Langmuir 2000, 16, 6131-6135
6131
Effect of Temperature and Molecular Weight on Binding between Poly(ethylene oxide) and Cationic Surfactant in Aqueous Solutions Khine Yi Mya,† Alexander M. Jamieson,‡ and Anuvat Sirivat*,† The Petroleum and Petrochemical College, Chulalongkorn University, Bangkok, Thailand, and Department of Macromolecular Science, Case Western Reserve University, Cleveland, Ohio 44106 Received December 6, 1999. In Final Form: April 20, 2000 The formation of a polymer-surfactant complex upon mixing a nonionic polymer, poly(ethylene oxide) (PEO), with a cationic surfactant, hexadecyltrimethylammonium chloride (HTAC), is studied by observing the changes in conductivity, specific viscosity (ηsp), and hydrodynamic radius (Rh) in terms of the solution temperature and PEO molecular weight. The conductivity data show clearly that an interaction between PEO and HTAC occurs at a temperature above 25 °C, as indicated by a decrease in the critical association concentration in the presence of PEO relative to the critical micelle concentration of a surfactant in the absence of PEO. Diminishing hydrophilicity of PEO upon increasing temperature is suggested to induce a stronger interaction between PEO and HTAC. The higher the temperature and the PEO molecular weight are, the more pronounced the peaks in ηsp and Rh will be. The amount of surfactant molecules per chain varies with molecular weight, but it is independent of temperature. Our results suggest that the strength of the polymer surfactant interaction increases with the increasing hydrophobicity of each component.
Introduction The interaction between water-soluble polymers and surfactants is of considerable importance, and it has been a field of intense research in recent years.1 These studies have generated insight into the structure of the polymersurfactant complexes and how the properties of a surfactant solution change upon the addition of polymer. In general, nonionic polymers interact strongly with anionic surfactants but weakly with cationic surfactants. This is interpreted as due to the bulkiness of the cationic headgroup,2 to a more favorable interaction between anionic surfactants and the hydration shell of the polymers,3 or to an electrostatic repulsion between the polymer and the surfactant due possibly to protonation of the polymer.4 Carlsson et al.5 have reported that the more hydrophobic the polymer is, the greater the binding of surfactant to polymer will be. Several studies have convincingly demonstrated6-9 that cationic surfactants such as tetradecyltrimethylammonium chloride and bromide (TTAC and TTAB) and hexadecyltrimethylammonium chloride and bromide (HTAC and HTAB) bind to nonionic cellulose derivatives such as hydroxypropyl cellulose (HPC) and ethyl(hydroxyethyl) cellulose (EHEC). * Corresponding author. † The Petroleum and Petrochemical College. ‡ Case Western Reserve University. (1) Goddard, E. D.; Ananthapadmanabhan, K. P. Interactions of Surfactants with Polymers and Proteins; CRC Press: Boca Raton, FL, 1993. (2) Ruckenstein, E.; Huber, G.; Hoffmann, L. Langmuir 1987, 3, 382. (3) Witte, F. M.; Engberts, J. B. F. N. J. Org. Chem. 1987, 52, 4767. (4) Moroi, Y.; Akisida, H.; Saito, M.; Murata, R. J. Colloid Interface Sci. 1977, 61, 233. (5) Carlsson, A.; Karlstrom, G.; Lindman, B. J. Phys. Chem. 1989, 93, 3673. (6) Winnik, F. M.; Winnik, M. A.; Tazuke, S. J. Phys. Chem. 1987, 91, 594. (7) Zana, R.; Binana-Limbel’e, W.; Kamenka, N.; Lindman, B. J. Phys. Chem. 1992, 96, 5461. (8) Wang, G.; Olofsson, G. J. Phys. Chem. 1995, 99, 5588. (9) Goldszal, A.; Costeux, S.; Djabourov, M. Colloids Surf., A 1996, 112, 141.
