Langmuir 2007, 23, 7065-7071
7065
Salt Effect on the Nanostructure of Strong Polyelectrolyte Brushes in Amphiphilic Diblock Copolymer Monolayers on the Water Surface Ploysai Kaewsaiha, Kozo Matsumoto,† and Hideki Matsuoka* Department of Polymer Chemistry, Kyoto UniVersity, Kyoto 615-8510, Japan ReceiVed NoVember 29, 2006. In Final Form: March 10, 2007 The nanostructure of a spread monolayer of diblock copolymers of poly(hydrogenated isoprene) and poly(styrenesulfonate) at the air/water interface were studied by in situ X-ray reflectivity as a function of the brush density and salt concentration. When the monolayer was compressed beyond the “critical brush density”, its nanostructure changed from a flat, adsorbed “carpet” layer to a “carpet + brush” structure. The critical brush density was found to be about 0.12 nm-2, independent the brush length and salt concentration under a low-salt condition. The brush formation behavior was considered to be controlled by an electrostatic interaction between polyelectrolyte chains rather than by a steric hindrance. This might be because the distance between the chains at the critical point is rather long and also because of the effect of the salt on the critical brush density. The critical brush density increased at higher added salt concentration beyond 1 M. As a result, we found a new structure transition behavior of the polymer brushes between carpet-only and carpet + brush structures, which was induced by salt addition. Finally, we succeeded in the controlled release of salt ions from the salted brush layer by changing the brush density by compression of the monolayer.
Introduction Polymer brushes have been the focus of a large number of experimental and theoretical studies.1-11 An interesting category of these systems is the charged brush. Apart from fundamental interests, charged polymer brushes can be used to generate metallic nanoparticles within the brush layer using the confinement of gold or silver ions12 and can also be valuable in technological applications such as lubrication,13 drug delivery,14 and protein adsorption.15 Pincus16 and Borisov et al.17 were the first to predict that the counterions are mostly localized within the brush layer. The thickness of the brush has been shown to result from a balance of the osmotic pressure within the brush and the configurational elasticity of the chain. Recent theoretical work in this area6-8 has * To whom correspondence should be addressed. E-mail:matsuoka@ star.polym.kyoto- u.ac.jp. † Present address: Molecular Engineering Institute, Kinki University, Fukuoka 820-8555, Japan. (1) De Gennes, P. G. Macromolecules 1980, 13, 1069. (2) Milner, S. T. Science 1991, 251, 905. (3) Halperin, A.; Tirrell, M.; Lodge, T. P. AdV. Polym. Sci. 1992, 100, 31. (4) Rigoberto, C., Advincula, W. J., Brittain, K. C., Caster, J. R., Eds.; Polymer Brushes; Wiley: New York, 2004. (5) Zhulina, E. B.; Borisov, O. V.; Priamitsyn, V. A. J. Colloid Interface Sci. 1990, 137, 495. (6) Israels, R.; Leermakers, F. A. M.; Fleer, G. J. Macromolecules 1994, 27, 3087. (7) Lyatskaya, Y. V.; Leermakers, F. A. M.; Fleer, G. J.; Zhulina, E. B.; Birshtein, T. M. Macromolecules 1995, 28, 3562. (8) Zhulina, E. B.; Birshtein, T. M.; Borisov, O. V. Macromolecules 1995, 28, 1491. (9) Zhao, B.; Brittain, W. J. J. Am. Chem. Soc. 1999, 121 (14), 3557. (10) Tran, Y.; Auroy, P.; Lee, L. T. Macromolecules 1999, 32, 8952. (11) Lemieux, M.; Usov, D.; Minko, S.; Stamm, M.; Shulha, H.; Tsukruk, V. V. Macromolecules 2003, 36 (19), 7244. (12) Sharma, G.; Ballauff, M. Macromol. Rapid Commun. 2004, 25, 547. (13) Raviv, U.; Giasson, S.; Kampf, N.; Gohy, J. F.; Jerome, R.; Klein, J. Nature (London) 2003, 425, 163. (14) Constancis, A.; Meyrueix, R.; Bryson, N.; Huille, S.; Grosselin, J. M.; Gulik- Krzywicki, T.; Soula, G. J. Colloid Interface Sci. 1999, 217, 357. (15) Wittemann, A.; Haupt, B.; Ballauff, M. Phys. Chem. Chem. Phys. 2003, 1671; Macromolecules 1991, 24, 6335. (16) Pincus, P. A. Macromolecules 1991, 24, 2912. (17) (a) Borisov, O. V.; Birshtein, T. M.; Zhulina, E. B. J. Phys. (Paris) 1991, 1, 512. (b) Zhulina, E. B.; Borisov, O. V.; Birshtein, T. M. Macromolecules 1999, 32, 8189.
