2412
Langmuir 2008, 24, 2412-2420
Synthesis, Characterization, Adsorption, and Interfacial Rheological Properties of Four-Arm Anionic Fluorosurfactants Ayc¸ a Ertekin,† Charles M. Kausch,‡ Yongsin Kim,‡ and Richard R. Thomas*,‡ Department of Polymer Engineering, Polymer Engineering Academic Center, The UniVersity of Akron, Akron, Ohio 44325-0301, and OMNOVA Solutions, Inc., 2990 Gilchrist Road, Akron, Ohio 44305-4418 ReceiVed October 8, 2007. In Final Form: NoVember 9, 2007 Four-arm oligo(fluorooxetane) tetraols containing -CF3 and -C2F5 groups were prepared in reasonable yields by cationic, ring-opening polymerization of fluorinated oxetane monomers using a tetrafunctional, alkoxylated polyol as initiator and BF3‚THF as catalyst. The tetraols were then converted to ammonium sulfate salts using oleum followed by neutralization with ammonium hydroxide in excellent yields. The four-arm oligo(fluorooxetane) sulfates (1 ) -CF3, 2 ) -C2F5) have an architecture characterized by a hydrophobic core of oligo(fluorooxetane) arms with a hydrophilic sulfate shell and initiator. The four-arm anionic oligo(fluorooxetane)s are surface active with critical micelle concentration values ≈4.2 × 10-6 and 2.4 × 10-6 mol/L for 1 and 2, respectively. Surface tension isotherms in pH 8 buffered solution were measured and data fitted parametrically to the Davies surface tension isotherm equation. Molecular areas at saturation were estimated to be ∼89 and ∼85 Å2 with ∆Gads ) -12.7 and -13.2 kcal/mol for 1 and 2, respectively. The results are compared to two-arm, bolaamphiphilic analogues of 1 and 2 and a small molecule, long perfluoroalkyl-chain (-C8F17), anionic fluorosurfactant (Kausch, C. M.; Kim, Y.; Russell, V. M.; Medsker, R. E.; Thomas, R. R. Langmuir 2003, 19, 7182). Dynamic surface tension data for 1 and 2 were analyzed using the Ward-Tordai mass transport equation to yield concentration-dependent diffusion coefficients. In the concentration range ≈10-6 mol/L, diffusion coefficients were estimated to be ≈1 - 3 × 10-5 cm2/s. Dilational interfacial rheological parameters for 1 and 2 were measured. Values of |E| and E′ were found to be larger than those of the two-arm analogues of the same perfluoroalkyl chain length while E′′and φ were found to be smaller. The magnitude of these values reflects the difference in adsorption strength and mass transport and/or relaxation between the two different architectures.
Introduction The simplest class of surfactants is characterized by the presence of single hydrophilic and hydrophobic functional groups on a molecule. The next levels of architectural complexity are either the gemini- or bolaamphiphilic-type surfactants. Beyond this level of diversity, few studies have been conducted and little is known regarding more complicated structural types and effect on interfacial behavior. This would include those with either dendrimeric/hyperbranched architecture1 or surfactants with the number of hydrophilic and/or hydrophobic groups exceeding two.2 Several reports have appeared detailing the synthesis and some adsorption properties (particularly π-A isotherms) of hyperbranched/dendrimeric amphiphiles; however, a comprehensive examination of surface tension isotherm and interfacial rheological parameters was lacking.3-5 An example of a coreshell suprabranched amphiphile using an oxetane monomer and Lewis acid catalysis is given by Xu et al.6 In addition, several interesting polyelectrolytes having dendrimeric architectures with 8-327 and 368,9 carboxylic acid groups have been described; however, no details of adsorption behavior were reported. * To whom correspondence should be addressed. E-mail: richard.thomas@ omnova.com. † The University of Akron. ‡ OMNOVA Solutions, Inc. (1) Newkome, G. R.; Moorefield, C. N.; Vo¨gtle, F., Dendrimers and Dendrons: Concepts, Syntheses, Applications; Wiley: New York, 2002. (2) Mu¨ller, P. U.; Akpo, C. C.; Sto¨ckelhuber, K. W.; Weber, E. AdV. Colloid Interface Sci. 2005, 114-114, 291. (3) Sui, G.; Micic, M.; Huo, Q.; Leblanc, R. M. Colloids Surf., A: Physico. Eng. Asp. 2000, 171, 185. (4) Ariga, K.; Urakawa, T.; Michiue, A.; Sasaki, Y.; Kikuchi, J. i. Langmuir 2000, 16, 9147. (5) Istratov, V.; Kautz, H.; Kim, Y.-K.; Schubert, R.; Frey, H. Tetrahedron 2003, 59, 4017. (6) Xu, Y.; Gao, C.; Kong, H.; Yan, D.; Luo, P.; Li, W.; Mai, Y. Macromolecules 2004, 37, 6264.
The benefits of greater architectural dimensionality on adsorption properties of bolaamphiphilic10-12 and gemini surfactants12-20 have been the subject of several investigations. The lack of study on these more complicated structural types is true particularly for fluorosurfactants. Previous work from this laboratory has detailed the synthesis, characterization, adsorption, and interfacial rheological properties of a series of oligomeric, bolaamphiphilic-type (two-arm) fluorosurfactants based on oligo(fluorooxetanes).21,22 The perfluoroalkyl (Rf) group of the oligo(fluorooxetane) was limited to -CF3 and -C2F5. In summary, previous work demonstrated that oligomers containing either -CF3 or -C2F5 groups were capable of achieving air/water surface tensions in the 30-25 mN/m range. (7) van Hest, J. C. M.; Baars, M. W. P. L.; Elissen-Roma´n, C.; van Genderen, M. H. P.; Meijer, E. W. Macromolecules 1995, 28, 6689. (8) Newkome, G. R.; Moorefield, C. N.; Baker, G. R.; Johnson, A. L.; Behera, R. K. Angew. Chem., Int. Ed. Engl. 1991, 30, 1176. (9) Newkome, G. R.; Moorefield, C. N.; Baker, G. R.; Saunders, M. J.; Grossman, S. H. Angew. Chem., Int. Ed. Engl. 1991, 30, 1178. (10) Jong, L. I.; Abbott, N. L. Langmuir 1998, 14, 2235. (11) Muzzalupo, R.; Ranieri, G. A.; La Mesa, C. Langmuir 1996, 12, 3157. (12) Pestman, J. M.; Terpstra, K. R.; Stuart, M. C. A.; van Doren, H. A.; Brisson, A.; Kellogg, R. M.; Engberts, J. B. F. N. Langmuir 1997, 13, 6857. (13) Diamant, H.; Andelman, D. Langmuir 1994, 10, 2910. (14) Diamant, H.; Andelman, D. Langmuir 1995, 11, 3605. (15) Eastoe, J.; Rogueda, P.; Howe, A. M.; Pitt, A. R.; Heenan, R. K. Langmuir 1996, 12, 2701. (16) Alami, E.; Abrahmse´n-Alami, S.; J., E.; Heenan, R. K. Langmuir 2002, 19, 18. (17) Menger, F. M.; Keiper, J. S.; Azov, V. Langmuir 2000, 16, 2062. (18) Song, L. D.; Rosen, M. Langmuir 1996, 12, 1149. (19) Wettig, S. D.; Nowak, P.; Verrall, R. E. Langmuir 2002, 18, 5354. (20) Boschkova, K.; Feiler, A.; Kronberg, B.; Stålgren, J. J. R. Langmuir 2002, 18, 7930. (21) Kausch, C. M.; Leising, J. E.; Medsker, R. E.; Russell, V. M.; Thomas, R. R.; Malik, A. A. Langmuir 2002, 18, 5933. (22) Kausch, C. M.; Kim, Y.; Russell, V. M.; Medsker, R. E.; Thomas, R. R. Langmuir 2003, 19, 7182.