Winnik et al.6 studied complex formations between HPC and HTAC and found that an interaction occurs at a surfactant concentration lower than the critical micelle concentration (cmc), which corresponds to the concentration at which free micelles appear in the system in the absence of a polymer. The term “critical aggregation concentration” (cac) has been coined to denote the surfactant concentration at which surfactant molecules form aggregates upon interacting with the polymer chain. Frequently, interaction between surfactant and polymer is manifested as a decrease in the cac relative to the cmc. For example, Zana et al.7 studied the interaction of EHEC with HTAB and HTAC and determined the cac as well as the micellar aggregation number, N, using time-resolved fluorescence quenching. They reported that, in the presence of a polymer, the cac decreases below the cmc, and furthermore, they found a lower value of N for polymerbound micelles. Other studies have investigated the association between more flexible polymers and cationic surfactants. Brackman and Engberts10,11 reported on the interaction of cationic surfactants with poly(vinyl methyl ether) (PVME), poly(propylene oxide) (PPO), poly(ethylene oxide) (PEO), and poly(vinyl pyrrolidone) (PVP) polymers. Complex formation with surfactant micelles was observed for both PVME and PPO polymers, indicated by a reduction in both the cac relative to the cmc and by a decrease in the aggregation number, N, of bound micelles. On the other hand, PEO and PVP polymers showed no interaction with the surfactant. Of particular relevance to the present paper, Anthony and Zana12 studied the effect of temperature on the interaction of PEO with TTAB and a nonionic surfactant, n-dodecyl octaoxyethylene glycol monoether (C12E8), using conductivity and time-resolved fluorescence quenching methods. Their results show that an interaction between PEO and TTAB occurs only at temperatures (10) Brackman, J. C.; Engberts, B. F. N. Langmuir 1992, 8, 424. (11) Brackman, J. C.; Engberts, B. F. N. Langmuir 1991, 7, 2097. (12) Anthony, O.; Zana, R. Langmuir 1994, 10, 4048.
10.1021/la991588u CCC: $19.00 © 2000 American Chemical Society Published on Web 06/22/2000
6132
Langmuir, Vol. 16, No. 15, 2000
above 35 °C, evidenced by the fact that the cac decreases below the cmc. In addition, they determined12 that above 35 °C, the PEO-bound TTAB micelles not only have smaller N values but also are more ionized relative to those micelles formed in the absence of polymer. In contrast, PEO shows no interaction with C12E8 even at higher temperatures.12 The binding of charged surfactant micelles to flexible nonionic polymers has been demonstrated to cause a characteristic change in the hydrodynamic properties of the solution. Specifically, upon addition of the anionic surfactant sodium dodecyl sulfate (SDS) to PEO, the specific viscosity, ηsp, increases to a maximum value and then decreases.13,14 The phenomenon has been shown to occur because of expansion of the polymer coil due to electrostatic repulsions between the bound micelles. The maximum corresponds to the point of saturation of bound micelles,13,14 and the subsequent decrease in viscosity on further addition of surfactant is due to the screening of charge interactions by excess counterions. A similar behavior was subsequently observed15 for the binding of cationic micelles (HTAB) to the semiflexible polymer HPC, and it was further shown that a maximum occurs in the hydrodynamic radius, Rh, determined via dynamic lightscattering (DLS) analysis, which correlates with the viscosity maximum. The present work examines the effect of temperature and the polymer molecular weight on the hydrodynamic properties of solutions containing poly(ethylene oxide) (PEO) and the cationic surfactant, HTAC. Experimental Section Materials and Sample Preparation. Poly(ethylene oxide) (PEO), with the quoted molecular weights of 1 × 105, 6 × 105, 9 × 105, and 4 × 106 g/mol, was purchased from Aldrich Chemical Co. All PEO samples were dried in a vacuum oven for 