led to a phase diagram in which several distinct regimes can be identified.18 For a polyelectrolyte brush, the ion concentration inside the brush is considerably higher than that in the bulk phase as long as the concentration of salt added to the system is low. This regime is called the “osmotic brush” regime as the osmotic pressure of the ions in the brush layer is responsible for the extent of swelling of the brush. At sufficiently high added salt concentration Cs, the strong polyelectrolyte brush thickness decreases with increasing ionic strength of the solution (i.e., increasing salt concentration) due to screening of the charges on the chains, which strongly reduces electrostatic repulsion between the individual segments. In this region the brush thickness L is predicted to follow a scaling relationship, L ∝ Cs-1/3.17 This regime is known as the “salted brush” regime. In our experimental study, an exponent of -0.15 for the PSS brush length as a function of the salt concentration has been found.19 Here, we further studied the effect of the salt concentration on the nanostructure of the polyelectrolyte brush and found a surprising feature of brush formation behavior. The intrinsic counterions in the hydrophilic layers were studied under various conditions. In addition, we found a new structural transition behavior of the polymer brushes, which enables us to control the catch and release of salt ions from the salted brush by changing the brush density. A unique system for this study is a polymer brush at the air/ water interface. This can be achieved by utilizing an amphiphilic diblock copolymer with a long hydrophobic chain that cannot be dissolved in water. When the polymer is spread on the water surface, the hydrophobic block behaves as an anchor and the ionic hydrophilic block forms a polyelectrolyte brush.23 The great (18) Ahrens, H.; Fo¨rster, S.; Helm, C. A. Phys. ReV. Lett. 1998, 81, 4172. (19) Kaewsaiha, P.; Matsumoto, K.; Matsuoka, H. Langmuir 2004, 20, 6754. (20) Guenoun, P.; Davis, H. T.; Tirrel, M.; Mays, J. W. Macromolecules 1996, 29, 3965. (21) Fo¨rster, S.; Hermsdorf, N.; Bo¨ttcher, C.; Linder, P. Macromolecules 2002, 35, 4096. (22) (a) Guenoun, P.; Schlachi, A.; Sentenac, D.; Mays, J. W.; Benattar, J. J. Phys. ReV. Lett. 1995, 74, 3628. (b) Balastre, M.; Li, F.; Schorr, P.; Yang, J.; Mays, J. W.; Tirrell, M. V. Macromolecules 2002, 35, 9480. (23) Hamley, I. W. Block Copolymers in Solution: Fundamentals and Applications; Wiley: New York, 2005; Chapter 5.
10.1021/la063462k CCC: $37.00 © 2007 American Chemical Society Published on Web 05/19/2007
7066 Langmuir, Vol. 23, No. 13, 2007
Kaewsaiha et al.
Table 1. Characteristics of (Ip-h2)m-b-(SS)n Block Copolymers ma:nb
Mn
Mw/Mnc
degree of sulfonationd
215:31 220:55
20 600 25 000
1.10 1.11
0.92 0.87
a Number-average degree of polymerization of the PIp-h2 segment determined by 1H NMR for parent polymers in CDCl3. b Number-average degree of polymerization of the PSS segment determined by GPC for parent polymers (chloroform as an eluent with polystyrene standards). c Number-average molecular weight distribution of the block copolymer determined by GPC for parent polymers after hydrogenation (chloroform as an eluent with polystyrene standards). d Determined by elementary analysis.
