- Email: [email protected]
Structure of water in the vicinity of amphoteric polymers as revealed by Raman spectroscopy
Journal of Colloid and Interface Science 313 (2007) 461–468 www.elsevier.com/locate/jcis
Structure of water in the vicinity of amphoteric polymers as revealed by Raman spectroscopy Hiromi Kitano ∗ , Kyoko Nagaoka, Susumu Tada, Makoto Gemmei-Ide Department of Applied Chemistry, Graduate School of Science and Engineering, University of Toyama, Toyama 930-8555, Japan Received 27 February 2007; accepted 2 May 2007 Available online 22 May 2007
Abstract The structure and hydrogen bonding of water in an aqueous solution of amphoteric copolymers (poly(MA-r-DMAPMA), 3 × 103 < Mw < 104 ) composed of various ratios of methacrylic acid (MA) and N -[3-(dimethylamino)propyl]methacrylamide (DMAPMA) were analyzed using the band shapes of the O–H stretching in the polarized Raman spectra. The number of hydrogen bonds disrupted due to the presence of one monomer residue (Ncorr value) evaluated for poly(methacrylic acid) was largely positive, and with an increase in the content of the DMAPMA residue, the Ncorr value became smaller, and after passing a minimum (which was still slightly positive) at a roughly equivalent molar ratio (P(M47 D53 ); M, methacrylic acid; D, N -[3-(dimethylamino)propyl]methacrylamide), increased again. This is in a significant contrast with the largely positive Ncorr values for the homopolymers of MA and DMAPMA, and other ordinary polyelectrolytes. The small Ncorr value for P(M47 D53 ) was comparable to those for water-soluble nonionic polymers such as poly(ethylene glycol) and zwitterionic polymers such as polycarboxybetaine. These results suggested that the balance of electric charges in polymeric materials is important to be inert to the structure of vicinal water. © 2007 Elsevier Inc. All rights reserved. Keywords: Amphoteric polymer; Hydrogen-bonded network; O–H stretching; Raman spectroscopy; Water structure
1. Introduction Water, which is ubiquitous on Earth, plays an essential role in living bodies. Hydrogen bonding (abbreviated as H-bonding hereafter) between the oxygen atom of the water molecule and hydrogen atom of a neighboring water molecule causes the anomalousness in physical properties of liquid water [1]. Furthermore, a significantly fluctuated H-bond network in the liquid state is induced by a non-linear bent structure of water molecules. Vibrational spectroscopic methods (abbreviated as VSs hereafter) including infrared spectroscopy and Raman spectroscopy have very often been used to analyze the structure of water reflecting orientation of water molecules (i.e., vibration-averaged structure, V-structure) [2–5]. This is because these methods utilize smaller observation times (τ = 10−13 –10−14 s) [6–8] than the relaxation times of the rotational rearrangement (τR ) of wa* Corresponding author.
E-mail address: [email protected] (H. Kitano). 0021-9797/$ – see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2007.05.009
ter molecules in the liquid phase (τ = 10−11 –10−12 s). Furthermore, the vibrational relaxation time of chemical bonds used as a probe in the VSs is smaller than the average lifetime of the H-bond between water molecules (in the order of 10−12 s). Therefore, the VSs can detect the changes in the H-bonded networks between water molecules in addition to the rotational rearrangement of water molecules. Among VSs, Raman spectroscopy is suitable for samples with plenty of water (aqueous solutions), because the intensity of Raman scattering for water is small. Infrared spectroscopy is suitable for a small amount of water sorbed into polymer films due to a large extinction coefficient of O–H stretching vibration bands. On the other hand, other techniques such as differential scanning calorimetry (DSC), X-ray diffraction, dielectric dispersion, and nuclear magnetic resonance (NMR), can detect the state of water based on the diffusion property of the water molecule, but cannot directly detect the changes in the water structure, because these techniques utilize larger observation times (τ ∼ = 101 , 101 , 100 –10−11 , and 10−6 –10−11 s, respectively) compared with the τR or the lifetime of H-bond between water molecules.
462
H. Kitano et al. / Journal of Colloid and Interface Science 313 (2007) 461–468
The interaction between water and the polymer chain very often determines the physical properties of polymer systems in all concentration regions from dilute aqueous solutions to hydrated solids. Therefore, the properties of water in polymer solutions have extensively been analyzed by NMR [9,10], DSC [11,12], sound velocity measurement [13], pulsed-field gradient NMR [14], neutron scattering [15], dielectric measurement [16], etc. As mentioned above, however, it is difficult to reveal the H-bonded network structure of water using these methods. In these fifteen years, we have been interested in analyses of the structure of water in aqueous polymer solutions from a Raman scattering attributable to the O–H stretching vibration band of water (2800–3800 cm−1 ) [17–29]. It was found that, when the molecular weight of the polymers was small enough to avoid a formation of “pseudo-network structure” (i.e., entanglement of the polymer chains), the number of H-bonds disrupted due to the presence of one monomer residue (Ncorr value) of a polyelectrolyte with an ionic group in the side chain (sodium polyacrylate, poly-L-lysine hydrobromide, etc.) was much larger than those for water-soluble nonionic polymers (poly(N -vinyl pyrrolidone) (PVPy) and poly(ethylene glycol) (PEG)) [18,21,22]. Moreover, the Ncorr value for a polymer with an ionizable group in the side chain (poly(acrylic acid) (HPAA), for example) was smaller than those for the polyelectrolytes with fully ionized groups and larger than those for the nonionic polymers [18]. Quite recently, it was found that homo- and random copolymers of zwitterionic phosphobetaine [22,24], sulfobetaine [23], and carboxybetaine vinyl monomers [25] did not disturb the Hbonded network structure of water significantly. Therefore, it is highly probable that zwitterionic polymers in general do not significantly affect the structure of water. There is another kind of ionic polymers whose total charges are neutralized: namely, amphoteric polymer (polyampholyte) composed of both anionic and cationic monomer residues. In a broad sense, amphoteric polymers can also be included into a category of “zwitterionic” polymers [30–33]. More than fifty years ago, the properties of an aqueous solution of amphoteric (zwitterionic) polymers prepared by the radical copolymerization of cationic and anionic vinyl monomers were studied [34,35]. Afterwards, zwitterionic polymers based on carboxy-, sulfo- and phospho-betaine monomer units were also investigated [36–47]. Meanwhile, zwitterionic compounds are characterized by having excellent dermatological properties, and frequently used in shampoos and other cosmetic products [48–51]. Biocompatibility and blood-compatibility (no thrombus formation) of carboxybetaine, phosphobetaine and sulfobetaine copolymers have often been reported, too [25,52–57]. In this work, the polarized Raman scattering of the O–H stretching vibration band of water has been examined in an aqueous solution of amphoteric random copolymer, poly(methacrylic acid-r-N -[3-(dimethylamino)propyl]methacrylamide), poly(MA-r-DMAPMA). As mentioned above, the unique properties of zwitterionic polymers might be strongly related to the interaction of the polymers with water. Therefore, Raman
Scheme 1. Chemical structure of amphoteric copolymer, poly(MA-rDMAPMA).
spectroscopy would be an excellent technique to explore the structure of water in the vicinity of the amphoteric polymers. 2. Materials and methods 2.1. Materials N -[3-(Dimethylamino)propyl]methacrylamide (DMAPMA) was from Aldrich, Milwaukee, WI. Methacrylic acid (MA) from Wako Pure Chemicals, Osaka, Japan, was distilled under the reduced pressure. Other reagents were commercially available. A Milli-Q grade water was used for the preparation of the sample solutions. 2.2. Preparation of copolymers (Scheme 1) DMAPMA (9.0 mL) and MA (4.2 mL) were copolymerized in water (100 mL) at 65 ◦ C for 24 h using 2,2 -azobis(2amidinopropane) dihydrochloride (V-50, Wako Pure Chemicals, 271 mg) and 2-mercaptoethanol (0.7 mL) as initiator and chain transfer reagent, respectively. After evaporation of the solvent, the obtained polymer was dissolved in water again, and purified by ultrafiltration using Amicon membranes (YM-3 (exclusion limit, 3 × 103 ) and YM-10 (104 )) to obtain a fraction between 3 × 103 and 104 daltons. The purified polymer was finally lyophilized (poly(MA-r-DMAPMA), 1.0 g). The weight-average and number-average molecular weights (Mw and Mn ) of the copolymer were determined to be 5.1 × 103 and 3.5 × 103 (Mw /Mn = 1.46), respectively, by gel permeation chromatography (GPC) (Waters HPLC system; column, Wako Gel G-40, Wako Pure Chemicals; mobile phase, 0.1 M LiCl in H2 O–N ,N -dimethylformamide (DMF) (4:1)) using pullulan standards (Showa Denko, Tokyo, Japan). Using a similar procedure, copolymers with various compositions were obtained. The composition of the copolymers was determined by 1 H NMR. Poly(MA-r-DMAPMA) are referred as P(Mx Dy ) (x and y are the mole fractions of MA and DMAPMA, respectively) hereafter. Results of the polymerization are summarized in Table 1. 2.3. Raman spectroscopy In the Raman spectra between 2500 and 4000 cm−1 for aqueous solution of various amphoteric polymers, the component of the O–H stretching band of water centered at 3250 cm−1
H. Kitano et al. / Journal of Colloid and Interface Science 313 (2007) 461–468
463
Table 1 Characteristics of various poly(MA-r-DMAPMA)s Code P(M100 D0 ) P(M67 D33 ) P(M47 D53 ) P(MnD89 ) P(M0 D100 )
Composition ratio (%) MA
DMAPMA
1/[η] (g/dL)/pX a
100 67 47 11 0
0 33 53 89 100
30.1/0.059 24.4/0.037 14.7/0.020 12.8/0.014 17.5/0.018
Mw (kD)
Mw / Mn
pHb
9.6 2.1 5.1 3.8 7.0
1.25 1.45 1.46 1.42 1.42
3.20 4.94 7.68 9.96 10.80
a The p values correspond to the 1/[η] values. X b pH value at p = 0.01. X
Fig. 2. Reduced viscosity vs concentration of amphoteric copolymers at 25 ◦ C in water: (") P(M100 D0 ), (2) P(M67 D33 ), (P) P(M47 D53 ), (F) P(M11 D89 ), (a) P(M0 D100 ); (· · ·) a guide to the eye.