10.1021/la7031175 CCC: $40.75 © 2008 American Chemical Society Published on Web 02/12/2008
Four-Arm Anionic Fluorosurfactants
While higher than the surface tensions exhibited by longer perfluoroalkyl-chain (eg., -C8F17) fluorosurfactants, the combination of flexible alkylene oxide units along with the entropic freedom to arrange effectively at an air/water interface afforded by an oligomer, allowed for surface tension reductions that were substantially lower than for small molecules possessing the same -CF3 or -C2F5 Rf groups. A trend of increasing air/water interfacial dilational viscoelastic (|E|) and elastic (E′) moduli was noted with increasing Rf chain length. Somewhat unexpectedly, the diffusion of the four-arm, anionic, oligo(fluorooxetanes) to the air/water interface was found to be comparable to that observed for typical, small molecule, -C8F17 fluorosurfactants and bolaamphiphilic analogues at a given concentration. The current study focuses on the next level of surfactant architectural complexity; namely, molecules containing mutliple hydrophiles/hydrophobes. The synthesis and characterization of four-arm, anionic oligo(fluorooxetanes) are discussed. In addition, surface tension isotherms were recorded from which adsorption parameters and kinetics (diffusion coefficient) were evaluated. Parameters obtaining by solving the Davies surface tension isotherm were compared to data obtained on films of a precursor four-arm tetraol cast on water and studied using a LangmuirBlodgett film balance. Finally, air/water dilatational interfacial rheological parameters for the four-arm, anionic, oligo(fluorooxetane)s are examined and discussed. Experimental Section Materials. Water used for synthesis was distilled. Water for Langmuir-Blodgett trough experiments was distilled and purified further using a Corning Mega-Pure MP-6A still to a typical resistivity >18 MΩ/cm. 3-Methyl-3-[(2,2,2-trifluoroethoxy)methyl]oxetane and 3-methyl-3-[(3,3,3,2,2-pentafluoropropoxy)methyl]oxetane monomers were prepared as described previously.21 The pH 8 solutions were prepared using solid pHydrion pH 8 buffer (Aldrich). Polyol 4640 tetrafunctional alkoxylated polyol initiator (hydroxyl number ) 630 ( 25; nominal molecular weight ) 360) was from Perstorp Polyols, Inc. (Toledo, OH). The boron trifluoride-tetrahydrofuran (BF3‚THF) complex was from Honeywell, Inc. 1,2-Dichloroethane (OmniSolve, LC grade) was from EMD. Unless stated otherwise, materials were used as received. Instrumentation. NMR spectroscopy was performed using a Varian Unity 400 spectrometer with probe frequencies of 399.945 and 100.575 MHz for 1H and 13C observation, respectively. Chemical shifts are reported against an internal standard of tetramethylsilane in CDCl3 solvent. Langmuir Film Balance Studies. π-A measurements on monolayers of the tetraol precursors 1a and 2a were studied using a Nima Type 611D Langmuir trough with an area ≈600 cm2. The trough area and tensiometer with filter paper probe were calibrated prior to use. Solutions of 1a and 2a were prepared in CHCl3 at ∼1 mg/mL (Aldrich Reagent Grade) and cast on water in ∼2 µL increments. Approximately 10-15 min were allowed for solvent evaporation before compression/expansion of the monolayer at 100 cm2/min. Equilibrium and Dynamic Surface Tension and Interfacial Dilational Rheological Measurements. Equilibrium and dynamic surface tensions and interfacial rheological measurements were performed using an oscillating bubble rheometer (The Tracker, ThetaDyne Corporation, Charlotte, NC). Details regarding this instrument have been given previously.23,24 The air-water interface was created by injecting a known volume of air into an inverted stainless steel needle attached to an airtight syringe. The tip of the needle was placed in a quartz cuvette containing the surfactant solution. Prior to each measurement, the cuvette and needle were (23) Hambardzumyan, A.; Aguie´-Be´ghin, V.; Panaı¨otov, I.; Douillard, R. Langmuir 2003, 19, 72. (24) Saulnier, P.; Boury, F.; Malzert, A.; Heurtault, B.; Ivanova, T.; Cagna, A.; Panaı¨otov, I.; Proust, J. E. Langmuir 2001, 17, 8104.
