pubs.acs.org/Langmuir © 2009 American Chemical Society
Equilibrium and Surface Rheology of Monolayers of Insoluble Polycations with Side Chains Beatriz Miranda,† Hani M. Hilles,‡ Ramon G. Rubio,‡ Hernan Ritacco,‡ Deodato Radic,† Ligia Gargallo,† Michele Sferrazza,§ and Francisco Ortega*,‡ † Departamento de Quı´mica Fı´sica, Facultad de Quı´mica, Pontificia Universidad Cat olica de Chile, Vicu~ na Mackenna 4860, Casilla 306, Santiago 22, Chile, ‡Departamento de Quı´mica Fı´sica I, Facultad de Quı´mica, Universidad Complutense, 28040-Madrid, Spain, and §D epartement de Physique, Universit e Libre de Bruxelles, Boulevard du Triomphe, CP223 Bruxelles, Belgium
Received May 18, 2009. Revised Manuscript Received July 8, 2009 We have studied monolayers of poly(n-tetradecyl 4-vinylpyridinium-co-4-vinylpyridine) bromide with different degrees of quaternization at the air-water interface. The isotherms (surface pressure vs area) present several phase transitions: at low monolayer coverage, there is a phase transition over a characteristic area that increases on increasing the quaternization degree. This behavior can be rationalized in terms of a mean-field theory of 2D semiflexible polymeric chains and could be an indication of a disorder-order transition from a 2D isotropic liquid (IL) at low surface concentration to a 2D nematic phase (N) at higher concentrations. Low-frequency oscillatory strain experiments show that at low surface coverage the monolayers exhibit highly nonlinear behavior, even for low strain amplitude, whereas at higher surface coverage the response is linear for strains higher than 20%. In addition, stress relaxation experiments show a minimum in the characteristic times that coincide with the transition area. These unexpected results at low surface coverage might be characteristic of the system or related to the fact that the oscillatory experiments do not strictly correspond to constant surface-coverage conditions. However, they are in agreement with high-frequency viscoelasticity, obtained by surface quasielastic light scattering, that shows that the dilational viscosity is higher at low surface concentration than for concentrations beyond the surface phase transition. At higher coverage, there is a second phase transition, after which the isotherms present hysteresis, which is not observed below. Ellipsometry indicates that, after this transition, the monolayer thicken, which may be related to 3D growth into a multilayer.
I. Introduction Polyelectrolytes are linear macromolecule chains containing a large number of charged or chargeable groups that, in a polar solvent such as water, dissociate into charges associated to the polymer backbone and counterions dispersed in the solution. They are used in many different fields because of their ability to adsorb at interfaces and modify surface properties and because of the interactions between colloidal particles and their environment. Charged macromolecules can be used to construct ordered ultrathin solid films by successive layering of alternating anionic and cationic polymeric coatings;layer-by-layer deposition (soluble polyelectrolytes);or by the Langmuir-Blodgett technique (insoluble polyelectrolytes), and they can be used in a wide range of technologies, including lubrication, controlled flocculation of colloidal dispersions, adhesion, and stabilization.1-3 To construct Langmuir-Blodgett (LB) films efficiently, characterization of the behavior of the polyelectrolyte in the monolayers, which will serve to order precursors of the solid-supported thin film, needs to be performed. In fact, noncharged polymer monolayers (PMs) are distinct to normal low-molecular-weight surfactants in many respects, their phase diagrams apparently being simpler.4 At very low surface concentrations, Γ, PMs usually do not show 2D gas to 2D liquid transitions at measurable surface *To whom correspondence should be addressed:
[email protected]. (1) Dautzenberg, H.; Jaeger, W.; K€otz, B. P. J.; Seidel, C.; Stscherbina, D. K. Polyelectrolytes: Formation, Characterization and Application; Hanser Publishers: Munich, 1994. (2) Oosawa, F. Polyelectrolytes; Marcel Dekker: New York, 1971. (3) F€orster, S.; Schmidt, M. Adv. Polym. Sci. 1995, 120, 50. (4) Jones, R. A. L.; Richards, R. W. Polymers at Surfaces and Interfaces; Cambridge University Press: Cambridge, U.K., 1999.