polyampholytes at 25 ◦ C was determined using an Ubbelohde dilution type viscometer (Type 0B; Kusano, Tokyo, Japan). The time needed for sample solutions to pass through the capillary of the viscometer was determined within the uncertainties of ±0.1 s. Fig. 1. Raman spectra of O–H stretching region: (a) I (solid) and I⊥ (broken) spectra of pure water at 25 ◦ C. I⊥ /ρO–H spectrum is also shown by (· · ·); (b) the collective band of pure water.
was highly polarized, and diminished in the spectra at a perpendicular position (Fig. 1) [58–64]. The polarized O–H stretching band of water (“collective band”) is ascribed to the H2 O molecule executing ν1 vibrations all in phase with each other but the vibrational amplitude varying from molecule to molecule in strongly H-bonded water clusters [65–68]. The intensity of the collective band (Ic ) was separated from the spectra using equation [58–63] (Fig. 1), Ic = I − I⊥ /ρO–H ,
(1)
where I⊥ and I are the intensities of the spectra observed with the polarizer oriented perpendicular and parallel to the incident laser beam, respectively. The depolarization ratio, ρO–H , is an indicator of symmetry of the vibration mode, and was adjusted to minimize the contribution of large wavenumber region to the Ic value. The area of Ic was normalized by [58–63], I (w) dw, C = Ic (w) dw (2) where w is the Raman shift in cm−1 . 2.4. Viscosity measurements The reduced viscosity (ηsp /φ, ηsp = (η − η0 )/η0 , φ: concentration of polymer in g/dL) for aqueous solution of various
3. Results and discussion 3.1. Viscosity behavior of amphoteric polymer solutions The viscosity behaviors of various kinds of copolymers in water are shown in Fig. 2. The reduced viscosity (ηsp /φ) for an aqueous solution of polyMA (P(M100 D0 )) monotonously decreased with a decrease in the polymer concentration. This is in a good contrast with the increase of ηsp /φ with the decrease in the concentration of poly(MA-r-DMAPMA)s (P(M67 D33 ), P(M47 D53 ) and P(M11 D89 )) and polyDMAPMA (P(M0 D100 )), which is a typical behavior of the aqueous solution of ordinary polyelectrolytes: A decrease in the concentration of polyelectrolyte results in an increased Debye length, and the resulting increase in the electrostatic repulsion between the charged groups within the polymer chain induces expansion of the polymer chain [69,70]. However, it was reported that polyelectrolyte molecules such as sodium poly(styrene sulfonate) do not completely stretch out even in an infinitely diluted solution [71,72]. When the DMAPMA content in the copolymer was not in excess to that of MA (P(M67 D33 )), a slight increase in ηsp /φ upon dilution was observed in a low concentration region. With a further increase in the content of DMAPMA, the copolymer (P(M47 D53 )) indicated a less significant polyelectrolyte behavior, while a further increase in the content of DMAPMA (P(M11 D89 )) and polyDMAPMA (P(M0 D100 )) resulted in a much clearer behavior of polyelectrolyte. This is because the methacrylic acid residues in the polymers are weakly acidic
464
H. Kitano et al. / Journal of Colloid and Interface Science 313 (2007) 461–468
and not deprotonated to a large extent, while the DMAPMA residues are mostly protonated (pKa : polyMA, 6.62 at the degree of neutralization of 0.50 [73]; MA monomer, 4.65 [34]; polyDMAPMA, 8.00). In the figure, the ηsp /φ value at a relatively high concentration changed with the polymer composition. At this moment, there is no definitive explanation for the difference in the ηsp /φ value, because the composition and the degree of polymerization (DP) of polymers examined were different from each other. However, this point is not important in the present study because the aim is to clarify the influence of net-charge of the polymer. Assuming that the dissociation behavior of carboxylate and ammonium groups in polymers is indicated by pH = n log(α/1 − α) + pKa ,
(3)
the degree of dissociation (α) can be calculated [74,75]. The n value for a small molecular weight acid is unity, whereas that for polyelectrolytes is larger than unity due to the electrostatic effect. As for P(M47 D53 ), the degrees of dissociation of the MA and DMAPMA residues are calculated to be 77 and 41%, respectively, when the n value is assumed to be 2 [74,75]. Consequently, the ratio of anionic and cationic groups in the copolymer is calculated to be 100:88, suggesting that the electro-neutralization along the polymer chain is roughly realized. In the cases of P(M67 D33 ) and P(M11 D89 ), on the contrary, the ratio of charged groups leaned towards cationic (26:100) and anionic groups (100:79), respectively. These values are consistent with the viscosity behavior of the homo- and copolymers examined here. That is, in contrast with the noticeable increase in viscosity with the decrease in concentration for the copolymers P(M67 D33 ) and P(M11 D89 ), that for P(M47 D53 ) was relatively less significant due to the approximate electroneutralization of the polymer. It has been frequently pointed out that charge-imbalanced polyampholytes undergo a transition from a “globular coil” in a charge-balanced state to an “extended chain” with a large excess of charge on the polymer via a “beaded-necklace” morphology [76–79]. Such a tendency was consistent with a zwitterionic polymer, sulfobetaine polymer: At a large degree of quaternization, sulfobetaine polymer showed a linear relationship between the ηsp /φ value and the φ value more clearly [23]. However, the sulfobetaine polymer almost completely quaternized showed a very poor solubility (anti-polyelectrolyte effect) [23], which is in contrast with the large solubility of P(M47 D53 ). In the presence of comparable amount of DMAPMA residues, the MA residue is significantly deprotonated, which induces electro-neutralization of the side chain to form an ion pair, while no effective stacking between the cationic and anionic groups of neighboring monomer residues would appear, probably because the distance between the carboxylate and tertiary ammonium groups might be long enough to avoid a disadvantageous strong ionic association. As for ordinary zwitterionic polymers, a disadvantageous inter-tether ionic association could be weakened by the introduction of hydrophilic comonomer residues between the zwitterionic monomer residues [41,42].
3.2. Structure of water in aqueous polymer solutions The O–H stretching Raman band of liquid water gave a broad band (Fig. 1a) composed of several overlapping components which were attributed to the unperturbed O–H stretching band by the intra- and inter-molecular vibrational coupling of the O–H oscillators [3]. The relative intensities of the collective bands (C, Fig. 1b) are reduced by the decoupling of the O–H oscillators: (1) when the hydrogen bond between the coupled O–H oscillators is broken by translational or rotational rearrangement of the water molecules, and (2) when the stretching frequency of an O–H oscillator is quite different from that of the O–H oscillator which is combined by a H-bond with the former one [58]. The defect probability, Pd , that an O–H oscillator is excluded from the H-bonding network of water molecules because of an unfavorable position or orientation, is defined as Pd =
C w − Cx , Cw
(4)
where Cw and Cx are the intensities of the collective band of pure water and aqueous solution at a certain temperature, respectively. The Cx values of aqueous solutions of simple electrolytes [58] and polar substances were smaller than that of pure water at the same temperature. This is caused, as mentioned above, by the exclusion of the O–H oscillator of hydrating water (HOH · · ·solute and/or H2 O · · ·solute) from the collective O–H oscillators of the water cluster. On the contrary, those of hydrophobic substances such as t -butanol [62] (mol fraction (pX ), 0.04) and tetra-n-butyl ammonium hydroxide [63] (pX = 0.013) were higher than that of pure water, due to an enhancement of the H-bond between water molecules in hydrophobic hydration shells around the hydrocarbon moiety [80]. These results show that the vibrational band attributed to the O–H oscillators of water cluster can be distinguished from that of the water hydrated to solutes by using the polarized Raman spectroscopy, and that this method has an ability to detect the changes in the H-bonded network structure of water even in dilute–semi-dilute aqueous solutions (pX = 0.01–0.05 (ca. 0.55–2.8 mol/L)). In a dilute solution, polymer coils are separated from each other and the intermolecular interaction between the coils is sufficiently weak, and therefore, the solution can be described as a “non-ideal gas.” When the polymer concentration (φ) approaches a critical value (φ ∗ ), the polymer coils begin to overlap and make a pseudo-network. In the concentration region (φ > φ ∗ ), the polymer solutions are called “semi-dilute” solutions [81]. Based on a previously suggested criterion between the low and high-concentration regimes for random coils in a good solvent, 1 < φ[η] < 10 [82], the value of φ ∗ can be roughly related to the intrinsic viscosity [η] as φ∗ ∼ = 1/[η].