Langmuir, Vol. 24, No. 6, 2008 2413 cleaned by sonication followed by thorough rinsing in distilled water and acetone. The image of the drop formed at the needle tip was captured by a CCD camera, digitized and analyzed by software employing the Laplace equation to obtain the surface tension at the solution-air interface. For equilibrium and dynamic surface tension measurements, the surface tension was monitored as a function of time and the interface was assumed to be equilibrated when the surface tension did not change with time within experimental scatter over a period ≈1 h. Interfacial rheological measurements were performed by oscillating the bubble volume to a maximum change of 20% (∆A/A ≈ 0.13) of the original drop volume. Data for five oscillation cycles were collected and averaged. For both materials examined, 1 and 2, dilational interfacial rheological parameters were independent of strain (∆A/A over a range of 0.1-0.25) and, therefore, in the linear viscoelastic regime. Synthesis of Four-Arm Oligo(fluorooxetane) Tetraol Precursor 1a. A 1 L, three-necked, jacketed reaction flask equipped with a magnetic stirrer, 125 mL pressure-equalizing addition funnel, nitrogen inlet and outlet, temperature probe, and reflux condenser were allowed to equilibrate at 30 °C. The reactor was charged with 153 g of CH2Cl2, 39.2 g of tetrafunctional Polyol 4640 polyol (113 mmol), and 6.33 g of BF3‚THF (45.24 mmol). The reaction mixture was allowed to stir for 30 min. 3-Methyl-3-[(2,2,2-trifluoroethoxy)methyl]oxetane (250 g, 1.36 mol) was added over 90 min. The temperature reached 42 °C after a 30 min induction period. The reaction was allowed to stir for 10 h after which another portion of CH2Cl2 (232 g) was added. The CH2Cl2 solution was then washed with 5% sodium bicarbonate solution (152 g, 90 mmol), followed by 246 mL of water to remove the BF3‚THF. The solution was dried with magnesium sulfate, solvent was removed, and four-arm oligo(fluorooxetane) tetraol 1a (258.94 g) was isolated at 59% yield based on initiator. The number-average degree of polymerization was determined after esterification of the terminal hydroxyl groups with trifluoroacetic anhydride by 1H NMR and found to be 11.4 equating to Mn ≈ 4300 g/mol. 1H NMR (CDCl3) δ 0.85-0.99 (-CH3, 3 H), 3.15-3.69 (-CH2O-, 8 H), 3.73-3.89 (-OCH2CF3, 2 H); 13C NMR (CDCl3) δ 16.9-17.3 (-CH3), 40.9-41.4 (backbone -CH2O), 68.3-69.6 (-OC*H2CF3, q, J13C-19F ) 34.2 Hz), 73.4-73.9 (pendent -CH2O-), 124.1 (-CF3, q, J13C-19F ) 279 Hz). Synthesis of Four-Arm Oligo(fluorooxetane) Tetraol Precursor 2a. A 500 mL, three-necked reaction flask equipped with a magnetic stirrer, 125 mL pressure-equalizing addition funnel, nitrogen inlet and outlet, temperature probe, heating mantle, and reflux condenser were allowed to equilibrate at 30 °C. The reactor was charged with 57.9 g of CH2Cl2, 9.25 g of tetrafunctional Polyol 4640 polyol (27 mmol), and 1.49 g of BF3‚THF (10.67 mmol). The reaction mixture was allowed to stir for 30 min. 3-Methyl-3-[(3,3,3,2,2-pentafluoropropoxy)methyl]oxetane (100.0 g, 0.43 mmol) was added over 90 min. The temperature reached 42 °C after a 20 min induction period. The temperature reached a maximum of 47 °C. The reaction was allowed to stir for 10 h after which another portion of CH2Cl2 was added (87.40 g). The CH2Cl2 solution was then washed with 5% sodium bicarbonate solution (35.9 g, 0.02 mol sodium bicarbonate), followed by 93 mL of water to remove BF3‚THF. The solution was then dried with magnesium sulfate and solvent was removed to yield four-arm oligo(fluorooxetane) tetraol 2a (86 g, 57% yield based on initiator). The number-average degree of polymerization was determined after esterification of the terminal hydroxyl groups with trifluoroacetic anhydride by 1H NMR and found to be 16.2 equating to Mn ≈ 5600 g/mol. 1H NMR (CDCl3) δ 0.85-0.99 (-CH3, 2.9 H), 3.15-3.69 (-CH2O-, 8 H), 3.73-3.89 (-OCH2CF3, 2 H); 13C NMR (CDCl ) δ 16.9-17.3 (-CH ), 40.9-41.4 (backbone 3 3 -CH2O-), 68.3-69.6 (-OC*H2CF2-, t, J13C-19F ) 34.2 Hz), 73.473.9 (pendent -CH2O-), 113.1 (-CF3-, t of q, J13C-19F ) 255, 37 Hz), 118.8 (-CF2-, q of t, J13C-19F ) 286, 35 Hz). Synthesis of Four-Arm, Anionic Oligo(fluorooxetane) 1. A 1 L, three-necked, jacketed reaction flask was equipped with a condenser, magnetic stirrer, temperature probe, and addition funnel. The jacket temperature was set to -5 °C. The reactor was then charged with 200 g of 1a (103 mmol), and 200 g of tetrahydrofuran. The reaction was allowed to cool to -5 °C, and dropwise addition
2414 Langmuir, Vol. 24, No. 6, 2008 of 20% fuming sulfuric acid (120.7 g, 0.34 mol) was started. An exothermic reaction was observed. The addition rate was regulated to keep the maximum temperature under 15 °C. A maximum temperature of 12.5 °C was observed over the 2 h addition. A conversion of 90% was observed by 1H NMR. Additional portions of 20, 10.4, 14.7, and 15.5 g of 20% fuming sulfuric acid were added. A conversion of 98% or better of hydroxyl groups to sulfate groups was observed by 1H NMR. The acid polymer solution was collected and placed in an addition funnel connected to a 2 L reactor equipped with an ice bath, temperature probe, reflux condenser, and mechanical stirrer. The acid polymer solution was added to 314.5 g of concentrated ammonium hydroxide and 188.7 g of water while keeping the maximum temperature below 48 °C. The reaction mixture was placed in a separatory funnel and the bottom aqueous layer was removed. The organic layer was found to be 42.5 wt % solids. A 5 wt % sodium bicarbonate solution was added (69.3 g, 1.5 wt % based on polymer solids). The solvent was removed under reduced pressure while keeping the maximum temperature under 46 °C to yield 343 g of four-arm, anionic oligo(fluorooxetane) 1 (91% yield based on 1a). Using the degree of polymerization value measured for 1a, Mn ≈ 4300 g/mol. Synthesis of Four-Arm, Anionic Oligo(fluorooxetane) 2. A 1 L, three-necked, jacketed reaction flask was equipped with a condenser, magnetic stirrer, temperature probe, and addition funnel. The jacket temperature was set to -5 °C. The reactor was then charged with 200 g of 2a (36 mmol) and 200 g of tetrahydrofuran. The reaction was allowed to cool to -5 °C and dropwise addition of 20% fuming sulfuric acid (107.4 grams, 0.3 mol) was started. An exothermic reaction was observed. The addition rate was regulated to keep the maximum temperature under 15 °C. A maximum temperature of 12.5 °C was observed over the 2 h addition. A conversion of 90% was observed by 1H NMR. Additional portions of 20, 10.4, 14.7, and 15.5 g of 20% fuming sulfuric acid were added. A conversion of 98% or better of hydroxyl groups to sulfate groups was observed by 1H NMR. The acid polymer solution was collected and placed in an addition funnel connected to a 2 L reactor equipped with an ice bath, temperature probe, reflux condenser, and mechanical stirrer. The acid polymer solution was added to 181.9 g of concentrated ammonium hydroxide and 109 g of water while keeping the maximum temperature below 48 °C. The reaction mixture was placed in a separatory funnel and the bottom aqueous layer was removed. The organic layer was found to be 70.4 wt % solids. A 5 wt % sodium bicarbonate solution was added (69.3 g, 1.5 wt % based on polymer solids). The solvent was removed under reduced pressure while keeping the maximum temperature under 46 °C to yield 204.8 g of four-arm, anionic oligo(fluorooxetane) 2 (96% yield based on 2a). Using the degree of polymerization value measured for 2a, Mn ≈ 6000 g/mol.