1 day at room temperature before the solution was prepared. Hexadecyltrimethylammonium chloride (HTAC) was provided by Unilever Holding Inc. and used as received. Deionized sterile water was purchased from the Thai Pharmaceutical Organization. Prior to solution preparation, the water was filtered through 0.2 µm Millipore membrane filters to remove dust particles. The polymer solutions were prepared by percentage (w/v) in prefiltered deionized sterile water and stored at room temperature. Before measurements, all sample solutions were centrifuged at 8000 rpm for 1 h and then filtered with 0.22 or 0.45 µm Millipore filters, depending on the concentration and molecular weight of the polymer, to remove the residual dust particles. Methods. A conductivity meter (Orion model 160) was used to determine the cmc for pure surfactant solution and the cac for polymer-surfactant complex solutions. Kinematic viscosity measurements were performed using Cannon-Ubbelohde viscometers with appropriate efflux times for each solution. Specific viscosities of PEO solutions were computed neglecting the density difference between solution and solvent. The average molecular weights of the polymers were determined by viscosity measurements to be Mw ) 1.04 × 105, 5.97 × 105, 8.86 × 105, and 4.00 × 106 g/mol, deduced from the Mark-Houwick-Sakurada equation,16 where K ) 1.25 × 10-4 (in dL/g) and a ) 0.75 for the unperturbed PEO coil at 30 °C, consistent with the manufacturer’s specified values. Dynamic light-scattering (DLS) measurements were performed with an instrument (Malvern Instruments Company, model 4700) which uses vertically polarized laser light of wavelength 514.5 nm. The translational diffusion coefficients were obtained by dynamic light- scattering analysis of the (13) Goddard, E. D. Colloids Surf. 1986, 19, 255. (14) Chari, K.; Antalek, B.; Lin, M. Y.; Sinha, S. K. J. Chem. Phys. 1994, 100 (7), 5294. (15) Hormnirun, P.; Sirivat, A.; Jamieson, A. M. Polymer 2000, 41, 2127. (16) Bailey, F. E., Jr.; Koleske, J. B. Poly (Ethylene Oxide); Chemicals and Plastics Technical Center, Union Carbide Corporation, New York; Academic Press: New York, 1976.
Mya et al. normalized intensity autocorrelation function, g2(τ), which is related to the electric field autocorrelation function,17 g1(τ), through the Siegert equation:
g2(τ) ) 1 + β|g1(τ)|2
(1)
where β is a spatial coherence factor determined by the collection efficiency of the detection apparatus. The function g1(τ) can be expressed as the Laplace transform of the distribution of relaxation rates (Γ):
g1(τ) )
∫G(Γ) exp(-Γτ)dΓ
(2)
Cumulant analysis is applied to determine the mean decay rate (Γ h ):
ln g1(τ) ) Γ hτ +
1 µ2 1 µ3 (Γ h τ)2 (Γ h τ)3 + ... 2 2! Γ 3! Γ3
(3)
The second moment, µ2, represents the variance of the distribution of relaxation rates; µ3 represents the skewness of the distribution. Γ h is related to the apparent diffusion coefficient, Dapp, as
Γ h ) Dappq2
(4)
Here, q is the scattering wave vector, which depends on the scattering angle θ (q ) 4πn/λ sin(θ/2)). The center of mass diffusion (Dm) of the polymer chain is obtained by extrapolation of Dapp to zero scattering angle using the following equation:
Dapp ) Dm (1 + Cq2Rg2 + ...)
(5)
Here, C is a coefficient determined by the slowest internal mode of motion in the particle and by the size, flexibility, and polydispersity of the polymer.18 The diffusion coefficient at infinite dilution (D0) can be obtained by extrapolation to zero polymer concentration, Cp, as
Dm ) D0 (1 + kDcp + ...)
(6)
where kD characterizes the concentration dependence of Dm due to thermodynamic and hydrodynamic interactions. The hydrodynamic radius can then be calculated from D0 using the StokesEinstein equation
Rh ) kBT/6πηsD0
(7)
Here, kB is Boltzmann’s constant, T is the absolute temperature, and ηs is the viscosity of the solvent. In our analysis of dilute solutions of PEO-containing surfactant, we have neglected the concentration correction for eq 6 and determined an approximate value of Rh from Dm at finite concentrations.