(5)
From the value of [η] for aqueous solutions of amphoteric polymers at 25 ◦ C (Table 1), it was concluded that all the polymers examined here give dilute solutions at pX = 0.01.
H. Kitano et al. / Journal of Colloid and Interface Science 313 (2007) 461–468
3.3. Estimation of Ncorr values for polymers
Table 2 The Ncorr values for various polymersa
To elucidate the effects of the properties of the monomer residue of various polymers, the number of H-bond defects introduced into the H-bonded network structure of water per one monomer unit of a polymer, N , was calculated from the defect probability (Pd ) using equation, N = Pd /fX ,
465
(6)
where fX is the number of monomer units per one O–H oscillator in a water molecule [18,20–22,83]. It should be mentioned here that the liquid water has an intrinsic defect in the H-bonded network structure. At the estimation of Pd values, this factor is not necessary to be considered, because the Pd values are calculated by the ratio of (Cw − Cx ) and Cw values. However, the number of O–H oscillators H-bonded in pure water at certain temperature should be smaller than the total number of O–H oscillators. Therefore, the N value was corrected using the C values for perfect ice and pure water (Cice and Cw , respectively) as Ncorr = N × (Cw /Cice ) [58]. It should be mentioned here that the Ncorr value represents the degree of perturbation on the H-bonded network structure of water. Since the derivation of Ncorr value includes several assumptions, it does not precisely indicate the actual number of H-bonds that each group of polymers perturb. Similar to small molecular solutes mentioned in the previous section, ionized and polar groups in water-soluble polymer chains and counter ions may disturb the H-bond between the water molecules and raise the Ncorr value, whereas hydrophobic moieties such as hydrocarbon chains may enhance the H-bond and reduce the Ncorr value. The exposure area of the chemical groups to water is also important to the structure of water around them. The effect of various water-soluble polymers on the C value at the constant molar fraction of monomer units (pX = 0.01) was examined, and the Ncorr values obtained for the polymers, in which the H-bond defects are localized in the hydration shell of polymer chains and their counter ions, were determined and are shown in Table 2. The Ncorr values of ordinary polyelectrolytes with one kind of ionic side group (sodium polyacrylate (NaPAA) [18,22], sodium poly(ethylene sulfonate) (NaPES), poly-L-lysine HBr salt (PLL·HBr) [22] and poly[(3-(methacryloyl)amino)propyltrimethylammonium chloride] (polyMAPTAC)) [26] are much larger than those of water-soluble nonionic polymers (PEG and PVPy) [18,22], indicating that ionic groups and the counterions of polyelectrolytes strongly disturb the structure of water in their hydration shells (electrostatic hydration). The very small Ncorr values for nonionic water-soluble polymers (PEG and PVPy) coincide with the previous report that oligo(ethylene glycol)s (degree of polymerization, 4–7) dissolves in water without much perturbation to the water structure [84]. 3.4. The Ncorr values for amphoteric copolymers Table 2 and Fig. 3 show that the amphoteric copolymer, P(M47 D53 ), had relatively the smallest positive Ncorr value,
Polymer
Ncorr value
Reference
NaPES PLL·HBr PolyMAPTAC NaPAAb HPAAc NaPolyMA PEG PVPy PMPC PSBB PCMB P(M100 D0 ) P(M68 D32 ) P(M47 D53 )
5.1 5.5 9.4 5.9 2.3 9.0 0.7 0.6 −0.7 0.7 −0.27 6.0 4.3 0.7d 5.4e 3.6f 4.6 5.1
[22] [22] [26] [18] [18] [26] [22] [22] [22] [23] [25] [26] This work This work This work This work This work This work
P(Mn D89 ) P(M0 D100 )
a In H O at 25 ◦ C and p = 0.01 unless mentioned. The uncertainties of X 2 Ncorr values are ±1 at the most. b M = 6.5 kD at p = 0.05. w X c HPAA (M = 5 kD) at p = 0.05. w X d pH 7.68. e After the addition of 0.1 M HCl (pH 4.83). f After the addition of 0.1 M NaOH (pH 9.35).
Fig. 3. The effect of content of MA and DMAPMA residues in amphoteric copolymers on the Ncorr value at 25 ◦ C.
and, with the increase or decrease in the composition ratio of MA, the Ncorr value increased. It is worth noting that the Ncorr values for the copolymers were smaller than that of polyMA (P(M100 D0 )) and polyDMAPMA (P(M0 D100 )). The MA and DMAPMA residues have ionizable groups (–COOH and –N(CH3 )2 , respectively), and they are partly ionized under neutral conditions as mentioned in Section 3.1. Therefore, we had expected that the water structure in aqueous solutions of copolymers, poly(MA-r-DMAPMA), would be more significantly disturbed in comparison with that in polyMA and polyDMAPMA solutions, because the amount of ionic species in the copolymers is much larger than that in polyMA and polyDMAPMA. The experimental result was opposite to the expectation.
466
H. Kitano et al. / Journal of Colloid and Interface Science 313 (2007) 461–468
The large Ncorr values for polyDMAPMA and polyMA were not contradictory to the general knowledge and our previous reports [18–26]. However the disturbing effects of the amphoteric copolymers on the water structure were much smaller than that of the ordinary polyelectrolytes. This is in accordance with the previous result that the zwitterionic polymer, poly[1-carboxyN ,N -dimethyl-N -(2 -methacryloyloxyethyl)methanaminium inner salt] (PCMB, carboxybetaine polymer), did not disturb the structure of water (Ncorr = −0.27 for PCMB (Mw , 11.4 kD)) [25]. The slightly negative Ncorr value for PCMB is probably due to the small error associated with separation of the collective band from the O–H stretching band, and therefore, they can be regarded as zero approximately. A similar tendency was observed for poly[4-sulfo-N ,N -dimethyl-N -(3 methacryloylaminopropyl)butanaminium inner salt] (PSBB, sulfobetaine polymer) [23], and poly(2-methacryloyloxyethyl phosphorylcholine) (PMPC, phosphobetaine polymer), too [22]. Using 1 H NMR, El Seoud and coworkers showed that a sulfobetaine surfactant, 3-sulfo-N ,N -dimethyl-N -(n-dodecyl)propanaminium inner salt, does not affect the state of interfacial water, due to strong intermolecular interactions between the head ions of neighboring molecules in the micelles [85]. Previously, we found that the Ncorr values for the zwitterionic surfactants were close to zero, too [86]. On the other hand, Halle and coworkers reported that the dynamic sense of water in the vicinity of polymers (adsorbent) was strongly perturbed [87,88], which is in conflict with the report by El Seoud and coworkers and by us. This confliction seems to be caused by the difference in the ordinary (poly)electrolyte and the zwitterionictype one. Actually, El Seoud et al. also pointed out that, with an increase in the distance between the anionic and cationic moieties in zwitterionic molecule, the structure of neighboring water became similar to that around the ordinary electrolytes: in other words, the two kinds of claims by El Seoud et al. and Halle et al. were the same in case of the zwitterionic molecule whose distance between anionic and cationic moieties was long (number of methylene groups 3) [85]. Furthermore, the C value of aqueous amino acid solutions under neutral conditions were found to be slightly smaller than that of pure water, and almost constant in spite of the difference in hydrophobicity of the side chain [89]. In other words, when both α-amino and α-carboxyl groups of amino acids are ionized (zwitterionic), the Ncorr value for the amino acids is slightly positive and almost independent of the hydrophobicity of the side chain. The small Ncorr values evaluated for the zwitterionic compounds with a small molecular weight and those for the amphoteric copolymers might be due to the same reason. As discussed in Section 3.1, when the n value in Eq. (3) is assumed to be 2, the ratio of anionic and cationic groups in P(M47 D53 ) is calculated to be 100:88, suggesting that the electro-neutralization along the polymer chain is roughly realized, whereas, in the cases of P(M67 D33 ) and P(M11 D89 ), the ratio of charged groups leaned towards cationic (26:100) and anionic groups (100:79), respectively. The deprotonated carboxylate group and protonated tertiary amino group, which are in close proximity to each
other in P(M47 D53 ), might counteract the electrostatic hydration: It has generally been suggested that the partially negative oxygen atom of the water molecule is toward the positively charged ions in solution, whereas the negatively charged ions attract the partially positive hydrogen atoms of water, creating their hydration shells [90]. When the cation and anion are in close proximity, the orientation of the hydrating water might be largely disturbed resulting in the partial collapse of electrostriction. Such a tendency would be more significant when the number of cationic and anionic groups are becoming close each other, because, in the linear polymer system, the equalization of oppositely charged groups statistically means relatively the closest proximity between the oppositely charged groups. In the present polymer system, such an equalization of the number of oppositely charged ionic groups could be realized by the deprotonation of MA residues by the neighboring DMAPMA residues as discussed above (Fig. 3 and Table 2). By the addition of 0.1 M of HCl and NaOH to the solution of P(M47 D53 ), the electro-equalization within the polymer molecules was collapsed (ratio of anionic and cationic groups calculated by Eq. (3) (n = 2): 10:100 and 100:14, respectively), and the Ncorr values for P(M47 D53 ) became larger than that dissolved in pure water (Table 2), supporting the hypothesis mentioned above. The hydrocarbon main chain in the amphoteric polymers might be hidden behind the bulky cationic and anionic groups from water to minimize the free energy of the system, and consequently, the effect of the main chain on the Ncorr value for the polymers might not appear clearly. Recently we have reported that the Ncorr values for α,ω-amino acids increased with an increase in the distance between the amino and carboxyl groups [91], which is in accordance with the 1 H NMR study by El Seoud and coworkers on the zwitterionic surfactants as discussed above. Høiland reported that, when hydration sheaths for ionizable groups of a dicarboxylic acid are in close proximity and overlapped, an additivity rule for the hydration of each component is not applicable [92]. These results support the tendency observed in this work. Previously, it was reported that self-assembled monolayers (SAMs) of zwitterionic sulfides or 1:1 mixture of cationic and anionic sulfides resist against the non-specific adsorption of various proteins from a buffer solution [93]. We found that a brush of zwitterionic telomer is resistant against the non-specific adsorption of proteins, too [94]. It has very often been pointed out that the absence of non-specific adsorption of serum proteins has been considered as an essential factor for usability of polymeric materials in medical fields [52,53]. Therefore, there is a possibility that the mildness of the amphoteric compounds to the structure of water at their surface plays an important role in the resistance against non-specific adsorption of proteins, resulting in the appearance of blood- and bio-compatibilities. Previously, it was found that linear and branched polyethyleneimines modified with acrylic acid showed no acute toxicity in mice up to dosage of 1 g/kg [95]. The zwitterionic form was attributed to the non-toxic nature of the amino acid type polymers, supporting the tendency observed in this work.