Results and Discussion Synthesis and Characterization of Four-Arm, Anionic Oligo(fluorooxetane)s. The precursor tetraols to 1a and 2a were prepared by initiation of fluorooxetane monomer ring-opening polymerization using a tetrafunctional, hydroxyl-terminated poly(ethylene oxide)-co-poly(propylene oxide) oligomer. The alkylene oxide initiator was necessary to ensure water dispersibility as solubility decreases substantially as the degree of polymerization of the flourooxetane monomer increases. The synthesis of fluorooxetane monomers was described previously.21,22 The tetraol precursors were then converted to the ammonium sulfate esters using oleum followed by neutralization with ammonium hydroxide. The synthetic scheme for 1 and 2 is shown in Figure 1. The tetraol precursors 1a and 2a were prepared in reasonable yields of 59% and 57%, respectively, while the final conversion to the ammonium sulfate salts proceeded to 91% and 96%, respectively, with overall yields of 54% and 55%, respectively, for 1 and 2. The oligomeric nature of 1a and 2a makes them amenable to number-average degree of polymerization analysis by 1H NMR spectroscopic investigation. 1H and 13C NMR spectra
Ertekin et al.
Figure 1. Synthesis of four-arm, anionic poly(fluorooxetane)s 1 and 2. Rf ) -CF3 and -C2F5 for 1 and 2, respectively, and x ≈ 11.4 and 16.2 for 1 and 2, respectively.
are shown in panels A and B of Figure 2, respectively, for 2 along with structural identification (Figure 2D). There is substantial overlap in the key -CH2O- region at 3-4 ppm, including backbone, pendent chain, and terminal moieties, in the 1H NMR spectrum making quantification difficult. Terminal -CH2O- identification becomes possible upon esterification using trifluoroacetic anhydride (Figure 2C; only the terminal -CH2OC(dO)CF3 is shown) as this resonance is now shifted downfield to ∼4.28 ppm. The ratio of the terminal alkoxy-tointernal alkoxy signals can then be used to establish a numberaverage degree of polymerization that was determined to be 11.4 and 16.2 for 1a and 2a, respectively. The number-average values used for 1 and 2 are taken to be identical based on a previous study demonstrating, under current experimental conditions, that sulfation using oleum results neither in main-chain scission, polymer degradation, nor subsequent molecular weight reduction.22 Another benefit of trifluoroacetic anhydride derivatization is establishing the symmetricalness of oligo(fluorooxetane) polymerization from the tetrafunctional initiator. Theoretically, ringopening polymerization of the fluorooxetane monomer can occur either with near symmetry using all four possible hydroxyl groups present on the initiator or, at the other extreme, quite asymmetrically using one initiator hydroxyl group from which to grow the oligo(fluorooxetane). Of course, all polymer growth between the two extremes is possible. In an independent experiment, the initiator was subjected to derivatization with trifluoroacetic anhydride. The resulting 1H NMR spectrum exhibited a signal at ∼4.5 ppm attributable to the -CH2OC(dO)CF3 group of the esterified initiator. Figure 2C for 2 shows clearly the near absence of a resonance at ∼4.5 ppm indicating that trifluoroacetic anhydride esterification occurs at a terminal oxetane unit (Figure 2D) and not from the initiator and that oligo(fluorooxetane) growth transpires in a somewhat symmetrical fashion about the initiator core. In previous work, symmetrical fluorooxetane polymerization from a neopentyl group was noted as well.25 NMR data for 1a are nearly identical to those of 2a with the exception of differences in 13C NMR spectra in the (25) Wesdemiotis, C.; Pingitore, F.; Polce, M. J.; Russell, V. M.; Kim, Y.; Kausch, C. M.; Connors, T. H.; Medsker, R. E.; Thomas, R. R. Macromolecules 2006, 39, 8369.
Four-Arm Anionic Fluorosurfactants
Langmuir, Vol. 24, No. 6, 2008 2415
Figure 2. Proton (A) and 13C (B) NMR spectra for 2a in CDCl3. Peaks indicated by † and * indicate residual fluorooxetane monomer and CDCl3, respectively. Shown in (C) is the terminal -CH2OC(dO)CF3 region after esterification using trifluoroacetic anhydride. Structures of 2a and trifluoroacetate ester derivative, along with assignments, are shown in (D) and (E), respectively.
region attributable (∼120 ppm) to the different perfluoroalkyl groups. Dynamic and Equilibrium Surface Tension and Adsorption Parameters. The four-arm anionic oligo(fluorooxetane)s 1 and 2 are amphiphilic and, therefore, surface active. Air-water surface tension isotherms were recorded for each material in pH 8 buffer solutions and are shown in Figures 3 and 4, respectively, using an oscillating bubble rheometer. To characterize further the adsorption properties of 1 and 2, surface tension data were fitted parametrically to the Davies surface tension isotherm due to the ionic nature of materials examined. The Davies surface tension isotherm is a variation of the Langmuir isotherm and accounts for the effect of double layer charging on adsorption properties and is given in eq 126,27
γ ) γ0 + RTΓ∞ ln(1 - θ) -
[
( )]
(1)
Figure 3. Surface tension isotherm for the four-arm anionic (-CF3) oligo(fluorooxetane) 1 in pH 8 buffer. Solid line is from nonlinear least-squares fitting using the Davies surface tension isotherm (eq 1) equation.
where γ0 is the surface tension of the solvent, R is the gas constant, T is absolute temperature, Γ∞ is surface excess at saturation, z is the surfactant valence, is dielectric constant of water, C1∞ is the bulk surfactant concentration, C3∞ is the bulk concentration of any added electrolyte co-ion, F is Faraday’s constant and Ψs is the surface potential. Assuming that each mer behaves similarly
in regards to adsorption, a pseudo-single-surfactant approach28 was adopted for use in surface tension isotherm data analysis. The term θ is the quotient of surface excess at a given concentration, Γ, and Γ∞ and given by
zFΨs 4RT (2 RT(C1∞ + C3∞))1/2 1 - cosh zF 2RT
(26) Davies, J. T. Proc. R. Soc. London, Ser. A 1958, 245, 417. (27) Datwani, S. S.; Stebe, K. J. Langmuir 2001, 17, 4287.