Figure 1. π-A isotherm for (Ip-h2)215-b-(SS)31 monolayers on water at 0 and 1 M NaCl concentrations in the subphase.
advantage of this system is the brush density can be varied continuously by compression or expansion of the monolayer. Liquid interfaces are also generally very smooth and therefore ideal for reflectivity studies. In addition, the surface pressure is experimentally accessible, which is not the case for the solid/ liquid interface. The system of choice in this study comprises a hydrophobic block of poly(hydrogenated isoprene) (PIp-h2) and a strong polyelectrolyte block, poly(styrenesulfonate) (PSS). These block copolymers have been studied at the air/water interface19 and in solution.24 When the polymer forms a monolayer at the water surface, it is shown that there are two characteristic regions in the hydrophilic layer. In the “carpet layer” region the local structure is determined by non-Coulombic interactions, while in the “brush layer” region electrostatic interactions dominate. Materials and Methods Block Copolymer Synthesis. The diblock copolymer used is poly(hydrogenated isoprene)-b-poly(styrenesulfonate) (PIph2-bPSS), synthesized and characterized in the same way in our previous studies.19,24 PIp-b-PS diblocks have been synthesized by living anionic polymerization, and the polystyrene (PS) block was sulfonated after hydrogenation of the PIp block (about 90% 1,4-addition with 9-10% 3,4-addition). The characteristics of the polymers thus synthesized are summarized in Table 1. π-A Isotherm Measurement. For π-A isotherm measurements, we used an FSD-220 controller and a Langmuir-Blodgett (LB) trough (130 mm × 60 mm) made of aluminum coated with Teflon (USI System, Fukuoka, Japan). The sample copolymer was diluted in chloroform and then spread on the water surface in the LB trough with a microsyringe to prepare the monolayer. We used Milli-Q ultrapure water for the subphase and 0.001 M sodium phosphate aqueous solution for pH adjustment of the subphase, which resulted in counterion exchange of the SO3- group from H+ to Na+. That is, the hydrophobic segment PSS was transformed into sodium poly(24) Kaewsaiha, P.; Matsumoto, K.; Matsuoka, H. Langmuir 2005, 21, 9938.
Figure 2. X-ray reflectivity of (Ip-h2)215-b-(SS)31 monolayers on water containing NaCl at various concentrations in the subphase at chain densities of (a) 0.10 nm-2, (b) 0.12 nm-2, (c) 0.15 nm-2, and (d) 0.24 nm-2. The profiles were shifted downward by one decade. The solid lines are the curves fitted by the box model.
(styrenesulfonate) (NaPSS). A 30 min period was allowed for solvent evaporation. Then the surface was compressed by moving a Teflon barrier at a rate of 0.01 mm/s. The measurement was performed at room temperature. X-ray Reflectivity Measurement. The nanostructure of the monolayer on the water surface was directly investigated by in situ X-ray reflectivity (XR) measurement. The XR measurements were performed with a RINT-TTR-MA (Rigaku Corp., Tokyo, Japan) in which the X-ray generator and detector rotate vertically around the sample stage. The LB trough (130 mm × 60 mm) was mounted on the sample stage. The details of the XR apparatus and data analysis have been fully described elsewhere.25-27 The model fitting was carried out in a smaller q (scattering vector) range with good statistics, and the fitting quality of this range was checked by the R (%) value
R(%) )
∑(log x - log ∑(log x )
x
i 10 0
10
10
xic)2
i 2 c
× 100
(xi0 ) ith data point, xic ) ith calculated data point) as in our previous studies.19,25-27
Result and Discussion Structural Transition Behavior of the Polyelectrolyte Brushes by the Addition of Salt. The effect of salt on the monolayer nanostructure of (Ip-h2)215-b-(SS)31 was examined directly by measuring the XR at a constant area with changing salt concentration. Figure 1 shows the π-A isotherms of the (Ip-h2)215-b-(SS)31 monolayer with and without 1 M salt addition. (25) Mouri, E.; Kaewsaiha, P.; Matsumoto, K.; Matsuoka, H.; Torikai, N. Langmuir 2004, 20, 10604. (26) Mouri, E.; Furuya, Y.; Matsumoto, K.; Matsuoka, H. Langmuir 2004, 20, 8062. (27) Mouri, E.; Matsumoto, K.; Matsuoka, H.; Torikai, N. Langmuir 2005, 21, 1840.