H. Kitano et al. / Journal of Colloid and Interface Science 313 (2007) 461–468
4. Conclusion In contrast with ordinary polyelectrolytes carrying one kind of ionized group, the amphoteric copolymer with comparable contents of cationic and anionic monomer residues had a very small disturbing effect on the structure of the hydrogen bonding network of water molecules in dilute solutions due to the inter-tether proximity between the oppositely charged groups (counteraction of the electrostatic hydration effects). The weak disturbing effect of amphoteric polymers on the water structure could be attributed to the intrinsic property of the polymers themselves, and not to the polymer–polymer coordination. Taking account of biocompatibility of zwitterionic polymers, a small disturbing effect of charge-balanced polymer materials on the structure of vicinal water (in other words, resemblance of the water structure in the vicinity of the polymers to that of free water) seemed to be important to afford the polymers biocompatibility. Acknowledgments This work was supported by Grants-in-Aid (16205015, 17750105 and 19350055) from the Japan Society for the Promotion of Science. The authors are indebted to the Toyama Prefecture New Century Foundation, Toyama, Japan, for its financial support. M.G.-I. wishes to thank the Sekisui Science Foundation, Osaka, Japan, for its financial support. The authors are grateful to Mr. K. Takaha of this laboratory for his technical assistance. References [1] S. Scheiner, Hydrogen Bonding: A Theoretical Perspective, Oxford Univ. Press, Oxford, 1997. [2] D. Eisenberg, W. Kauzmann, The Structure and Properties of Water, Clarendon Press, London, 1969. [3] G.E. Walrafen, in: F. Franks (Ed.), Water: A Comprehensive Treatise, vol. 1, Plenum Press, New York, 1972, chap. 2. [4] G.E. Walrafen, in: W.A.P. Luck (Ed.), Structure of Water and Aqueous Solutions, Verlag Chemie, Weinheim, 1974. [5] E. Westhof (Ed.), Water and Biological Macromolecules, McMillan Press, Houndmills, 1993. [6] C.H. Collie, J.B. Hasted, D.M. Riston, Proc. Phys. Soc. 60 (1948) 145. [7] E.H. Grant, T.J. Buchanan, H.F. Cook, J. Chem. Phys. 26 (1957) 156. [8] S.K. Garg, C.P. Smyth, J. Chem. Phys. 43 (1965) 2959. [9] S. Katayama, S. Fujiwara, J. Am. Chem. Soc. 101 (1979) 4485. [10] F.X. Quinn, E. Kampf, G. Smyth, V.J. McBriety, Macromolecules 21 (1988) 3191. [11] N. Murase, K. Gonda, T. Watanabe, J. Phys. Chem. 90 (1986) 5420. [12] H. Ohno, M. Shibayama, E. Tsuchida, Makromol. Chem. 184 (1983) 1017. [13] K. Gekko, H. Noguchi, Biopolymers 10 (1994) 1513. [14] A.C. Rarnes, T.W.N. Bieze, J.E. Enderby, J.C. Leyte, J. Phys. Chem. 98 (1994) 11527. [15] T.W.N. Bieze, A.C. Barnes, C.J.M. Huige, J.C. Leyte, J. Phys. Chem. 98 (1994) 6568. [16] N. Shinyashiki, Y. Matsumura, N. Miura, S. Yagihara, S. Mashimo, J. Phys. Chem. 98 (1994) 13169. [17] T. Terada, Y. Maeda, H. Kitano, J. Phys. Chem. 97 (1993) 3619. [18] Y. Maeda, N. Tsukida, H. Kitano, T. Terada, J. Yamanaka, J. Phys. Chem. 97 (1993) 13903.
467
[19] T. Terada, T. Inaba, H. Kitano, Y. Maeda, N. Tsukida, Macromol. Chem. Phys. 195 (1994) 3261. [20] N. Tsukida, Y. Maeda, H. Kitano, Macromol. Chem. Phys. 197 (1996) 1681. [21] N. Tsukida, H. Muranaka, M. Ide, Y. Maeda, H. Kitano, J. Phys. Chem. B 101 (1997) 6676. [22] H. Kitano, K. Sudo, K. Ichikawa, M. Ide, K. Ishihara, J. Phys. Chem. B 104 (2000) 11425. [23] H. Kitano, M. Imai, K. Sudo, M. Ide, J. Phys. Chem. B 106 (2002) 11391. [24] H. Kitano, M. Imai, T. Mori, M. Gemmei-Ide, Y. Yokoyama, K. Ishihara, Langmuir 19 (2003) 10260. [25] H. Kitano, S. Tada, T. Mori, K. Takaha, M. Gememei-Ide, M. Tanaka, M. Fukuda, Y. Yokoyama, Langmuir 21 (2005) 11932. [26] H. Kitano, K. Takaha, M. Gemmei-Ide, Phys. Chem. Chem. Phys. 8 (2006) 1178. [27] Y. Maeda, H. Kitano, Spectrochim. Acta A 51 (1995) 2433. [28] Y. Maeda, H. Kitano, Trends Phys. Chem. 6 (1997) 269. [29] Y. Maeda, M. Ide, H. Kitano, J. Mol. Liq. 80 (1999) 149. [30] H. Morawetz, Macromolecules in Solution, Interscience Publishers, New York, 1965. [31] S.E. Kudaibergenov, Polyampholytes: Synthesis, Characterization and Application, Kluwer Academic/Plenum Publishers, New York, 2002. [32] A.B. Lowe, C.L. McCormick, Chem. Rev. 102 (2002) 4177. [33] S. Kudaibergenov, W. Jaeger, A. Laschewsky, Adv. Polym. Sci. 201 (2006) 157. [34] T. Alfrey, H. Morawetz, J. Am. Chem. Soc. 74 (1952) 436. [35] G. Ehrlich, P. Doty, J. Am. Chem. Soc. 76 (1954) 3764. [36] H. Ladenheim, H. Morawetz, J. Polym. Sci. 26 (1957) 251. [37] K. Rosenheck, A. Katchalsky, J. Polym. Sci. 32 (1958) 511. [38] R. Hart, D. Timmerman, J. Polym. Sci. 28 (1958) 638. [39] W.-F. Lee, C.-C. Tsai, Polymer 35 (1994) 2210. [40] W.-F. Lee, Y.-M. Tu, J. Appl. Polym. Sci. 72 (1999) 1221. [41] E.E. Kathmann, L.A. White, C.L. McCormick, Polymer 38 (1997) 871. [42] R.S. Armentrout, C.L. McCormick, Macromolecules 33 (2000) 2944. [43] S.A. Ali, A. Rasheed, M.I.M. Wazeer, Polymer 40 (1999) 2439. [44] I.V. Berlinova, I.V. Dimitrov, R.G. Kalinova, N.G. Vladimirov, Polymer 41 (1999) 831. [45] S.F. Lascelles, F. Malet, R. Mayada, N.C. Billingham, S.P. Armes, Macromolecules 32 (1999) 2462. [46] C. Viklund, K. Irgum, Macromolecules 33 (2000) 2539. [47] J. Virtanen, M. Arotçaréna, B. Heise, S. Ishaya, A. Laschewsky, H. Tenhu, Langmuir 18 (2002) 5360. [48] B. Jönsson, B. Lindman, K. Holmberg, B. Kronberg, Surfactants and Polymers in Aqueous Solution, Wiley, Chichester, 1998, pp. 21–23. [49] D. Saint-Leger, G. Hussler, G. Kaba, Eur. Pat. Appl. EP 728473. [50] T. Murayama, Japan Patent 2000-219670A. [51] K. Takano, K. Ide, Japan Patent 11012596A. [52] T. Nakaya, Y.J. Li, Prog. Polym. Sci. 24 (1999) 143. [53] A.L. Lewis, Colloids Surf. B Biointerfaces 18 (2000) 261. [54] Y.L. Yuan, F. Ai, J. Zhang, X.B. Zang, J. Shen, S.C. Lin, J. Biomater. Sci. Polym. Ed. 13 (2002) 1081. [55] Y. Yuan, X. Zang, F. Ai, J. Zhou, J. Shen, S. Lin, Polym. Int. 53 (2004) 121. [56] A.B. Lowe, M. Vamvakaki, W.A. Wasall, L. Wong, N.C. Billingham, S.P. Armes, A.W. Lloyd, J. Biomed. Mater. Res. 52 (2000) 88. [57] S.L. West, J.P. Salvage, E.J. Lobb, S.P. Armes, N.C. Billingham, A.L. Lewis, J.G.W. Hanlon, A.W. Llyod, Biomaterials 25 (2004) 1195. [58] J.L. Green, A.R. Lacey, M.G. Sceats, J. Phys. Chem. 90 (1986) 3958. [59] J.L. Green, A.R. Lacey, M.G. Sceats, S.J. Henderson, R.J. Speedy, J. Phys. Chem. 91 (1987) 1684. [60] J.L. Green, A.R. Lacey, M.G. Sceats, J. Chem. Phys. 86 (1987) 1841. [61] J.L. Green, A.R. Lacey, M.G. Sceats, Chem. Phys. Lett. 134 (1987) 385. [62] J.L. Green, A.R. Lacey, M.G. Sceats, Chem. Phys. Lett. 137 (1987) 527. [63] J.L. Green, M.G. Sceats, A.R. Lacey, J. Chem. Phys. 87 (1987) 3603. [64] N. Karger, H.-G. Lüdemann, M.G. Sceats, Ber. Bunsenges. Phys. Chem. 99 (1995) 1104. [65] S.A. Rice, M.G. Sceats, J. Phys. Chem. 85 (1981) 1108. [66] M.G. Sceats, S.A. Rice, in: F. Franks (Ed.), Water—A Comprehensive Treatise, vol. 7, Plenum Press, New York, 1983, chap. 2.