θ)
βC1∞/R Γ ) Γ∞ exp(zFΨs/RT) + βC1∞/R
(2)
2416 Langmuir, Vol. 24, No. 6, 2008
Ertekin et al.
Figure 4. Surface tension isotherm for the four-arm anionic (-C2F5) oligo(fluorooxetane) 2 in pH 8 buffer. Solid line is from nonlinear least-squares fitting using the Davies surface tension isotherm (eq 1) equation.
where β/R reflects the adsorption strength and is related to the free energy of adsorption through the following.
∆Gads ) - RT ln(β/R)
(3)
To make eqs 1 and 2 soluble, the surface potential, Ψs, must be related to the surface charge density, σ, through the GouyChapman relationship 4
σ ) zFΓeq ) (2RT
∑ i)1
Ci∞)1/2 sinh
( ) zFΨs 2RT
(4)
where C2∞ is the bulk surfactant co-ion concentration and C4∞ is the bulk added counterion concentration. Numerically, Ψs was found by using a root finder with eq 4. The two adjustable parameters, β/R and Γ∞, were evaluated using a LevenbergMarquardt nonlinear least-squares algorithm29 iteratively by minimizing the difference (χ2) between experimental surface tension and theoretical values given by eq 1 while solving for Ψs and adjusting θ after each cycle. The results from parametric fitting of surface tension isotherm data according to eq 1 are shown in Table 1. An obvious feature from Figures 3 and 4 is the very low critical micelle concentrations (cmc) ≈ 4.2 × 10-6 and 2.4 × 10-6 mol/L observed for 1 and 2, respectively. These values are approximately an order of magnitude lower than those seen in a previous study of bolaamphiphilic (two-arm) analogues of the four-arm materials examined in the current work.22 The cmc values given here are also about an order of magnitude lower than typical anionic perfluorooctanoic acid-based fluorosurfactants (one-arm).22 The properties of long perfluoroalkyl chain surfactants have been the subject of numerous investigations.30-40 The low values of cmc (28) Daniel, R. C.; Berg, J. C. Langmuir 2002, 18, 5074. (29) Sprott, J. C., Numerical Recipes: Routines and Examples in Basic; Oxford University Press: Oxford, 1998; Chapter 13. (30) Duns, G. J.; Reeves, L. W.; Yang, D. W.; Williams, D. S. J. Colloid Interface Sci. 1991, 145, 270. (31) Eastoe, J.; Paul, A.; Rankin, A.; Wat, R.; Penfold, J.; Webster, J. R. P. Langmuir 2001, 17, 7873. (32) Fung, B. M.; Mamrosh, D. L.; O’Rear, E. A.; Frech, C. B.; Afzal, J. J. Phys. Chem. 1988, 92, 4405. (33) Guo, W.; Brown, T. A.; Fung, B. M. J. Phys. Chem. 1991, 95, 1829. (34) Ishikawa, Y.; Kuwahara, H.; Kunitake, T. Chem. Lett. 1989, 1737. (35) Lin, I. J. J. Phys. Chem. 1972, 76, 2019. (36) Muller, N.; Simsohn, H. J. Phys. Chem. 1971, 75, 942. (37) Sadtler, V. M.; Giulieri, F.; Krafft, M. P.; Riess, J. G. Chem. Eur. J. 1998, 4, 1952.
exhibited here are a reflection of the decreased solubility of four-arm surfactants compared to two-arm oligo(fluorooxetane)s and one-arm long perfluoroalkyl chain surfactants. A very low cmc value ≈2 × 10-6 mol/L has been reported for a three-arm, multifunctional hydrocarbon (-(CH2)10H hydrophobe) amphiphile.2 Surface excesses at saturation, Γ∞, and subsequent molecular areas measured for 1 and 2 are quite different from those found in earlier work on two-arm analogues. For 1 and 2, Γ∞ ) (1.87 ( 0.06) × 10-6 and (1.95 ( 0.06) × 10-6 mol/m2, respectively, leading to molecular areas of 88.6 ( 3 and 84.9 ( 3 Å,2 respectively. The molecular areas determined for 1 and 2 are ≈2× those determined for bolaamphiphilic versions of 1 and 2. Undoubtedly, the increased molecular area demands of 1 and 2 compared to bolaamphiphilic derivatives are a consequence of increased geometrical dimensionality of hydrophobic chains about a tetrahedral initiator core. Molecular area demands given in the current work exceed values of ∼30 Å2 observed for small molecule, single chain fluorosurfactants of the -(CF2)nF (nj ≈ 8) type.41-43 Single-chain fluorosurfactants are able to adsorb to the air-water interface such that the chain approaches orthogonality to the interface at saturation and with areas that reflect cross-sectional dimensions of an Rf chain. Clearly, such a configuration is not possible in the current case. For 1 and 2, the arrangement of perfluoroalkyl groups is more “brushlike” about pendent groups and that will have a significant influence on the ability of these perfluoroalkyl groups to pack at the air-water interface. Molecular dynamics calculations substantiate the differing packing arrangement of bolaamphiphilic anionic oligo(fluorooxetane) surfactants compared to small-molecule fluorosurfactants.44 The anionic ammonium sulfate groups, along with the relatively hydrophilic initiator core, are “anchored” in the water phase just below the interface while the “arms” of the bolaamphiphile form “loops” that extend into the air phase. Such a configuration has been observed and hypothesized on triblock, nonionic, ethoxylated surfactants.45,46 While 1 and 2 do not form stable monolayer films at the air-water interface, the tetraol precursors 1a and 2a do. It is assumed that the hydroxyl groups of 2a will function equivalently to the ammonium sulfate groups of 2 as “anchors” at the airwater interface and that its behavior will be a reasonable facsimile of 2. Films of 2a were cast on a Langmuir trough from CHCl3 and compression/expansion isotherms were recorded. The isotherm is shown in Figure 5. It is important to note that the tetraols 1a and 2a are relatively viscous, water-insoluble liquids and, therefore, the collapse observed typically for solid surfactants would not be expected upon monolayer compression. Rather, surface pressure should reach a compressibility-limited point near the value of the pure liquid surface tension. While the viscosity of 2a precludes direct measurement of surface tension with present instrumentation, the limiting value of π corresponds to a reasonable surface tension of pure 2a ≈ 29 mN/m. Annotated on the isotherm are estimates of transition molecular areas of ∼157, 732, and 1630 Å2 determined by the intersection of tangents constructed through linear sections and extrapolation to the (38) Tadros, T. F. J. Colloid Interface Sci. 1980, 74, 196. (39) Taylor, C. K. Paint Coatings Ind. 1999, May, 58. (40) Ulmius, J.; Lindman, B. J. Phys. Chem. 1981, 85, 4131. (41) Bernett, M. K.; Zisman, W. A. J. Phys. Chem. 1963, 67, 1534. (42) Hann, R. A. Molecular Structure and Monolayer Properties. In LangmuirBlodgett Films; Roberts, G., Ed.; Plenum: New York, 1990; p 17. (43) Acero, A. A.; Li, M.; Lin, B.; Rice, S. A.; Goldmann, M.; Azouz, I. B.; Goudot, A.; Rondelez, F. J. Chem. Phys. 1993, 99, 7214. (44) Stephenson, B. C.; Beers, K. J. J. Phys. Chem. B 2006, 110, 19393. (45) Leclerc, E.; Daoud, M. Macromolecules 1997, 30, 293. (46) Be´ghin-Aquie´, V.; Leclerc, E.; Daoud, M.; Douillard, R. J. Colloid Interface Sci. 1999, 214, 143.
