this dewetting behavior to a reduction in γcw with increasing density. Note that eq 3, when under CO2 atmosphere, should be written (75) Ulman, A. An Introduction to Ultrathin Organic Films: From Langmuir Blodgett to Self-Assembly; Academic Press: San Diego, CA, 1997. (76) Clear, S. C.; Nealey, P. F. J. Colloid Interface Sci. 1999, 213, 238-250. (77) Hiemenz, P.; Rajagopalan, R. Principles of Colloid and Surface Chemistry; Marcel Dekker: New York, 1997.
12078 Langmuir, Vol. 23, No. 24, 2007
as γcw cos θ ) γsc - γsw, where γsc is the surface energy between the C8TS-modified substrate and CO2. While γsw remains approximately constant within the conditions shown here, γsc is expected to decrease as the density of CO2 increases due to an increase in the solvent power of CO2.1 Therefore, the reduction in the interfacial energy at the solid-CO2 interface, coupled with the reduction in the tension of the CO2-water interface with increasing density, leads to an increase in the negativity of the cos θ value and, hence, the increase in the contact angle. Upon addition of the surfactant, the contact angle of this solution in air reduces dramatically from 103 to 88°. This change can be attributed to a reduction in the γsw to a value lower than that of the γsc. With the introduction of CO2, the contact angle of water at the three phase contact line is seen to again go above 90°, and it increases with CO2 density. This increase in θ is caused by a decrease in the γsc. The θ is then seen to increase with density due to an increase in solvent power of CO2 that leads to both an enhanced interaction with the substrate (reduction in γsc) and a decrease in γcw, as shown above. A change of slope (variation of θ with density) is also observed when in the presence of surfactant. Different from the pure water case, however, the θ value seems to level off more quickly (levels off at lower density compared to the system without surfactant) when in the presence of the amphiphile. This can be attributed to the fact that the reduction in γcw is more significantly affected by the presence of surfactant at higher pressuressmore so than the effect of CO2 density.
Conclusions In this work we investigated the interfacial properties of biodegradable and biocompatible nonionic lactide (LA)-based amphiphiles at the CO2-water (C|W) and CO2-water-solid (C|W|S) interfaces. A series of amphiphiles with the general structure LAmEOnLAm were synthesized for this study. A novel high-pressure tensiometer/contact angle goniometer that allows us to measure both interfacial tension (γ) and contact angle (θ) in tandem was also developed. Before going into the surfactant-modified interfaces, we investigated the γ of the binary C|W interface at 298, 308, and 318 K and CO2 densities of up to ∼0.92 g‚mL-1 CO2. The different isotherms are seen to approximately collapse onto a single curve when plotted as a function of CO2 density. These results can be explained in terms of the effect of temperature on the CO2-CO2, CO2-water, and water-water interactions. The tension-pressure isotherms were shown to follow the same trend as the CO2 solubility parameter isotherms, thus suggesting that the solvent power of CO2 alone controls the γ of the binary interface. Different from some previously published results,19,58 but in agreement with more recent tensiometric data,54 we did not observe any cusps on the tension vs pressure diagrams. The experimental
Bharatwaj et al.
results shown here are in excellent agreement with (thus corroborating) previous computational studies of the C|W interface.40 The LAmEOnLAm surfactants were shown to be highly active at the C|W interface even at low (0.01 wt %) overall surfactant concentrations. The γ of the binary C|W interface was reduced to as low as 2.5 mN‚m-1. The γ is significantly impacted as the size of the hydrophile increases from EO7 to EO14, but little change is seen when further increase to EO23. The results are discussed in terms of the ability of surfactants with different molecular weights and interfacial structures to screen CO2water interactions. A scan in the hydrophilic-to-CO2-philic balance reveals an optimum surfactant structure (minimum tension) at 50% EO for all series. When contrasted to other nonfluorinated/ non-silicon-based CO2-philes, the ester moiety is seen to be fairly promising. It can reduce the γ of the C|W interface to a larger extent when compared to propylene oxide (PO)-based copolymers, which were shown to produce very stable emulsions of water and CO2.19 At the same wt % concentration, they are seen to be more interfacially active than the microemulsion-forming, branched hydrocarbons.27 However, the overall solubility of the amphiphiles discussed here is limited compared to some of those surfactants. The surfactant LA4EO14LA4, which showed an optimum balance at the C|W interface for the EO14 series, is capable of emulsifying equal volumes of CO2 and water at 298 K, 13.6 MPa, and 1 wt % overall concentration. The resulting emulsion morphology is water-continuous (CO2 dispersed droplets in water), and stabilities similar to those observed for PO-based amphiphiles78 were determined by visual observation of the contents of the high-pressure cell. In situ contact angle measurements of water against a silane-modified substrate under compressed CO2 atmosphere show that the aqueous phase dewets the substrate as the pressure is increased. This phenomenon is similar to recently reported results.1 We attribute the dewetting of the aqueous phase to a decrease in γ of the C|W interface with increasing pressure since the γ of the water-solid substrate remains approximately constant and that of the C8TS-modified substrate-CO2 is expected to decrease with density (see force balance on the inset in Figure 8). The presence of LA1EO14LA1 improves the wetting on the hydrophobic substrate. The same dewetting behavior as a function of CO2 density is also observed when in the presence of the surfactant. Acknowledgment. We thank WSU and the EPA (Grant No. R831504) for financial support and BASF for providing us with the PO-based surfactants. LA701831V (78) da Rocha, S. R. P.; Psathas, P. A.; Klein, E.; Johnston, K. P. J. Colloid Interface Sci. 2001, 239, 241-253.