468
[67] [68] [69] [70]
[71] [72] [73] [74] [75] [76] [77] [78] [79] [80] [81] [82]
H. Kitano et al. / Journal of Colloid and Interface Science 313 (2007) 461–468
F.H. Stillinger, Science 209 (1980) 451. D.E. Hare, C.M. Sorensen, J. Chem. Phys. 93 (1990) 25. R.M. Fuoss, U.P. Strauss, J. Polym. Sci. 246 (1948) 602. H. Dautzenberg, W. Jaeger, J. Kotz, B. Philipp, Ch. Seidel, D. Stcherbina, Polyelectrolytes: Formation, Characterization and Application, Hanser, Munich, 1994. H. Vink, Polymer 33 (1992) 3771. J. Yamanaka, H. Matsuoka, H. Kitano, M. Hasegawa, N. Ise, J. Am. Chem. Soc. 112 (1990) 587. E.V. Anufrieva, T.M. Birshtein, T.N. Nekrasova, O.B. Ptitsyn, T.V. Shveleva, J. Polym. Sci. C 16 (1968) 3519. A. Katchalsky, P. Spitnik, J. Polym. Sci. 2 (1947) 432. A. Katchalsky, J. Polym. Sci. 13 (1957) 69. Y. Kantor, M. Kardar, Europhys. Lett. 27 (1994) 643. Y. Kantor, M. Kardar, Phys. Rev. E 51 (1995) 1299. Y. Kantor, M. Kardar, Phys. Rev. E 52 (1995) 835. D.B. Thomas, Y.A. Vasilieva, R.S. Armentrout, C.L. McCormick, Macromolecules 36 (2003) 9710. A. Kauzmann, Adv. Protein Chem. 14 (1959) 39. P.-G. de Gennes, Scaling Concepts in Polymer Physics, Cornell Univ. Press, Ithaca, 1979. W.W. Graessley, Adv. Polym. Sci. 16 (1974) 1.
[83] N.J. Tao, S.M. Lindsay, A. Rupprecht, Biopolymers 28 (1989) 1019. [84] T. Sato, H. Niwa, A. Chiba, J. Chem. Phys. 108 (1998) 4138. [85] O.A. El Seoud, R.C. Bazito, G.K. Barlow, Prog. Colloid Polym. Sci. 111 (1998) 151. [86] H. Kitano, M. Imai, M. Gemmei-Ide, K. Takaha, J. Colloid Interface Sci. 269 (2004) 459. [87] K. Fontell, B. Halle, Liq. Cryst. 16 (1994) 235. [88] S. Wiesner, E. Kurian, F.G. Prendergast, B. Halle, J. Mol. Biol. 286 (1999) 233. [89] M. Ide, Y. Maeda, H. Kitano, J. Phys. Chem. B 101 (1997) 7022. [90] R.H. Garrett, C.M. Grisham, Biochemistry, Saunders College Publishing, Fort Worth, 1995, p. 36. [91] H. Kitano, K. Takaha, M. Gemmei-Ide, J. Colloid Interface Sci. 283 (2005) 452. [92] H. Høiland, J. Chem. Soc. Faraday I 71 (1975) 797. [93] R.E. Holmlin, X. Chen, R.G. Chapman, S. Takayama, G.M. Whitesides, Langmuir 17 (2001) 2841. [94] H. Kitano, A. Kawasaki, H. Kawasaki, S. Morokoshi, J. Colloid Interface Sci. 282 (2005) 340. [95] S. Kobayashi, L. Gross, G. Muacevic, H. Ringsdorf, Makromol. Chem. 184 (1983) 793.
Structure of water in the vicinity of amphoteric polymers as revealed by Raman spectroscopy Hiromi Kitano ∗ , Kyoko Nagaoka, Susumu Tada, Makoto Gemmei-Ide Department of Applied Chemistry, Graduate School of Science and Engineering, University of Toyama, Toyama 930-8555, Japan Received 27 February 2007; accepted 2 May 2007 Available online 22 May 2007
Abstract The structure and hydrogen bonding of water in an aqueous solution of amphoteric copolymers (poly(MA-r-DMAPMA), 3 × 103 < Mw < 104 ) composed of various ratios of methacrylic acid (MA) and N -[3-(dimethylamino)propyl]methacrylamide (DMAPMA) were analyzed using the band shapes of the O–H stretching in the polarized Raman spectra. The number of hydrogen bonds disrupted due to the presence of one monomer residue (Ncorr value) evaluated for poly(methacrylic acid) was largely positive, and with an increase in the content of the DMAPMA residue, the Ncorr value became smaller, and after passing a minimum (which was still slightly positive) at a roughly equivalent molar ratio (P(M47 D53 ); M, methacrylic acid; D, N -[3-(dimethylamino)propyl]methacrylamide), increased again. This is in a significant contrast with the largely positive Ncorr values for the homopolymers of MA and DMAPMA, and other ordinary polyelectrolytes. The small Ncorr value for P(M47 D53 ) was comparable to those for water-soluble nonionic polymers such as poly(ethylene glycol) and zwitterionic polymers such as polycarboxybetaine. These results suggested that the balance of electric charges in polymeric materials is important to be inert to the structure of vicinal water. © 2007 Elsevier Inc. All rights reserved. Keywords: Amphoteric polymer; Hydrogen-bonded network; O–H stretching; Raman spectroscopy; Water structure
1. Introduction Water, which is ubiquitous on Earth, plays an essential role in living bodies. Hydrogen bonding (abbreviated as H-bonding hereafter) between the oxygen atom of the water molecule and hydrogen atom of a neighboring water molecule causes the anomalousness in physical properties of liquid water [1]. Furthermore, a significantly fluctuated H-bond network in the liquid state is induced by a non-linear bent structure of water molecules. Vibrational spectroscopic methods (abbreviated as VSs hereafter) including infrared spectroscopy and Raman spectroscopy have very often been used to analyze the structure of water reflecting orientation of water molecules (i.e., vibration-averaged structure, V-structure) [2–5]. This is because these methods utilize smaller observation times (τ = 10−13 –10−14 s) [6–8] than the relaxation times of the rotational rearrangement (τR ) of wa* Corresponding author.
E-mail address: [email protected] (H. Kitano). 0021-9797/$ – see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2007.05.009
ter molecules in the liquid phase (τ = 10−11 –10−12 s). Furthermore, the vibrational relaxation time of chemical bonds used as a probe in the VSs is smaller than the average lifetime of the H-bond between water molecules (in the order of 10−12 s). Therefore, the VSs can detect the changes in the H-bonded networks between water molecules in addition to the rotational rearrangement of water molecules. Among VSs, Raman spectroscopy is suitable for samples with plenty of water (aqueous solutions), because the intensity of Raman scattering for water is small. Infrared spectroscopy is suitable for a small amount of water sorbed into polymer films due to a large extinction coefficient of O–H stretching vibration bands. On the other hand, other techniques such as differential scanning calorimetry (DSC), X-ray diffraction, dielectric dispersion, and nuclear magnetic resonance (NMR), can detect the state of water based on the diffusion property of the water molecule, but cannot directly detect the changes in the water structure, because these techniques utilize larger observation times (τ ∼ = 101 , 101 , 100 –10−11 , and 10−6 –10−11 s, respectively) compared with the τR or the lifetime of H-bond between water molecules.