Four-Arm Anionic Fluorosurfactants
Langmuir, Vol. 24, No. 6, 2008 2417
Table 1. Adsorption Parameters from Davies Isotherm Fit, ∆Gads and cmc Values for Four-Arm, Anionic Oligo(fluorooxetane)s 1 and 2 material
Γ (×106 mol/m2)
area (Å2/molecule)
β/R (×10-9 m3/mol)
∆Gads (kcal/mol)
cmc (×106 mol/L)
χ2
1 2
1.87 ( 0.06 1.95 ( 0.06
88.6 ( 3 84.9 ( 3
2.23 ( 0.4 4.90 ( 1
-12.7 ( 0.1 -13.2 ( 0.1
4.17 2.43
5.18 1.38
abscissa. On the basis of simple molecular modeling of a molecule of 2a in free space and geometric arguments, a fully extended molecule is estimated to occupy an excluded area ≈3600 Å2. An increase in π to this compressed area would neither be anticipated nor is observed. Evolution from the gaseous-to-liquid expanded state occurs at a molecular area ≈1630 Å2. The liquid expandedto-liquid condensed transition occurs near 732 Å2. Upon further compression, the molecule will distort to accommodate its neighbors and the interface. The approximate cross-sectional area of a single “arm” of 2a is ∼500-1000 Å2. At the compressibility limit of pure, condensed film of 2a, the transition area (∼157 Å2) approximates well results from molecular dynamics calculations for saturation adsorption on bolaamphiphilic analogues of 1 and 2 (∼153 and 165 Å2/molecule, respectively)44 and is about 2× the results for saturation areas estimated from solution of the Davies surface tension isotherm in the present work (∼85 Å2/molecule for 2). The previous
Figure 5. π-A isotherm for four-arm tetraol 2a at the air-water interface for compression (solid line) and expansion (dashed line). Values indicated on curve are estimated transition areas.
discussion must be considered with caution. While film studies of 2a may capture the salient features of the configurational entropy of 2, it certainly does not depict enthalpy accurately. Compression of a film of 2a will result in intermolecular hydrogen-bonding attraction, while doing so with a film of 2 should lead to intermolecular Coulombic repulsion. Undoubtedly, these differences will affect isotherm behavior. Shown also in Figure 5 is the expansion cycle of the isotherm and substantial hysteresis is noted. Furthermore, the hysteresis is larger as the molecular area is increased. If the discussion regarding interpretation of transition areas were accurate, vide supra, then this would be the region of overlap and incipient intermolecular hydrogen-bonding interactions.4 Intermolecular hydrogen-bonding interactions would not be expected to contribute significantly to π near the compressibility limit at low molecular areas. The four-arm anionic oligo(fluorooxetane)s demonstrate large adsorption strengths, β/R, with the consequence of large negative heats of adsorption, ∆Gads, calculated to be -12.7 ( 0.1 and -13.2 ( 0.1 kcal/mol for 1 and 2, respectively. The large negative heats of adsorption give an indication of the low reversibility of adsorption that will become obvious during the discussion, vide infra, of interfacial dilational rheological parameters. The heats of adsorption shown here can be compared to those measured for bolaamphiphilic analogues of 1 and 2 and small-molecule, single-chain fluorosurfactants of the -(CF2)nF (nj ≈ 8) type in the -7 to -8 kcal/mol range.22 Considering the large molar mass of 1 (Mn ≈ 4300 g/mol) and 2 (Mn ≈ 6000 g/mol) and architecture consisting of a fluorinated corona about an initiator core, mass transport compared to the bolaamphiphilic versions of 1 and 2 was examined. Diffusion coefficients were obtained by solution of the mass transfer equation given by Ward and Tordai47,48
Γ(t) ) 2
(Dπ)
1/2
[C0t1/2 -
1/2
Csd[(T - t)1/2]]
(5)
where Γ(t) is surface excess as a function of t, D is the diffusion coefficient, C0 is the bulk surfactant concentration, T is the time needed to reach an equilibrium surface tension, and Cs is the subsurface concentration as a function of t and calculated from eq 6 after solving for C1∞ as a function of θ.
Cs ) C1∞ )
Figure 6. Dynamic surface tensions for four-arm anionic (-CF3) oligo(fluorooxetane) 1 along with fits according to eq 5. Shown are 5 × 10-8 (square, solid), 1 × 10-7 (circle, dash), 2.49 × 10-7 (triangle, dot), 5.5 × 10-7 (inverted triangle, dash-dot), 1.1 × 10-6 (diamond, dash-dot-dot), and 2.98 × 10-6 mol/L (left triangle, short dash) solutions in pH 8 buffer.
∫0T
R/β exp(zFΨs/RT) 1-θ
(6)
The first term in brackets on the right side of eq 5 accounts for the adsorption rate of surfactant to the interface while the second term reconciles the desorption rate that becomes more important at long times and/or higher concentrations. Equation 5 was evaluated using a finite difference method.48 Shown in Figures 6 and 7 are dynamic surface tension data along with fits using eq 5. The estimated diffusion coefficients for 1 and 2 are concentration-dependent and decreasing functions of concentration ([M]) as observed previously for oligo(fluorooxetane) surfactants22 and are presented graphically in (47) Ward, A. F. H.; Tordai, L. J. Chem. Phys. 1946, 14, 453. (48) Aksenenko, E. V. WardTordai Software Package. In Studies in Surface Science 13; Mo¨bius, D., Miller, R., Eds.; Elsevier Science: Amsterdam, 2001.