Langmuir 2009, 25(21), 12561–12568
pressures, Π. Furthermore, at moderately higher surface concentrations, PMs do not show liquid-liquid transitions; instead, they present a continuous change from virial-like (Π ≈ Γ2) behavior to a Π ≈ Γy (y > 2) power law dependency, where the value of exponent y depend on the nature of the polymer-interface system.5-7 For very high surface concentrations, the system becomes very complex, and the monolayer is entirely formed from a meltlike state of the polymer, with y , 1.8 Furthermore, in the case of some very specific tacticities, 2D solid behavior has been reported for high values of Γ (e.g., isotactic poly(methyl methacrylate)).9 The 3D osmotic view of polymer solutions has been invoked in order to understand the experimental isotherms of polymer monolayers.10 The Π versus Γ behavior of PMs is derived by comparing the osmotic pressure of 3D polymer solutions and the surface pressure. In fact, the scaling analysis of the Π - Γ isotherms, in the so-called semidilute state of PMs, can be expressed as5 Π ¼ Γy ; y ¼
2ν 2ν -1
ð1Þ
where ν is the critical exponent for the relation between the radius of gyration and the molecular weight Rg ≈ M w ν
ð2Þ
(5) Vilanove, R.; Rondelez, F. Phys. Rev. Lett. 1980, 45, 1502. (6) Rivillon, S; Monroy, F; Ortega, F.; Rubio, R. G. Macromolecules 2003, 36, 4068. (7) Esker, A. R.; Kim, C.; Yu, H. Adv. Polym. Sci. 2007, 209, 59. (8) Cicuta, P.; Stanvik, E. J.; Fuller, G. G. Phys. Rev. Lett. 2003, 90, 236101. (9) Brinkhuis, R. H. G.; Schouten, A. J. Macromolecules 1991, 24, 1497. (10) des Cloizeaux, J.; Jannick, G. Polymers in Solution; Clarendon Press: Oxford, U.K., 1990.
Published on Web 08/18/2009
DOI: 10.1021/la901762u
12561
Article
Miranda et al.
Figure 1. Chemical scheme of poly(n-alkyl 4-vinylpyridinium-covinylpyridine) randomly quaternized. In our case, n=14.
The ν value in 2D is still a matter of controversy, and it is assumed to be 0.57 (y = 8.14) under Θ-solvent conditions and 0.75 (y = 3) under good-solvent conditions, corresponding to the Flory exponent for a self-avoiding trajectory in 2D.11,12 However, it is not clear whether a real 2D Θ or sub-Θ state exists for some PMs or just 3D aggregation takes place in some cases, preventing any 2D phase separation. Most of the studies of insoluble polymer Langmuir monolayers have been devoted to neutral polymers, and less effort has been dedicated to PMs of polyelectrolytes. Owing to the water solubility of polyelectrolytes, studies of Langmuir films are scarce. To overcome this solubility limitation, several strategies have been implemented, such as the use of partially ionized chains (e.g., random and block copolymers) or the use of hydrophobized polyelectrolytes. The last one has been the strategy used in the present work, in which we have used n-tetradecyl side chains bonded to a poly(4-vinylpyridinium) chain. Kawaguchi’s group13,14 has pointed out that PMs of poly(n-alkyl 4-vinylpyridinium) show a phase transition at surface pressures below 1 mN m-1. According to these authors, the nature of the transition is of the type LE-LC. More recently, Miranda et al.15 have shown that this family of polyelectrolytes also presents a second surface phase transition at surface pressures that depend on the quaternization degree and on the length of the alkyl side chain. Similar results have been presented by Davis et al.16 In this work, we will present a detailed study of the equilibrium isotherms and the dilatational viscoelasticity of monolayers of poly (n-tetradecyl 4-vinylpyridinium-co-4-vinylpyridine) bromide with different degrees of quaternization (i.e., charge and grafting density) at the air/water interface.