462
H. Kitano et al. / Journal of Colloid and Interface Science 313 (2007) 461–468
The interaction between water and the polymer chain very often determines the physical properties of polymer systems in all concentration regions from dilute aqueous solutions to hydrated solids. Therefore, the properties of water in polymer solutions have extensively been analyzed by NMR [9,10], DSC [11,12], sound velocity measurement [13], pulsed-field gradient NMR [14], neutron scattering [15], dielectric measurement [16], etc. As mentioned above, however, it is difficult to reveal the H-bonded network structure of water using these methods. In these fifteen years, we have been interested in analyses of the structure of water in aqueous polymer solutions from a Raman scattering attributable to the O–H stretching vibration band of water (2800–3800 cm−1 ) [17–29]. It was found that, when the molecular weight of the polymers was small enough to avoid a formation of “pseudo-network structure” (i.e., entanglement of the polymer chains), the number of H-bonds disrupted due to the presence of one monomer residue (Ncorr value) of a polyelectrolyte with an ionic group in the side chain (sodium polyacrylate, poly-L-lysine hydrobromide, etc.) was much larger than those for water-soluble nonionic polymers (poly(N -vinyl pyrrolidone) (PVPy) and poly(ethylene glycol) (PEG)) [18,21,22]. Moreover, the Ncorr value for a polymer with an ionizable group in the side chain (poly(acrylic acid) (HPAA), for example) was smaller than those for the polyelectrolytes with fully ionized groups and larger than those for the nonionic polymers [18]. Quite recently, it was found that homo- and random copolymers of zwitterionic phosphobetaine [22,24], sulfobetaine [23], and carboxybetaine vinyl monomers [25] did not disturb the Hbonded network structure of water significantly. Therefore, it is highly probable that zwitterionic polymers in general do not significantly affect the structure of water. There is another kind of ionic polymers whose total charges are neutralized: namely, amphoteric polymer (polyampholyte) composed of both anionic and cationic monomer residues. In a broad sense, amphoteric polymers can also be included into a category of “zwitterionic” polymers [30–33]. More than fifty years ago, the properties of an aqueous solution of amphoteric (zwitterionic) polymers prepared by the radical copolymerization of cationic and anionic vinyl monomers were studied [34,35]. Afterwards, zwitterionic polymers based on carboxy-, sulfo- and phospho-betaine monomer units were also investigated [36–47]. Meanwhile, zwitterionic compounds are characterized by having excellent dermatological properties, and frequently used in shampoos and other cosmetic products [48–51]. Biocompatibility and blood-compatibility (no thrombus formation) of carboxybetaine, phosphobetaine and sulfobetaine copolymers have often been reported, too [25,52–57]. In this work, the polarized Raman scattering of the O–H stretching vibration band of water has been examined in an aqueous solution of amphoteric random copolymer, poly(methacrylic acid-r-N -[3-(dimethylamino)propyl]methacrylamide), poly(MA-r-DMAPMA). As mentioned above, the unique properties of zwitterionic polymers might be strongly related to the interaction of the polymers with water. Therefore, Raman
Scheme 1. Chemical structure of amphoteric copolymer, poly(MA-rDMAPMA).
spectroscopy would be an excellent technique to explore the structure of water in the vicinity of the amphoteric polymers. 2. Materials and methods 2.1. Materials N -[3-(Dimethylamino)propyl]methacrylamide (DMAPMA) was from Aldrich, Milwaukee, WI. Methacrylic acid (MA) from Wako Pure Chemicals, Osaka, Japan, was distilled under the reduced pressure. Other reagents were commercially available. A Milli-Q grade water was used for the preparation of the sample solutions. 2.2. Preparation of copolymers (Scheme 1) DMAPMA (9.0 mL) and MA (4.2 mL) were copolymerized in water (100 mL) at 65 ◦ C for 24 h using 2,2 -azobis(2amidinopropane) dihydrochloride (V-50, Wako Pure Chemicals, 271 mg) and 2-mercaptoethanol (0.7 mL) as initiator and chain transfer reagent, respectively. After evaporation of the solvent, the obtained polymer was dissolved in water again, and purified by ultrafiltration using Amicon membranes (YM-3 (exclusion limit, 3 × 103 ) and YM-10 (104 )) to obtain a fraction between 3 × 103 and 104 daltons. The purified polymer was finally lyophilized (poly(MA-r-DMAPMA), 1.0 g). The weight-average and number-average molecular weights (Mw and Mn ) of the copolymer were determined to be 5.1 × 103 and 3.5 × 103 (Mw /Mn = 1.46), respectively, by gel permeation chromatography (GPC) (Waters HPLC system; column, Wako Gel G-40, Wako Pure Chemicals; mobile phase, 0.1 M LiCl in H2 O–N ,N -dimethylformamide (DMF) (4:1)) using pullulan standards (Showa Denko, Tokyo, Japan). Using a similar procedure, copolymers with various compositions were obtained. The composition of the copolymers was determined by 1 H NMR. Poly(MA-r-DMAPMA) are referred as P(Mx Dy ) (x and y are the mole fractions of MA and DMAPMA, respectively) hereafter. Results of the polymerization are summarized in Table 1. 2.3. Raman spectroscopy In the Raman spectra between 2500 and 4000 cm−1 for aqueous solution of various amphoteric polymers, the component of the O–H stretching band of water centered at 3250 cm−1
H. Kitano et al. / Journal of Colloid and Interface Science 313 (2007) 461–468
463
Table 1 Characteristics of various poly(MA-r-DMAPMA)s Code P(M100 D0 ) P(M67 D33 ) P(M47 D53 ) P(MnD89 ) P(M0 D100 )
Composition ratio (%) MA
DMAPMA
1/[η] (g/dL)/pX a
100 67 47 11 0
0 33 53 89 100
30.1/0.059 24.4/0.037 14.7/0.020 12.8/0.014 17.5/0.018
Mw (kD)
Mw / Mn
pHb
9.6 2.1 5.1 3.8 7.0
1.25 1.45 1.46 1.42 1.42
3.20 4.94 7.68 9.96 10.80
a The p values correspond to the 1/[η] values. X b pH value at p = 0.01. X
Fig. 2. Reduced viscosity vs concentration of amphoteric copolymers at 25 ◦ C in water: (") P(M100 D0 ), (2) P(M67 D33 ), (P) P(M47 D53 ), (F) P(M11 D89 ), (a) P(M0 D100 ); (· · ·) a guide to the eye.
polyampholytes at 25 ◦ C was determined using an Ubbelohde dilution type viscometer (Type 0B; Kusano, Tokyo, Japan). The time needed for sample solutions to pass through the capillary of the viscometer was determined within the uncertainties of ±0.1 s. Fig. 1. Raman spectra of O–H stretching region: (a) I (solid) and I⊥ (broken) spectra of pure water at 25 ◦ C. I⊥ /ρO–H spectrum is also shown by (· · ·); (b) the collective band of pure water.
was highly polarized, and diminished in the spectra at a perpendicular position (Fig. 1) [58–64]. The polarized O–H stretching band of water (“collective band”) is ascribed to the H2 O molecule executing ν1 vibrations all in phase with each other but the vibrational amplitude varying from molecule to molecule in strongly H-bonded water clusters [65–68]. The intensity of the collective band (Ic ) was separated from the spectra using equation [58–63] (Fig. 1), Ic = I − I⊥ /ρO–H ,
(1)
where I⊥ and I are the intensities of the spectra observed with the polarizer oriented perpendicular and parallel to the incident laser beam, respectively. The depolarization ratio, ρO–H , is an indicator of symmetry of the vibration mode, and was adjusted to minimize the contribution of large wavenumber region to the Ic value. The area of Ic was normalized by [58–63], I (w) dw, C = Ic (w) dw (2) where w is the Raman shift in cm−1 . 2.4. Viscosity measurements The reduced viscosity (ηsp /φ, ηsp = (η − η0 )/η0 , φ: concentration of polymer in g/dL) for aqueous solution of various
3. Results and discussion 3.1. Viscosity behavior of amphoteric polymer solutions The viscosity behaviors of various kinds of copolymers in water are shown in Fig. 2. The reduced viscosity (ηsp /φ) for an aqueous solution of polyMA (P(M100 D0 )) monotonously decreased with a decrease in the polymer concentration. This is in a good contrast with the increase of ηsp /φ with the decrease in the concentration of poly(MA-r-DMAPMA)s (P(M67 D33 ), P(M47 D53 ) and P(M11 D89 )) and polyDMAPMA (P(M0 D100 )), which is a typical behavior of the aqueous solution of ordinary polyelectrolytes: A decrease in the concentration of polyelectrolyte results in an increased Debye length, and the resulting increase in the electrostatic repulsion between the charged groups within the polymer chain induces expansion of the polymer chain [69,70]. However, it was reported that polyelectrolyte molecules such as sodium poly(styrene sulfonate) do not completely stretch out even in an infinitely diluted solution [71,72]. When the DMAPMA content in the copolymer was not in excess to that of MA (P(M67 D33 )), a slight increase in ηsp /φ upon dilution was observed in a low concentration region. With a further increase in the content of DMAPMA, the copolymer (P(M47 D53 )) indicated a less significant polyelectrolyte behavior, while a further increase in the content of DMAPMA (P(M11 D89 )) and polyDMAPMA (P(M0 D100 )) resulted in a much clearer behavior of polyelectrolyte. This is because the methacrylic acid residues in the polymers are weakly acidic
464
H. Kitano et al. / Journal of Colloid and Interface Science 313 (2007) 461–468
and not deprotonated to a large extent, while the DMAPMA residues are mostly protonated (pKa : polyMA, 6.62 at the degree of neutralization of 0.50 [73]; MA monomer, 4.65 [34]; polyDMAPMA, 8.00). In the figure, the ηsp /φ value at a relatively high concentration changed with the polymer composition. At this moment, there is no definitive explanation for the difference in the ηsp /φ value, because the composition and the degree of polymerization (DP) of polymers examined were different from each other. However, this point is not important in the present study because the aim is to clarify the influence of net-charge of the polymer. Assuming that the dissociation behavior of carboxylate and ammonium groups in polymers is indicated by pH = n log(α/1 − α) + pKa ,
(3)
the degree of dissociation (α) can be calculated [74,75]. The n value for a small molecular weight acid is unity, whereas that for polyelectrolytes is larger than unity due to the electrostatic effect. As for P(M47 D53 ), the degrees of dissociation of the MA and DMAPMA residues are calculated to be 77 and 41%, respectively, when the n value is assumed to be 2 [74,75]. Consequently, the ratio of anionic and cationic groups in the copolymer is calculated to be 100:88, suggesting that the electro-neutralization along the polymer chain is roughly realized. In the cases of P(M67 D33 ) and P(M11 D89 ), on the contrary, the ratio of charged groups leaned towards cationic (26:100) and anionic groups (100:79), respectively. These values are consistent with the viscosity behavior of the homo- and copolymers examined here. That is, in contrast with the noticeable increase in viscosity with the decrease in concentration for the copolymers P(M67 D33 ) and P(M11 D89 ), that for P(M47 D53 ) was relatively less significant due to the approximate electroneutralization of the polymer. It has been frequently pointed out that charge-imbalanced polyampholytes undergo a transition from a “globular coil” in a charge-balanced state to an “extended chain” with a large excess of charge on the polymer via a “beaded-necklace” morphology [76–79]. Such a tendency was consistent with a zwitterionic polymer, sulfobetaine polymer: At a large degree of quaternization, sulfobetaine polymer showed a linear relationship between the ηsp /φ value and the φ value more clearly [23]. However, the sulfobetaine polymer almost completely quaternized showed a very poor solubility (anti-polyelectrolyte effect) [23], which is in contrast with the large solubility of P(M47 D53 ). In the presence of comparable amount of DMAPMA residues, the MA residue is significantly deprotonated, which induces electro-neutralization of the side chain to form an ion pair, while no effective stacking between the cationic and anionic groups of neighboring monomer residues would appear, probably because the distance between the carboxylate and tertiary ammonium groups might be long enough to avoid a disadvantageous strong ionic association. As for ordinary zwitterionic polymers, a disadvantageous inter-tether ionic association could be weakened by the introduction of hydrophilic comonomer residues between the zwitterionic monomer residues [41,42].