Results and Discussion Effect of Temperature. Figure 1 shows the variation of the cmc and the cac as a function of temperature in the solutions of HTAC in water and in a PEO-water mixture using conductivity measurements. The polymer concentration was fixed at 0.1 g/dL [Mw (PEO) ) 5.97 × 105 g/mol], and the surfactant concentration was varied in all experiments. Within the temperature range of 25-50 °C, the cmc does not vary, whereas the cac decreases rapidly as temperature increases above 25 °C, indicating some interaction between PEO and HTAC. At 25 °C, the cmc and the cac values are essentially identical, indicating that negligible interaction takes place between PEO and HTAC at or below 25 °C. This result contrasts with the observations of Anthony and Zana12 that the onset of an (17) Brown, W.; Nicolar, T. Dynamic Light Scattering: The Method and Some Appllications; Oxford Science Publications: London, 1993. (18) Brown, W. Laser Light Scattering; Clarendon Press: Oxford, 1996; Chapter 13, p 448 and reference therein.
Poly(ethylene oxide) and Cationic Surfactant Binding
Langmuir, Vol. 16, No. 15, 2000 6133
Figure 1. Dependence of the cmc and the cac on temperature: PEO concentration, 0.1 g/dL; PEO molecular weight, 5.97 × 105 g/mol.
interaction between PEO and TTAB occurs only at temperatures above 35 °C. We can express the decrease in cac with rising temperature in terms of a reduction in the free energy of micellization. Using the phase separation model19 as a first approximation and neglecting changes in micelle size, we define the reduction in free energy as
-∆G ) RT ln[cmc/cac]
(8)
From this equation, we calculated how much PEO lowers the free energy of micellization in units of RT and found that the reduction in free energy is 0.05 RT units at 25 °C, which increases to 1.74 RT units at 50 °C. Figure 2 shows the variation of the specific viscosity with surfactant concentration at three temperatures. At low surfactant concentration, the specific viscosity decreases slightly with increasing temperature (see the insert to Figure 2). Here, the concentration of the surfactant is comparable to the cac, and only a few surfactant micelles bind to the polymer. PEO becomes less polar with increasing temperature, and hence, the chain contracts, resulting in a decrease in specific viscosity. At higher surfactant concentrations, an increase in solution specific viscosity is observed with increasing temperature. This indicates that a sufficient number of surfactant micelles bind to PEO to produce a chain expansion due to electrostatic repulsions between bound micelles. Figure 2 shows that the viscosity increase is more pronounced at higher temperatures, i.e., that the chain expansion is larger at higher temperatures. This result seems consistent with previous studies,7,12 which indicate that a substantial increase in the ionization of bound micelles occurs as temperature increases, which increases the strengths of electrostatic repulsions. At a certain surfactant concentration, the specific viscosity reaches a maximum but then decreases with a further increase in HTAC concentration. As noted above, the viscosity maximum has been interpreted13,14 to indicate the point at which PEO chains are saturated with surfactant micelles. The decrease in viscosity above the saturation point is (19) Tanford, C. The Hydrophobic Effect: Formation of Micelles in Biological Membranes; John Wiley & Sons: New York, 1980.
Figure 2. Variation of the specific viscosity with surfactant concentration at different temperatures: PEO concentration, 0.1 g/dL; PEO molecular weight, 5.97 × 105 g/mol. [Insert: Expanded plot of data in the surfactant concentration range (0-1.6 mM).]