II. Experimental Section II.1. Samples. The polylectrolytes used are poly(n-alkyl 4vinylpyridinium-co-4-vinylpyridine) salts of the general structure sketched in Figure 1. The polymers were prepared from commercially available poly(4-vinylpyridine) of average molecular weight 50 000 (Polysciences Ltd.). The synthesis proceeded by mixing the poly(4-vinylpyridine) with n-tetradecyl bromide in chloroform and refluxing for several weeks. The reaction was followed by FTIR spectroscopy to measure the relative area of the bands at 1600 and 1640 cm-1 corresponding to the pyridine ring and to the quaternized one, respectively. The reaction was stopped at the desired quaternization degree, and the final product was precipitated and washed several times with ethyl acetate. The purity and the final quaternization degree were checked by FTIR. (11) Vilanove, R.; Poupinet, D.; Rondelez, F. Macromolecules 1988, 21, 2880. (12) Monroy, F.; Ortega, F.; Rubio, R. G. J. Phys. Chem. 1999, 103, 2061. (13) Kawaguchi, M.; Itoh, S.; Takahashi, A. Macromolecules 1987, 20, 1052. (14) Kawaguchi, M.; Itoh, S.; Takahashi, A. Macromolecules 1987, 20, 1056. (15) Gargallo, L.; Miranda, B.; Rios, H.; Gonzalez-Nilo, F.; Radic, D. Polym. Int. 2001, 50, 825. (16) Davis, F.; Hodge, Ph.; Liu, X.-H.; Ali-Adib, Z. Macromolecules 1994, 27, 1957.
12562 DOI: 10.1021/la901762u
The solvent used to prepare the polyelectrolyte solutions was chloroform, which has good spreading characteristics on the water surface. The chloroform was obtained from Riedel de Haen (Germany) and was of chromasolv quality. The polyelectrolyte concentration of the spreading solutions was between 0.1 and 1 mg/mL, depending on the polymer and the particular experiment, and no effect of the concentration of the spreading solution on the isotherms was detected. Doubly distilled, deionized water from a Milli-Q-RG unit has been used in all of the experiments. Its resistivity was always higher than 18 mΩ cm-1, and its organic matter content was less than 3 ppb. Before the spreading of the polyelectrolyte solutions, the surface tension of water was checked by the plate method in order to ensure that there were no surfaceactive impurities. In addition, pure chloroform was spread in order to check that it did not contain any surface-active impurities. II.2. Isotherms and Dilatational Viscoelasticity. Isotherms and low-frequency dilatational viscoelasticity experiments have been performed on a Langmuir balance, Nima 702 model (Coventry, U.K.). The trough was thermostatted at 25.0 ( 0.1 C for all experiments, and two barriers made of Teflon were use to modify the surface area. After dropwise addition of the appropriate volume of the polymer solution to the surface of the water, a waiting time of 30 min was allowed before recording the isotherms at a barrier speed of 5 cm2/min. Compression-expansion cycles were carried out for all polymers, and no appreciable hysteresis was found until high surface pressure was reached (see below). Isotherms were also obtained point-by-point by sequential additions, and no differences were found with the ones obtained by continuous compression below the high-surface-pressure phase transition. Reported isotherms are the average of at least three isotherms that coincide within 5% in surface area. The monolayers were studied with stress relaxation, oscillatory strain, and surface quasi-elastic light scattering (SQELS) techniques, and a brief description of these techniques is presented in the following sections. II.2.1. Stress Relaxation. The stress relaxation, σ(t), has been obtained as a function of time, t, after a sudden uniaxial in-plane compression of the Langmuir film performed with two coupled barriers. The surface stress is defined as σ(t) ΔΠ(t) =Π(t) Π(t f ¥). The surface stress is dilational in nature, σD, and acts as a restoring force to recover the final equilibrium state, Π(t f ¥), of the film when the strain ceases. The surface dilation [θ = -ΔA/A0 = -(A0 - A)/A0 = ΔΓ/Γ, where A denotes area, A0 denotes the initial value, and Γ denotes the surface concentration] was adjusted to accomplish a linear stress-strain relationship (i.e., θ < 10%). To obtain significant results, the final equilibrium state should recover the expected value according to the equilibrium isotherm. In the present systems, this is a further confirmation of the lack of solubility of the polyelectrolytes. The dilational stress usually follows a single-exponential decay, although in some cases a sum of two or more exponentials is needed to explain the experiments t t σðtÞ ¼ σ 01 exp þ σ 02 2 exp þ ::: τ1 τ2
ð3Þ
In eq 3, σ0i and τi are the amplitudes and relaxation times, respectively.17,18 II.2.2. Oscillatory Strain. In this kind of experiment,19,20 the barriers of the Langmuir trough are driven to a sinusoidal motion of constant frequency, ω, that imposes an oscillatory strain (17) Rivillon, S.; Mu~noz, M. G.; Monroy, F.; Ortega, F.; Rubio, R. G. Macromolecules 2003, 36, 4068. (18) Cardenas, A. E.; Valera, A. I. Colloids Surf., A 1993, 79, 115. (19) Hilles, H. M.; Maestro, A.; Monroy, F.; Ortega, F.; Rubio, R. G.; Velarde, M. G. J. Chem. Phys. 2007, 126, 124904. (20) Noskov, B. A.; Loglio, G. Colloids Surf., A 1998, 143, 167.