Structural Transition and Controlled Salt Release
Langmuir, Vol. 23, No. 13, 2007 7067
Figure 3. δ profiles obtained by a box model fitting for Figure 1.
Figure 4. (a) Carpet layer thickness and (b) the brush layer thickness of (Ip-h2)215-b-(SS)31 monolayers on water containing NaCl at various concentrations as a function of the brush density. The dashed line in (b) shows an example of the change of the square of the radii of gyration 〈S2〉 for NaPSS in NaCl aqueous solution from ref 33.
Both curves show a smooth increase with decreasing area per molecule, which means formation of smooth, uniform monolayers. The two curves cross at about π ) 8 mN/m, and the salt-free condition shows a higher surface pressure in a small area region, which may reflect enhancement of the Coulomb interaction between brush chains. XR profiles and corresponding density profiles in different salt concentrations are shown in Figures 2 and 3, respectively. The XR profiles in Figure 2 show very clear
Kiessig fringes, and the box model fitting (solid line) shows excellent agreement with experimental data. The error bars based on R (%) are shown in each figure, and discussion about the accuracy in the fitting analysis has already been done in our previous paper.28 The density profiles normal to the surface in Figure 3 show formation of a carpet/brush double-layer structure at a higher brush density of more than 0.15 nm-2 and also shows carpet-only structure formation at a low brush density condition, i.e., 0.10 nm-2. The thicknesses of the carpet layer and the brush layer were evaluated from XR profiles and are presented in Figure 4 as a function of the salt concentration. The fitting parameters are shown in Table 2. The term “brush density” in this paper means the grafting density of PSS chains irrespective of the existence of a brush layer. The brush density was calculated from the hydrophobic layer thickness, i.e., the area per molecule, supposing that the hydrophobic layer should be a liquidlike layer. In the previous study, we examined the effect of salt addition on the monolayer nanostructure of (Ip-h2)220-b-(SS)55. The results in this study on the monolayer nanostructure of different PSS chain length (Ip-h2)215-b-(SS)31 polymers showed almost the same behavior. The π-A isotherm shifted toward the lower area per molecule at a high salt concentration. Also, no surface pressure dependence, hydrophilic chain length dependence, or salt concentration dependence on the carpet layer was confirmed.19,28 According to Table 2 and Figure 4, the thickness of the carpet layer of (Ip-h2)215-b-(SS)31 was also constant at 1-2 nm and did not show a remarkable dependence on the salt concentration. However, a small dependence of the carpet layer thickness on the salt concentration was confirmed at a high salt concentration, and this result will be discussed in a later section. On the other hand, the thickness of the brush layer was found to decrease after a critical salt concentration (Figure 4b, 0.24 nm-2 (0.2 M)). As was discussed before, the decrease in the brush layer thickness as a function of the salt concentration is due to the electrostatic screening effect by the salt ions. A very interesting observation in this study is the effect of salt at a lower brush density. Figure 2a shows the XR profile of the (Ip-h2)215-b-(SS)31 monolayer at a brush density of 0.10 nm-2 (28) Kaewsaiha, P.; Matsumoto, K.; Matsuoka, H. Langmuir 2007, 23, 20.
7068 Langmuir, Vol. 23, No. 13, 2007
Kaewsaiha et al.