interpreted13,14 as due to the effect of added counterions (Cl-) in screening the repulsive interactions among micelles bound to the PEO chains. Our results are generally consistent with the findings of previous studies,7,12,20 which indicate that a less polar polymer segment provides a better nucleus for surfactant self-assembly and that increase in temperature leads to a decrease in polarity and, hence, a more effective interaction between polymer and surfactant.7,12,20 An interesting feature of Figure 2 is that the position of the maximum in ηsp is independent of temperature, occurring at around 5 mM HTAC. This indicates that the saturation concentration of binding is independent of temperature, and it corresponds to 0.2 mol of HTAC per mole of PEO repeating unit. This agrees with previous findings that the saturation concentration of binding of an ionic surfactant to a polymer is only weakly dependent on polymer hydrophobicity.12 In Figure 3, we compare in more detail the temperature dependence of ηsp at HTAC concentrations below the cac versus the temperature dependence at concentrations far above the cac. Clearly, ηsp decreases below the cac with temperature, whereas above the cac, ηsp increases sharply with temperature. The variation in viscosity in dilute solution is dictated primarily by the hydrodynamic volume of the polymer chain, assuming no intermolecular association of chains. In the binary system of PEO in water, ηsp decreases as temperature increases. This behavior is well understood for PEO in aqueous solution.21 The PEO chains assume a more compact conformation at elevated temperatures, presumably because the poorer solvation of the PEO outweighs any weak electrostatic repulsions at higher temperature. Also evident in Figure 3 is that as a small amount of surfactant (c e cmc) is added to the polymer solution at a fixed temperature, ηsp decreases relative to the viscosity (20) Lindman, B.; Carlsson, A.; Karlstrom, G.; Malmsten; M. Adv. Colloid Interface Sci. 1990, 32, 183. (21) Kirk-Othmer. Encyclopedia of Chemical Technology, 4th ed.; John Wiley & Sons: New York, 1996; Vol. 19.
6134
Langmuir, Vol. 16, No. 15, 2000
Figure 3. Effect of temperature on specific viscosity at various surfactant concentrations for the PEO molecular weight of 5.97 × 105 g/mol. Solution concentrations were (b) 0.1 g/dL of PEO, (0) 0.1 g/dL of PEO + 0.15 mM HTAC, (4) 0.1 g/dL of PEO + 0.5 mM HTAC, ()) 0.1 g/dL of PEO + 2.5 mM HTAC, and (3) 0.1 g/dL of PEO + 5 mM HTAC.
of the pure polymer solution. Such a decrease has been observed by others14,22 at a surfactant concentration lower than one micelle per polymer coil. Above the cac and at a fixed temperature, ηsp increases with surfactant concentration, obviously due to electrostatic repulsions between the bound micelles. At 25 °C, a particular behavior is observed: no variation in ηsp occurs with surfactant concentration when it is well above cmc, suggesting a small or negligible interaction between the PEO and the HTAC. This result is consistent with that observed in the conductivity measurement (Figure 1). At the saturation point of binding (CHTAC ) 5 mM, Figure 2), ηsp increases most steeply with temperature, indicating that the electrostatic repulsion becomes most pronounced at the maximum binding condition. The present results may be compared with those of Anthony and Zana,12 who observed an interaction between PEO and TTAC only at temperatures above 35 °C. Thus, it appears that the more hydrophobic HTAC has a greater tendency for association with PEO. Figure 4 shows the apparent hydrodynamic radius (Rh) determined from the DLS experiments as a function of surfactant concentration at each temperature. Rh was calculated from eq 7 using Dm. Evidently, as surfactant is added, Rh increases to a maximum value, which occurs at a surfactant concentration similar to that of the specific viscosity, and then decreases again. The magnitude of the increase in Rh is larger at higher temperatures, again consistent with the viscosity data. These data confirm that the polymer coil expands as temperature and the surfactant concentration increase beyond the cac due to the binding of surfactant micelles to the PEO coils. Thus, the DLS results substantiate our interpretation of the viscosity measurements. The location of the maximum in the hydrodynamic radius corresponds to the maximum binding between PEO and HTAC. Beyond the saturation point, the hydrodynamic radius of polymer chain decreases due to the screening effect of Cl- counterions from the excess surfactant solution. (22) Mya, K. Y.; Jamieson, A. M.; Sirivat, A. Polymer 1999, 40, 5741.
Mya et al.