2418 Langmuir, Vol. 24, No. 6, 2008
Ertekin et al.
Figure 7. Dynamic surface tensions for four-arm anionic (-CF3) oligo(fluorooxetane) 1 along with fits according to eq 5. Shown are 3.92 × 10-8 (square, solid), 1.76 × 10-7 (circle, dash), 4.31 × 10-7 (triangle, dot), 9.31 × 10-7 (inverted triangle, dash-dot), and 2.98 × 10-6 mol/L (diamond, dash-dot-dot) solutions in pH 8 buffer.
Figure 8. Log-log plot of diffusion coefficients as a function of concentration for the four-arm, anionic oligo(fluorooxetane)s 1 (9) and 2 (b) along with linear least-squares fits shown as solid and dashed lines, respectively. Table 2. Diffusion Coefficient as a Function of Concentration for Four-Arm, Anionic Oligo(fluorooxetane)s 1 and 2 material 1
2
concentration (mol/L)
D (cm2/s)
5 × 10-8 1 × 10-7 2.49 × 10-7 5.5 × 10-7 7.49 × 10-7 1.1 × 10-6 2.98 × 10-6 3.92 × 10-8 1.76 × 10-7 4.31 × 10-7 9.31 × 10-7 2.98 × 10-6
1.9 × 10-3 2.1 × 10-3 5.8 × 10-4 9.2 × 10-5 6.1 × 10-5 4.7 × 10-4 3.6 × 10-5 8.2 × 10-3 8.2 × 10-4 2.4 × 10-4 3.6 × 10-5 3.0 × 10-5
Figure 8 and tabularly in Table 2. Linear least-squares fitting of the data yielded log D ) -1.2 ( 0.2 log [M] - 11 ( 1 and log D ) -1.4 ( 0.2 log [M] - 12 ( 1 for 1 and 2, respectively. While the molar mass of 1 and 2, (Mn ≈ 4300 and 6000 g/mol, respectively) is substantially higher compared to the less massive bolaamphiphilic analogues (Mn ≈ 1650-2000 g/mol) and a typical small molecule, -C8F17-type surfactant (Mn ≈ 700 g/mol) detailed previously, diffusion coefficients evaluated at a given
concentration (∼10-6 mol/L) are comparable. The StokesEinstein diffusion equation predicts D ∝ r-1 where r is hydrodynamic radius.49 It is evident from a comparison of the diffusion coefficients estimated for the three different architectures and Mn that r does not scale linearly with molecular weight. Undoubtedly, this is due to the increased geometrical dimensionality of 1 and 2 (four-arm) versus bolaamphiphilic (twoarm) and small-molecule (one-arm) surfactants and allows for increased molar mass without the concomitant increase in molecular volume expected for surfactants of lower dimensionality. This factor will have substantial consequences, vide infra, on dilational interfacial rheological behavior. The diffusion coefficients estimated using eq 5 assumes that dynamics are controlled solely by diffusion with no other relaxation mechanism operative. This may not be the case entirely, and this can be judged by the quality of fits shown in Figures 6 and 7. As the size and entropic configurational freedom of molecules increases, it is anticipated that structural rearrangement will present a barrier to occupying the surface and that the barrier will increase as a strong function of concentration. Often, this results in irreversible adsorption as in the case of many polyampholytes such as proteins.50-56 As is evidenced in Figures 6 and 7, the deviation from ideal Ward-Tordai diffusional behavior increases with increasing concentration. Molecular weight polydispersity and its influence on dynamics is another issue that needs to be reconciled. Dilational Interfacial Rheological Studies. As demonstrated in an earlier section, the four-arm anionic oligo(fluorooxetane) molecules 1 and 2 are surface active. In addition to reducing surface tension, the adsorption of molecules also imparts interfacial viscoelastic behavior. The viscoelastic properties attributed to an interface play a crucial role in determining the stability of foams and emulsions, as well as in the uniformity of film coatings. Despite the widespread applications of fluorosurfactants, there have been few published studies on the interfacial rheological properties of these materials at the airwater interface.57,58 Dilational interfacial rheological properties of the four-arm anionic oligo(fluorooxetane) surfactants 1 and 2 in pH 8 buffered water at room temperature have been measured using an oscillating bubble rheometer. This technique has been outlined elsewhere57,59-61 and will be covered here only briefly. In this technique, the air-water interface is subjected to a periodic expansion and contraction of the surface area and the change in the surface tension at the interface (∆γ) due to the change in the (49) Einstein, A. InVestigations on the Theory of Brownian MoVement; Dover: New York, 1956. (50) Benjamins, J.; Lucassen-Reynders, E. H. Surface Dilational Rheology of Proteins Adsorbed at Air/Water and Oil/Water Interfaces. In Proteins at Liquid Interfaces; Mo¨bius, D., Miller, R., Eds.; Elsevier Science: Amsterdam, 1998; p 341. (51) Bos, M. A.; van Vliet, T. AdV. Colloid Interface Sci. 2001, 91, 437. (52) Freer, E. M.; Yim, K. S.; Fuller, G. G.; Radke, C. J. Langmuir 2004, 20, 10159. (53) Petkov, J. T.; Gurkov, T. D.; Campbell, B. E.; Borwankar, R. P. Langmuir 2000, 16, 3701. (54) Patino, J. M. R.; Nin˜o, M. R. R.; Sa´nchez, C. C. J. Agric. Food Chem. 1999, 47, 3640. (55) Graham, D. E.; Phillips, M. C. J. Colloid Interface Sci. 1979, 70, 403. (56) Wang, Z.; Narsimhan, G. Langmuir 2005, 21, 4482. (57) Frømyr, T.; Hansen, F. K.; Kotzev, A.; Laschewsky, A. Langmuir 2001, 17, 5256. (58) Kausch, C. M.; Kim, Y.; Russell, V. M.; Medsker, R. E.; Thomas, R. R. Langmuir 2003, 19, 7354. (59) Cao, X.; Li, Y.; Jiang, S.; Sun, H.; Cagna, A.; Dou, L. J. Colloid Interface Sci. 2004, 270, 295. (60) Miller, R.; Wu¨stneck, R.; Kra¨gel, J.; Kretzschmar, G. Colloids Surf., A: Physico. Eng. Asp. 1996, 111, 75. (61) Casca˜o, Pereira, L. G.; The´odoly, O.; Blanch, H. W.; Radke, C. J. Langmuir 2003, 19, 2349.