Table 2. Nanostructure of the (Ip-h2)215-b-(SS)31 Monolayer on Water at Various Brush Densities and Salt Concentrations Determined by XRa Cs (M)
δSNa1 × 106
dSSNa1 (Å)
σSSNa1 (Å)
δSSNa2 × 106
dSSNa2 (Å)
σSSNa2 (Å)
δH2O × 106
σH2O (Å)
0 0.1 0.5 1
4.14 4.29 4.29 4.19
14 12 12 10
4 5 6 2
0.10 nm-2 b b b b
b b b b
b b b b
3.57 3.58 3.61 3.65
4 2 8 6
0 0.1 0.5 1
4.33 4.40 4.07 4.22
13 13 20 13
4 7 10 3
0.12 nm-2 3.94 3.83 3.73 b
26 27 22 b
3 2 2 b
3.57 3.58 3.61 3.65
19 9 3 8
0 0.1 0.5 1
3.92 4.23 3.89 4.04
18 11 19 18
3 3 3 2
0.15 nm-2 3.67 3.93 3.89 b
29 29 28 b
5 8 2 b
3.57 3.58 3.61 3.65
10 20 12 8
0 0.1 1
4.41 3.60 4.44
14 10 18
4 7 5
0.24 nm-2 4.14 4 4.28
32 34 26
6 4 2
3.57 3.58 3.65
14 10 4
a δ is defined as n ) 1 - δ - I, where n is the refractive index, d ) thickness of the layer, which is defined as the distance between the two x ri ri x interfaces, σx ) interface roughness represented as the standard deviation of the Gaussian function, x ) name of the layer (SSNa1, carpet layer; SSNa2, brush layer). b No brush layer.
Table 3. Nanostructure of the (Ip-h2)220-b-(SS)55 Monolayer on Water at Various Brush Densities Determined by XRa brush density (nm-2)
δSSNa1 × 106
dSSNa1 (Å)
σSSNa1 (Å)
δSSNa2 × 106
dSSNa2 (Å)
σSSNa2 (Å)
δH2O × 106
σH2O (Å)
0.12 0.15 0.18 0.19 0.23
4.47 4.18 4.30 4.29 4.29
10 14 11 12 12
3 5 5 5 6
b 3.83 3.92 3.96 4.07
b 75 81 83 85
b 6 4 10 6
3.57 3.57 3.57 3.57 3.57
9 23 23 23 30
a
Symbols are the same as in Table 2. b No brush layer.
Table 4. Nanostructure of the (Ip-h2)220-b-(SS)55 Monolayer on 1M NaCl Aqueous Solution at Various Brush Densities Determined by XRa brush density (nm-2)
δSSNa1 × 106
dSSNa1 (Å)
σSSNa1 (Å)
δSSNa1 × 106
dSSNa2 (Å)
σSSNa2 (Å)
δH2O × 106
σH2O (Å)
0.12 0.15 0.18 0.21 0.24 0.27
4.27 4.19 4.27 4.32 4.30 4.30
14 13 15 13 11 14
3 3 4 4 5 6
b 3.96 4.02 4.24 4.19 4.22
b 53 55 56 69 72
b 9 9 6 4 2
3.65 3.65 3.65 3.65 3.65 3.65
6 18 16 22 23 20
a
Symbols are the same as in Table 2. b No brush layer.
with various salt concentrations. By model fitting, we confirmed that monolayers at this brush density formed only a carpet layer under the hydrophobic layer. As reported previously, the carpetonly layer structure did not show any dependence on the salt concentration. Curve d in Figure 2a (1 M) could be reproduced by the twobox model, which means that the monolayer consists of only a carpet layer. The others could not, but satisfactory agreement was obtained by using a three-box model. As shown in Figure 4b, at a brush density lower than 0.15 nm-2 the brush layer thickness changed to zero when 1 M NaCl was added. In other words, it is a carpet-only structure under this condition. This is the first observation of a carpet-only/carpet + brush double-
layer transition induced by salt addition. This behavior is schematically shown in Figure 5. As reported previously,28 the carpet layer cannot be affected by salt due to the high ion condensation. After transition to the carpet-only layer, further salt addition would not affect the monolayer structure even in a high salt concentration such as 2 M NaCl. The same phenomenon can be seen in the polymergrafted charged colloidal systems, in which the hydrodynamic radius of the NaPSS-grafted particle was found to decrease after a critical salt concentration but did not change with further added salt.29 Salt Concentration Dependence of the Critical Brush Density. Parts a-d of Figure 6 are the histogram representations
Structural Transition and Controlled Salt Release
Langmuir, Vol. 23, No. 13, 2007 7069
Figure 6. Thickness of the PSS carpet layer (closed bars) and PSS brush layer (open bars) as a function of the brush density at various NaCl concentrations.