Figure 4. Hydrodynamic radius vs surfactant concentration at different temperatures: (O) 30 °C; (0) 35 °C; (4) 40 °C. PEO concentration was 0.1 g/dL, and PEO molecular weight was 5.97 × 105 g/mol.
Figure 5. Dependence of specific viscosity on surfactant concentration at different polymer molecular weights at 30 °C: PEO concentration, 0.1 g/dL. Molecular weights were ()) 1.04 × 105 g/mol, (O) 5.97 × 105 g/mol, (0) 8.86 × 105 g/mol, and (4) 4.00 × 106 g/mol.
Effect of Polymer Molecular Weight. To explore the effect of polymer molecular weight, the concentration of polymer was fixed at 0.1 g/dL at 30 °C, and the surfactant concentration was varied. Figure 5 shows the dependence of specific viscosity on the surfactant concentration at four different polymer molecular weights. In each case above the cac, ηsp increases due to the expansion of the PEO coil upon binding with surfactant micelles. The viscosity maximum is more prominent with increasing polymer molecular weight. This is expected because the hydrodynamic volume is an increasing function of molecular weightse.g., for flexible coils, ηsp ∼ Ma, with 0.5 < a < 0.8, depending on solvent quality.16
Poly(ethylene oxide) and Cationic Surfactant Binding
Figure 6. The ratio of ηsp(PEO + HTAC)/ηsp(PEO) as a function of surfactant concentration. The numbers described in the figure refer the molecular weights of polymer.
To illustrate this, we plot in Figure 6 the ratio ηsp(PEO + HTAC)/ηsp (PEO) as function of the surfactant concentration. There is approximate superposition of the data, with the exception of the lowest molecular weight, for which the viscosity change is perhaps too small to measure accurately. Also in Figure 5, we see that the position of the viscosity maximum is only weakly dependent on the PEO molecular weight, decreasing slightly as molecular weight increases. Thus, the mole ratio of the HTAC/PEO repeating unit at the saturation point decreases slightly with molecular weight, having a value of approximately 0.2 mole of HTAC/mole of PEO repeating unit for Mw ) 5.97 × 105 g/mol (weight ratio: 1.5 g of HTAC/mole of PEO), 0.17 mole of HTAC/mole of PEO repeating unit for Mw ) 8.86 × 105 g/mol (weight ratio: 1.2 g of HTAC/g of PEO), and 0.15 mole of HTAC/mole of PEO repeating unit for Mw ) 4.00 × 106 g/mol (weight ratio: 1.125 g of HTAC/g of PEO). Assuming the aggregation number of HTAC micelles to be 75, as reported by Zana et al.,7 this result implies there are 37 micelles per PEO chain when Mw ) 5.97 × 105 g/mol, 44 micelles per PEO chain when Mw ) 8.86 × 105 g/mol, and 185 micelles per PEO chain when Mw ) 4.00 × 106 g/mol. Note that this calculation gives an upper limit to the number of bound micelles, since it assumes all HTAC micelles are bound to PEO and there are no free micelles in the solution. The decrease in the saturation weight ratio of bound HTAC with increase of PEO molecular weight (Figure 6) reflects possibly that the probability of a micelle having multiple attachment sites to the same or different PEO chains increases with increase of molecular weight. Figure 7 shows the apparent values of Rh, derived from translational diffusion coefficients measured by DLS for various molecular weights of PEO at finite concentrations. The results are again consistent with the viscosity data,
Langmuir, Vol. 16, No. 15, 2000 6135
Figure 7. Hydrodynamic radius vs surfactant concentration of different PEO molecular weights at 30 °C: PEO concentration, 0.1 g/dL. Symbols identify the same data sets as in the figure 5. The numbers shown in Figure are the molecular weights of polymer.
exhibiting a maximum at an HTAC concentration which correlates with that observed in the viscosity data. Also, the magnitude of the increase in Rh is, as expected, greater for the higher molecular weights, since for flexible chains, Rh ∼ Mb, with 0.5 < b