3.2. Structure of water in aqueous polymer solutions The O–H stretching Raman band of liquid water gave a broad band (Fig. 1a) composed of several overlapping components which were attributed to the unperturbed O–H stretching band by the intra- and inter-molecular vibrational coupling of the O–H oscillators [3]. The relative intensities of the collective bands (C, Fig. 1b) are reduced by the decoupling of the O–H oscillators: (1) when the hydrogen bond between the coupled O–H oscillators is broken by translational or rotational rearrangement of the water molecules, and (2) when the stretching frequency of an O–H oscillator is quite different from that of the O–H oscillator which is combined by a H-bond with the former one [58]. The defect probability, Pd , that an O–H oscillator is excluded from the H-bonding network of water molecules because of an unfavorable position or orientation, is defined as Pd =
C w − Cx , Cw
(4)
where Cw and Cx are the intensities of the collective band of pure water and aqueous solution at a certain temperature, respectively. The Cx values of aqueous solutions of simple electrolytes [58] and polar substances were smaller than that of pure water at the same temperature. This is caused, as mentioned above, by the exclusion of the O–H oscillator of hydrating water (HOH · · ·solute and/or H2 O · · ·solute) from the collective O–H oscillators of the water cluster. On the contrary, those of hydrophobic substances such as t -butanol [62] (mol fraction (pX ), 0.04) and tetra-n-butyl ammonium hydroxide [63] (pX = 0.013) were higher than that of pure water, due to an enhancement of the H-bond between water molecules in hydrophobic hydration shells around the hydrocarbon moiety [80]. These results show that the vibrational band attributed to the O–H oscillators of water cluster can be distinguished from that of the water hydrated to solutes by using the polarized Raman spectroscopy, and that this method has an ability to detect the changes in the H-bonded network structure of water even in dilute–semi-dilute aqueous solutions (pX = 0.01–0.05 (ca. 0.55–2.8 mol/L)). In a dilute solution, polymer coils are separated from each other and the intermolecular interaction between the coils is sufficiently weak, and therefore, the solution can be described as a “non-ideal gas.” When the polymer concentration (φ) approaches a critical value (φ ∗ ), the polymer coils begin to overlap and make a pseudo-network. In the concentration region (φ > φ ∗ ), the polymer solutions are called “semi-dilute” solutions [81]. Based on a previously suggested criterion between the low and high-concentration regimes for random coils in a good solvent, 1 < φ[η] < 10 [82], the value of φ ∗ can be roughly related to the intrinsic viscosity [η] as φ∗ ∼ = 1/[η].
(5)
From the value of [η] for aqueous solutions of amphoteric polymers at 25 ◦ C (Table 1), it was concluded that all the polymers examined here give dilute solutions at pX = 0.01.
H. Kitano et al. / Journal of Colloid and Interface Science 313 (2007) 461–468
3.3. Estimation of Ncorr values for polymers
Table 2 The Ncorr values for various polymersa
To elucidate the effects of the properties of the monomer residue of various polymers, the number of H-bond defects introduced into the H-bonded network structure of water per one monomer unit of a polymer, N , was calculated from the defect probability (Pd ) using equation, N = Pd /fX ,
465
(6)
where fX is the number of monomer units per one O–H oscillator in a water molecule [18,20–22,83]. It should be mentioned here that the liquid water has an intrinsic defect in the H-bonded network structure. At the estimation of Pd values, this factor is not necessary to be considered, because the Pd values are calculated by the ratio of (Cw − Cx ) and Cw values. However, the number of O–H oscillators H-bonded in pure water at certain temperature should be smaller than the total number of O–H oscillators. Therefore, the N value was corrected using the C values for perfect ice and pure water (Cice and Cw , respectively) as Ncorr = N × (Cw /Cice ) [58]. It should be mentioned here that the Ncorr value represents the degree of perturbation on the H-bonded network structure of water. Since the derivation of Ncorr value includes several assumptions, it does not precisely indicate the actual number of H-bonds that each group of polymers perturb. Similar to small molecular solutes mentioned in the previous section, ionized and polar groups in water-soluble polymer chains and counter ions may disturb the H-bond between the water molecules and raise the Ncorr value, whereas hydrophobic moieties such as hydrocarbon chains may enhance the H-bond and reduce the Ncorr value. The exposure area of the chemical groups to water is also important to the structure of water around them. The effect of various water-soluble polymers on the C value at the constant molar fraction of monomer units (pX = 0.01) was examined, and the Ncorr values obtained for the polymers, in which the H-bond defects are localized in the hydration shell of polymer chains and their counter ions, were determined and are shown in Table 2. The Ncorr values of ordinary polyelectrolytes with one kind of ionic side group (sodium polyacrylate (NaPAA) [18,22], sodium poly(ethylene sulfonate) (NaPES), poly-L-lysine HBr salt (PLL·HBr) [22] and poly[(3-(methacryloyl)amino)propyltrimethylammonium chloride] (polyMAPTAC)) [26] are much larger than those of water-soluble nonionic polymers (PEG and PVPy) [18,22], indicating that ionic groups and the counterions of polyelectrolytes strongly disturb the structure of water in their hydration shells (electrostatic hydration). The very small Ncorr values for nonionic water-soluble polymers (PEG and PVPy) coincide with the previous report that oligo(ethylene glycol)s (degree of polymerization, 4–7) dissolves in water without much perturbation to the water structure [84]. 3.4. The Ncorr values for amphoteric copolymers Table 2 and Fig. 3 show that the amphoteric copolymer, P(M47 D53 ), had relatively the smallest positive Ncorr value,
Polymer
Ncorr value
Reference
NaPES PLL·HBr PolyMAPTAC NaPAAb HPAAc NaPolyMA PEG PVPy PMPC PSBB PCMB P(M100 D0 ) P(M68 D32 ) P(M47 D53 )
5.1 5.5 9.4 5.9 2.3 9.0 0.7 0.6 −0.7 0.7 −0.27 6.0 4.3 0.7d 5.4e 3.6f 4.6 5.1
[22] [22] [26] [18] [18] [26] [22] [22] [22] [23] [25] [26] This work This work This work This work This work This work
P(Mn D89 ) P(M0 D100 )
a In H O at 25 ◦ C and p = 0.01 unless mentioned. The uncertainties of X 2 Ncorr values are ±1 at the most. b M = 6.5 kD at p = 0.05. w X c HPAA (M = 5 kD) at p = 0.05. w X d pH 7.68. e After the addition of 0.1 M HCl (pH 4.83). f After the addition of 0.1 M NaOH (pH 9.35).