Figure 5. Schematic representation of the nanostructure change by salt concentration. (1) Osmotic brush region. The brush nanostructure is not influenced by salt addition in the bulk at low concentration. (2) Salted brush region. Added salt ions enter the brush layer. A screening effect occurs at salt concentration higher than the critical salt concentration, which results in shrinking of the brush chains. (3) Structural transition from the carpet + brush to carpet-only PSS layer by further addition of salt.
of the thickness of the carpet and brush layers as a function of the brush density at different salt concentrations. The closed bars represent the carpet layer thickness and open bars the brush layer thickness when it exists. As can be seen from the figures, a transition from a carpet-only structure to a carpet/brush doublelayered structure is observed at 0.12 nm-2 for 0-0.5 M NaCl and at 0.16 nm-2 for 1 M NaCl aqueous solution. In the absence of salt, PSS chains had a tendency to stretch into water due to the electrostatic interaction. However, after addition of salt, the electrostatic interaction between the brush chains was screened and the chain shrunk. As can be considered from the results in the previous section (Figures 3c and 4b), at a brush density lower than 0.15 nm-2 the brush layer could not form due to the screening effect at the higher salt condition. From Figure 4a, a small increase of the carpet layer thickness at a high salt concentration is noticed. This observation can be (29) Mohanty, P. S.; Harada, T.; Matsumoto, K.; Matsuoka, H. Macromolecules 2006, 39, 2016.
interpreted by joining of the brush chains into a carpet layer. As noticed from Figure 4b, at a higher salt condition, i.e., 1 M, above the critical salt concentration, the brush chains shrunk, and even the brush layer disappeared by the screening effect of the salt ions. A similar behavior was also observed for a weak acid brush for the transition by a brush density change. By compression of the monolayer, in other words, with increasing brush density, we observed a carpet-only to carpet + brush transition. Just after the brush layer formation, a small decrease of the carpet layer was observed.31 In conclusion, the critical brush density of the PIp-h2-b-PSS monolayer showed a higher value in a high salt concentration condition such as 1 M NaCl, i.e., above the critical salt concentration. In other words, the critical brush density is strongly correlated with the added salt concentration. Hence, it is fair to say that the Coulomb interaction is the major factor for brush layer formation for this strongly ionic polyelectrolyte brush. Mechanism of the Carpet to Brush Transition. There are many unknown points about brush layer formation of a spread diblock copolymer monolayer. Several works have been reported, and different hypotheses were formulated. The most common explanation is the increasing value of the surface coverage; i.e., the surface of the polymer chains has reached its plateau, upon further compression the hydrophilic blocks begin desorbing, thereby gradually forming a kind of brush, and the brushes stretch further upon lateral compression.30 However, the former explanation applies only in the case of a nonionic or weakly ionic polymer, which has no or weak electrostatic interaction in the monolayer. As was discussed previously,28 the critical brush density of PSS chains is 0.12 chain/nm2, independent of the PSS chain length. The formation of a brush layer in PIp-h2-b-NaPSS depends on the distance (30) Bijsterbosch, H. D.; de Haan, V. O.; de Graaf, A. W.; Mellema, M.; Leermakers, F. A. M.; Cohen, S. M. A.; van Well, A. A. Langmuir 1995, 11, 4467. (31) (a) Matsuoka, H.; Furuya, Y.; Kaewsaiha, P.; Matsumoto, K. Langmuir 2005, 21, 6845. (b) Matsuoka, H.; Furuya, Y.; Kaewsaiha, P.; Matsumoto, K. Macromolecules 2007, 40, 766.