Langmuir 2009, 25(21), 12561–12568
Miranda et al.
Article
perturbation in the film. The monolayer is strained in the Langmuir trough by this uniaxial in-plane oscillatory motion of amplitude, u. Depending on the dimensions of the trough, u can be varied from 0 to typically 5% of the initial area A0 to perform an experiment within the linear regime. It must be considered that in this kind of rheological experiment nonlinearity can be reached by too high a strain amplitude or strain rate and that to perform experiments at variable strain amplitudes with constant frequency the barrier speed c must be changed, u = c/w. Because the Langmuir trough has a fixed width, the strain ratio is actually a dilation/compression ratio. If the trough barriers are placed at a distance L for a given time t, then we have uðtÞ ¼
A0 -AðtÞ L0 - LðtÞ ¼ A0 L0
ð4Þ
The time dependence of the strain can be expressed as uðtÞ ¼
u0 ½1 þ expðiωtÞ 2
ð5Þ
where u0 is the strain amplitude. In the linear regime, the viscoelastic response follows the sinusoidal shape of frequency ω imposed by the strain ΠðtÞ ¼ Π0 þ σðtÞ
ð6Þ
σ0 σðtÞ ¼ ½1 þ exp iðωt þ jσ Þ 2 Here, σ0 is the stress amplitude, measured relative to the background pressure Π0, and φσ is the phase delay of the stress response due to the viscous response of the monolayer. In the linear viscoelastic regime, the low-frequency elasticity modulus, ε, and the dilational viscosity,κ, can be obtained as follows ε ¼ jεjcos jσ
ωK ¼ jεjsin φσ
ð7Þ
where σ0 jεj ¼ u0
ð8Þ
It must be stressed that in the oscillatory strain experiments (and in general in the so-called compressional surface rheological techniques) the surface coverage is changing during the experiment and as a consequence the surface pressure also changes. The analysis of the results of this technique assumes that the strain amplitude is small enough to have a linear Π-A dependence within the oscillation. If this condition is not fulfilled, then apparent nonlinear viscoelasticity may show up as a consequence of a nonlinear Π-A dependence. If this is so, then experiments with smaller strain amplitudes must be performed until linearity is obtained.
II.2.3. Surface Quasi-Elastic Light Scattering (SQELS). In this method, the light that is scattered in a quasi-elastic way by thermal capillary waves, naturally occurring in the surface of a liquid, is measured. Light scattering arises from refractive index fluctuations due to the capillary roughness, with typical amplitudes in the range of a few angstroms.21 The SQELS setup used here is the one described previously,22,23 where a polarized, very stable 25 mW He-Ne laser beam (λ = 632.8 nm) goes through a spatial filter that expands the beam and (21) Langevin, D., Ed., Light Scattering by Liquid Surfaces and Complementary Techniques; Marcel Dekker: New York, 1992 (22) Mu~noz, M. G.; Luna, L.; Monroy, F.; Rubio, R. G.; Ortega, F. Langmuir 2000, 16, 6657. (23) Monroy, F.; Ortega, F.; Rubio, R. G. J. Phys. Chem. B 1999, 103, 2061.
Langmuir 2009, 25(21), 12561–12568
ensures a Gaussian intensity profile. A couple of mirrors and a lens (L1) direct the beam through a transmission diffraction grating from Align-Rite (U.K.), slightly out of focus of L1. In all experiments reported here, we have used 5-μm-wide dark lines spaced 275 μm from each other. With a beam stirrer and a second lens L2, a 1:1 image of the grating is formed onto the liquid surface with an angle of incidence of 45. The light scattered by the capillary waves, at a given wavevector, is detected in heterodyne mode; with mixing on the photomultiplier (Hamamatsu), the different diffraction orders are reflected from the surface with the light scattered in the same direction. The output of the photomultiplier were analyzed using a spectrum analyzer (Stanford Research 760FFT). The spectrometer has been calibrated using several fluids (water, acetone, and diethyl-ether) and can be used in the wavevector range of 100