Four-Arm Anionic Fluorosurfactants
Langmuir, Vol. 24, No. 6, 2008 2419
Figure 9. Elastic, E′, (9) and viscous, E′′, (b) moduli for the fourarm anionic oligo(fluorooxetane) 1 at pH 8 as a function and frequency at 1.66 × 10-8 (A) and 1.10 × 10-6 mol/L (B).
interfacial area (∆A) is determined. For a sinusoidal variation in area of the interface (having an amplitude Aa, an equilibrium value of Ao and a frequency ω) represented by the equation,
∆A ) A - A0 ) Aa sin (ωt)
(7)
the change in surface tension can be described as
∆γ ) γ - γ0 ) γa sin(ωt + φ)
(8)
where γa is the measured amplitude of the surface tension, γo is the equilibrium surface tension, and φ is the phase angle. The surface viscoelastic dilational modulus is then defined as
|E| )
γa Aa/A0
(9)
that can be decoupled into real and imaginary components E′ and E′′, respectively, using the following equations
E′ ) |E| cos φ E′′ ) |E| cos φ
(10)
where E′ is the storage modulus representing elastic properties of the adsorbed layer and E′′ is the loss modulus arising due to the transport of molecules between the bulk and the surface or relaxation in the surface plane. Dilational interfacial rheological data as a function of frequency at two concentrations spanning isotherm extremes were collected on 1 and 2 in pH 8 buffered solutions and are shown in Figures 9 and 10, respectively. Overall, the response surfaces exhibited for |E|, E′, E′′, and φ as a function of concentration and frequency are typical of behavior expected for surfactants.60 For both 1 and 2 at two extremes in concentration, E′′ is rather invariant of frequency, while E′ exhibits a monotonic increase with increasing frequency. Such behavior would be expected for both the case of diffusional exchange62 and interfacial relaxation.63 For fluorosurfactants containing only -CF3 and -C2F5 groups, values of |E| and E′ observed for 1 and 2 are substantially larger (∼40-70 mN/m) than those observed in previous work on twoarm bolaamphiphilic analogues (∼10-20 mN/m).58 As in the (62) Lucassen, J.; van den Tempel, M. Chem. Eng. Sci. 1972, 27, 1283. (63) Noskov, B. A. Colloid. Polym. Sci. 1995, 273, 263.
Figure 10. Elastic, E’, (9) and viscous, E”, (b) moduli for the four-arm anionic oligo(fluorooxetane) 2 at pH 8 as a function and frequency at 1.47 × 10-8 (A) and 2.37 × 10-6 mol/L (B).
previous study, both |E| and E′ were found to increase with increasing perfluoroalkyl chain length. This is expected on the basis of the relative adsorption of perfluoroalkyl types.64 The loss modulus, E′′, is a measure of mass transport (diffusion) of molecules to and from the interface during dilation/contraction or relaxation in the surface plane and its contribution to |E| is given through the phase angle, φ (eq 8). For surfactants of the “soluble” type, both E′′ and φ are finite with 0 e φ e π/4 if relaxation is governed by diffusional exchange of molecules between the surface and near-surface.62 Values of E′′ and φ for 1 and 2 are lower (∼10-12 mN/m; 12-20°) than those observed previously for the two-arm analogues (10-20 mN/m; 30-50°) of the same perfluoroalkyl chain length.58 The decreased values of E′′ and φ for 1 and 2 versus those observed for the bolaamphiphilic analogues studied earlier may indicate a lesser contribution of diffusion, a change in relaxation mechanism, or a different relaxation time scale. A conclusion awaits a more detailed examination of rheolgical properties. The heats of adsorption for 1 and 2 (-12.7 ( 0.1 and -13.2 ( 0.1 kcal/mol, respectively) have greater negative values compared to the other materials (∼ -7 to -8 kcal/mol), suggesting stronger adsorption at the air-water interface. On average, molecules 1 and 2 will spend more time at the surface than just below and will be more reluctant to vacate the air-water interface upon dilation. As a consequence, there will be less mass transport contribution to E′′ and φ and, ultimately, |E|.
Conclusions Oligomeric, four-arm tetraols can be prepared readily from a fluorinated oxetane monomer and an oligo(alkylene oxide) tetraol initiator. Conversion of the oligomeric, four-arm tetraols to ammonium sulfate salts 1 and 2 was accomplished in high yield by a sulfation reaction using oleum, followed by neutralization with ammonium hydroxide. Both 1 and 2 are surface active with low cmc’s in the range ≈10-6 mol/L. Surface tension isotherms for 1 and 2 were measured and the data fitted parametrically to the Davies surface tension isotherm equation, yielding molecular areas at saturation ≈88.6 ( 3 and 84.9 ( 3 Å2/molecule, respectively, and ∆Gads ≈ -12.7 ( 0.1 and -13.2 ( 0.1 kcal/ mol, respectively. Dynamic surface tension data were fitted parametrically using the Ward-Tordai mass transport equation (64) Edwards, D. A.; Brenner, H.; Wasan, D. T. Interfacial Transport Processes and Rheology; Butterworth-Heinemann: Boston, 1991; Chapter 2.
2420 Langmuir, Vol. 24, No. 6, 2008
to provide diffusion coefficients that were found to be concentration-dependent. Despite being relatively more massive than two-arm analogues, the diffusion coefficients measured for 1 and 2 were found to be comparable at a given concentration (∼10-6 mol/L). This is a conseqence of the hyperbranched architecture of the four-arm species versus a two-arm analogue. Dilational interfacial rheological properties for 1 and 2 were measured using an oscillating bubble rheometer in pH 8 buffered solution. Dilational interfacial rheology was dominated by viscoelastic, |E|, and storage moduli, E′. The loss modulus, E′′, and phase angle, φ, were small and approached zero depending on concentration and frequency. The relative magnitudes of the rheological parameters indicate nearly irreversible adsorption of 1 and 2 to the air-water interface. This is reflected in the relatively large, negative ∆Gads found for 1 and 2 versus two-arm analogues
Ertekin et al.
and small fluorosurfactants with long (-(CF2)∼8F) perfluoroalkyl chains. Acknowledgment. OMNOVA Solutions, Inc. is acknowledged for financial support of A.E. during this work. The authors wish to thank Professor Mark Foster of the Polymer Science Department at the University of Akron for the use of his Langmuir trough. Supporting Information Available: 1H and 13C spectra for 1a and 2a in CDCl3, larger versions of structures shown in Figure 2 and |E|, E′ and E′′ versus ∆A/A for 1 in pH 8 solution. This material is available free of charge via the Internet at http://pubs.acs.org. LA7031175