7070 Langmuir, Vol. 23, No. 13, 2007
Kaewsaiha et al.
Figure 8. Brush layer thickness of (Ip-h2)220-b-(SS)55 monolayers on water containing NaCl at various concentrations along compression as a function of the brush density.
Figure 7. XR profiles for (Ip-h2)220-b-(SS)55 monolayers on (a) water and (b) 1 M NaCl aqueous solution at various surface pressures. The profiles were shifted downward by one decade. The solid lines are the curves fitted by the three-box model. The straight lines show the change of the fringe position.
between the polymer chains (2.88 nm) at the point of brush layer formation, and it is larger than that of a weakly ionic polymer, for example, poly(methacrylic acid) (1.44 nm for (Et2SB)34-b(PMAA)50)31 or poly(acrylic acid).32 Furthermore, the critical brush density of the PSS chain depends on the external salt concentration, but that of PAA does not. From these results, it is obvious that the main factor of brush layer formation of a strongly ionic amphiphilic diblock copolymer might be an electrostatic interaction in the hydrophilic layer, different from that of a weakly ionic polymer, which might be rather the steric repulsion or other factors. In Figure 4b, the decreasing profile for the radii of gyration of NaPSS, 〈S2〉, in aqueous solution with increasing salt concentration by Takahashi et al.33 was superimposed as a (32) Suetomi, Y.; Kaewsaiha, P.; Matsumoto, M.; Matsuoka, H. Polym. Prepr. Jpn. 2006, 55 (2), 4373 (33) Takahashi, A.; Kato, T.; Nagasawa, M. J. Phys. Chem. 1967, 71, 2001.
reference. They performed a systematic light scattering experiment for NaPSS in solution and found that 4.17 M NaCl(aq) is the Θ solvent for NaPSS. The highest salt conceentration in our experiment here, i.e., 2 M, is not so far from this Θ condition. In the bulk, the radius of gyration decreased rapidly with increasing salt concentration. In the case of the PSS brush layer, the PSS layer structure is not influenced until the critical salt concentration. However, over this salt concentration, a large amount of salt ions enter the brush layer and strongly screen the electrostatic interaction between the brush chains. Therefore, the drastic change of the brush layer thickness in Figure 4b is understandable. In addition, the formation of the brush was also seen as a transition from a “pancake” to a “brush” structure.30 However, from our study on both strongly ionic and weakly ionic polymer using different hydrophobic blocks, we observed that the carpet layer exists even after brushes are formed at lateral compression. Therefore, we have to reconsider the formation of the brush in the monolayer of the amphiphilic diblock copolymer at the air/ water interface as “a carpet/carpet + brush structural transition”. Nonlinear Swelling Behavior of Salted Brushes with Increasing Brush Density. In the salted brush regime, the ion concentration in the bulk solution is higher than that inside the brush layer and results in the entry of salt ions into the brush layer. When we compress the monolayer at this regime, the concentration of ions in the brush layer, i.e., counterions of PSS brushes and salt ions, will increase and there might be a point where the small-ion concentration becomes larger than that in the bulk. The XR profiles for the (Ip-h2)220-b-(SS)55 monolayer at various salt concentrations with compression are shown in Figure 7. In Figure 7a, at a salt-free condition, the shift of Kiessig fringes, indicated by straight lines in the figure, shows almost a linear change. However, in Figure 7b for the salted brush at 1 M NaCl, the shift line appears bent, indicating the critical salt condition for the shift line. Figure 8 shows the brush layer thickness as a function of the brush density at different salt concentrations. The osmotic brushes, as shown by the dashed line, show a linear increase of the brush length with compression. In contrast, the salted brushes, as shown by the solid line, show a sudden increase of the brush length along compression at a certain surface area; e.g., it increased suddenly at 0.25 nm-2 at a 2 M NaCl condition.
Structural Transition and Controlled Salt Release
Langmuir, Vol. 23, No. 13, 2007 7071
Figure 9. (a, b) δ profiles for hydrophilic layers obtained by a box model fitting for Figures 7a and b, showing the sudden decreasing electron density in the salted PSS layer during the compression. The arrows are drawn for clearity. (c) Schematic representation of salt ions squeezing out from the brush layer by compression of the monolayer.
The reason for the sudden increase in brush layer thickness is considered to be as follows. For the salted brush regime, the added salt ions are already inside the brush layer because the ion concentration in the bulk is higher than that in the brush layer. As a result, the concentration of total free ion in the brush layer was increased from 0.2 M19 to the same concentration in the subphase. When the monolayer is compressed, the ion concentration in the brush layer increases and becomes higher than that in the bulk, and then the excess salt ions are released from the brush layer into the bulk. As a result, the electrostatic interaction in the brush is enhanced, resulting in stretching of the brush, i.e., an increase of the brush layer thickness. Parts a and b of Figure 9 show the density profiles for the hydrophilic layer evaluated from Figure 7a and b. In the case of an osmotic brush (salt-free to 0.1 M NaCl(aq)) the electron density inside the brush layer linearly increased with compression (Table 3, Figure 9a). This is quite understandable because it is just an effect of compression. However, the electron density in the salted brush (Figure 9b) showed the characteristic behavior. When the monolayer was compressed from 20 to 22.8 mN/m (or the brush density was increased from 0.21 to 0.23 nm-2), the electron density did not simply increase, but once decreased and increased again (see also Table 4). This critical point showed agreement with the bending point of the XR profiles in Figure 7. The decrease in electron density should be attributed to a decrease of the ion concentration inside the brush layer because the brush density itself should increase. The salt ions escape at this point, and the brush layer acts like the osmotic brush at lateral compression (Figure 9b). In short, this observation means that small salt ions in the brush layer can be squeezed out by compression. This behavior is schematically shown in Figure 9c. This phenomenon certainly can be applied to the “controlled release” of small ions.
Conclusions Spread monolayers of diblock copolymers of poly(hydrogenated isoprene) and poly(styrenesulfonate) at the air/water interface were studied by X-ray reflectivity as a function of the brush density and salt concentration. At a low brush density, the
adsorbed PSS block forms a flat carpet layer. Upon compression beyond the “critical brush density” the brush layer was formed in addition to the carpet layer. The brush formation behavior of a strong polyelectrolyte is considered to be due to an electrostatic interaction in the hydrophilic layer rather than a steric hindrance because of the long distance between the chains at the critical brush density and dependence on the added salt concentration. The screening of electrostatic interaction in the PSS brush layer by the salt ions resulted in a decrease of the brush layer thickness and finally in a structural transition from a carpet + brush to a carpet-only structure at low brush density. There are many studies on the effect of salt on polyelectrolyte brushes and the decrease of the polyelectrolyte layer thickness as a function of the salt concentration. However, to the best of our knowledge, this is the first study on the structural transition of a polyelectrolyte brush to a carpet-only structure by salt addition. Furthermore, we succeeded in controlling counterions to and out of the brush region. The interesting feature of the system is that the XR measurement can be directly done at the air/water interface. The chain density can also be varied continuously using a movable Teflon barrier, which made it possible to investigate the counterion contribution in the brush layer along compression or expansion of the interface. In salted brush regimes, where the counterion concentration in the brush layer was lower than that in the bulk, small salt ions, which once entered from the bulk to the brush layer, can be squeezed out by compression. This is a very important discovery in the field of drug delivery systems. Acknowledgment. This work was supported by a Grantin-Aid for Scientific Research (A15205017) from the Ministry of Education, Culture, Sports, Science and Technology of Japan, to whom our sincere gratitude is due. This work was also supported by the Sasagawa Scientific Research Grant from The Japan Science Society and the 21st Century COE Program, COE for a United Approach to New Materials Science. LA063462K