Fig. 3. The effect of content of MA and DMAPMA residues in amphoteric copolymers on the Ncorr value at 25 ◦ C.
and, with the increase or decrease in the composition ratio of MA, the Ncorr value increased. It is worth noting that the Ncorr values for the copolymers were smaller than that of polyMA (P(M100 D0 )) and polyDMAPMA (P(M0 D100 )). The MA and DMAPMA residues have ionizable groups (–COOH and –N(CH3 )2 , respectively), and they are partly ionized under neutral conditions as mentioned in Section 3.1. Therefore, we had expected that the water structure in aqueous solutions of copolymers, poly(MA-r-DMAPMA), would be more significantly disturbed in comparison with that in polyMA and polyDMAPMA solutions, because the amount of ionic species in the copolymers is much larger than that in polyMA and polyDMAPMA. The experimental result was opposite to the expectation.
466
H. Kitano et al. / Journal of Colloid and Interface Science 313 (2007) 461–468
The large Ncorr values for polyDMAPMA and polyMA were not contradictory to the general knowledge and our previous reports [18–26]. However the disturbing effects of the amphoteric copolymers on the water structure were much smaller than that of the ordinary polyelectrolytes. This is in accordance with the previous result that the zwitterionic polymer, poly[1-carboxyN ,N -dimethyl-N -(2 -methacryloyloxyethyl)methanaminium inner salt] (PCMB, carboxybetaine polymer), did not disturb the structure of water (Ncorr = −0.27 for PCMB (Mw , 11.4 kD)) [25]. The slightly negative Ncorr value for PCMB is probably due to the small error associated with separation of the collective band from the O–H stretching band, and therefore, they can be regarded as zero approximately. A similar tendency was observed for poly[4-sulfo-N ,N -dimethyl-N -(3 methacryloylaminopropyl)butanaminium inner salt] (PSBB, sulfobetaine polymer) [23], and poly(2-methacryloyloxyethyl phosphorylcholine) (PMPC, phosphobetaine polymer), too [22]. Using 1 H NMR, El Seoud and coworkers showed that a sulfobetaine surfactant, 3-sulfo-N ,N -dimethyl-N -(n-dodecyl)propanaminium inner salt, does not affect the state of interfacial water, due to strong intermolecular interactions between the head ions of neighboring molecules in the micelles [85]. Previously, we found that the Ncorr values for the zwitterionic surfactants were close to zero, too [86]. On the other hand, Halle and coworkers reported that the dynamic sense of water in the vicinity of polymers (adsorbent) was strongly perturbed [87,88], which is in conflict with the report by El Seoud and coworkers and by us. This confliction seems to be caused by the difference in the ordinary (poly)electrolyte and the zwitterionictype one. Actually, El Seoud et al. also pointed out that, with an increase in the distance between the anionic and cationic moieties in zwitterionic molecule, the structure of neighboring water became similar to that around the ordinary electrolytes: in other words, the two kinds of claims by El Seoud et al. and Halle et al. were the same in case of the zwitterionic molecule whose distance between anionic and cationic moieties was long (number of methylene groups 3) [85]. Furthermore, the C value of aqueous amino acid solutions under neutral conditions were found to be slightly smaller than that of pure water, and almost constant in spite of the difference in hydrophobicity of the side chain [89]. In other words, when both α-amino and α-carboxyl groups of amino acids are ionized (zwitterionic), the Ncorr value for the amino acids is slightly positive and almost independent of the hydrophobicity of the side chain. The small Ncorr values evaluated for the zwitterionic compounds with a small molecular weight and those for the amphoteric copolymers might be due to the same reason. As discussed in Section 3.1, when the n value in Eq. (3) is assumed to be 2, the ratio of anionic and cationic groups in P(M47 D53 ) is calculated to be 100:88, suggesting that the electro-neutralization along the polymer chain is roughly realized, whereas, in the cases of P(M67 D33 ) and P(M11 D89 ), the ratio of charged groups leaned towards cationic (26:100) and anionic groups (100:79), respectively. The deprotonated carboxylate group and protonated tertiary amino group, which are in close proximity to each
other in P(M47 D53 ), might counteract the electrostatic hydration: It has generally been suggested that the partially negative oxygen atom of the water molecule is toward the positively charged ions in solution, whereas the negatively charged ions attract the partially positive hydrogen atoms of water, creating their hydration shells [90]. When the cation and anion are in close proximity, the orientation of the hydrating water might be largely disturbed resulting in the partial collapse of electrostriction. Such a tendency would be more significant when the number of cationic and anionic groups are becoming close each other, because, in the linear polymer system, the equalization of oppositely charged groups statistically means relatively the closest proximity between the oppositely charged groups. In the present polymer system, such an equalization of the number of oppositely charged ionic groups could be realized by the deprotonation of MA residues by the neighboring DMAPMA residues as discussed above (Fig. 3 and Table 2). By the addition of 0.1 M of HCl and NaOH to the solution of P(M47 D53 ), the electro-equalization within the polymer molecules was collapsed (ratio of anionic and cationic groups calculated by Eq. (3) (n = 2): 10:100 and 100:14, respectively), and the Ncorr values for P(M47 D53 ) became larger than that dissolved in pure water (Table 2), supporting the hypothesis mentioned above. The hydrocarbon main chain in the amphoteric polymers might be hidden behind the bulky cationic and anionic groups from water to minimize the free energy of the system, and consequently, the effect of the main chain on the Ncorr value for the polymers might not appear clearly. Recently we have reported that the Ncorr values for α,ω-amino acids increased with an increase in the distance between the amino and carboxyl groups [91], which is in accordance with the 1 H NMR study by El Seoud and coworkers on the zwitterionic surfactants as discussed above. Høiland reported that, when hydration sheaths for ionizable groups of a dicarboxylic acid are in close proximity and overlapped, an additivity rule for the hydration of each component is not applicable [92]. These results support the tendency observed in this work. Previously, it was reported that self-assembled monolayers (SAMs) of zwitterionic sulfides or 1:1 mixture of cationic and anionic sulfides resist against the non-specific adsorption of various proteins from a buffer solution [93]. We found that a brush of zwitterionic telomer is resistant against the non-specific adsorption of proteins, too [94]. It has very often been pointed out that the absence of non-specific adsorption of serum proteins has been considered as an essential factor for usability of polymeric materials in medical fields [52,53]. Therefore, there is a possibility that the mildness of the amphoteric compounds to the structure of water at their surface plays an important role in the resistance against non-specific adsorption of proteins, resulting in the appearance of blood- and bio-compatibilities. Previously, it was found that linear and branched polyethyleneimines modified with acrylic acid showed no acute toxicity in mice up to dosage of 1 g/kg [95]. The zwitterionic form was attributed to the non-toxic nature of the amino acid type polymers, supporting the tendency observed in this work.
H. Kitano et al. / Journal of Colloid and Interface Science 313 (2007) 461–468
4. Conclusion In contrast with ordinary polyelectrolytes carrying one kind of ionized group, the amphoteric copolymer with comparable contents of cationic and anionic monomer residues had a very small disturbing effect on the structure of the hydrogen bonding network of water molecules in dilute solutions due to the inter-tether proximity between the oppositely charged groups (counteraction of the electrostatic hydration effects). The weak disturbing effect of amphoteric polymers on the water structure could be attributed to the intrinsic property of the polymers themselves, and not to the polymer–polymer coordination. Taking account of biocompatibility of zwitterionic polymers, a small disturbing effect of charge-balanced polymer materials on the structure of vicinal water (in other words, resemblance of the water structure in the vicinity of the polymers to that of free water) seemed to be important to afford the polymers biocompatibility. Acknowledgments This work was supported by Grants-in-Aid (16205015, 17750105 and 19350055) from the Japan Society for the Promotion of Science. The authors are indebted to the Toyama Prefecture New Century Foundation, Toyama, Japan, for its financial support. M.G.-I. wishes to thank the Sekisui Science Foundation, Osaka, Japan, for its financial support. The authors are grateful to Mr. K. Takaha of this laboratory for his technical assistance. References [1] S. Scheiner, Hydrogen Bonding: A Theoretical Perspective, Oxford Univ. Press, Oxford, 1997. [2] D. Eisenberg, W. Kauzmann, The Structure and Properties of Water, Clarendon Press, London, 1969. [3] G.E. Walrafen, in: F. Franks (Ed.), Water: A Comprehensive Treatise, vol. 1, Plenum Press, New York, 1972, chap. 2. [4] G.E. Walrafen, in: W.A.P. Luck (Ed.), Structure of Water and Aqueous Solutions, Verlag Chemie, Weinheim, 1974. [5] E. Westhof (Ed.), Water and Biological Macromolecules, McMillan Press, Houndmills, 1993. [6] C.H. Collie, J.B. Hasted, D.M. Riston, Proc. Phys. Soc. 60 (1948) 145. [7] E.H. Grant, T.J. Buchanan, H.F. Cook, J. Chem. Phys. 26 (1957) 156. [8] S.K. Garg, C.P. Smyth, J. Chem. Phys. 43 (1965) 2959. [9] S. Katayama, S. Fujiwara, J. Am. Chem. Soc. 101 (1979) 4485. [10] F.X. Quinn, E. Kampf, G. Smyth, V.J. McBriety, Macromolecules 21 (1988) 3191. [11] N. Murase, K. Gonda, T. Watanabe, J. Phys. Chem. 90 (1986) 5420. [12] H. Ohno, M. Shibayama, E. Tsuchida, Makromol. Chem. 184 (1983) 1017. [13] K. Gekko, H. Noguchi, Biopolymers 10 (1994) 1513. [14] A.C. Rarnes, T.W.N. Bieze, J.E. Enderby, J.C. Leyte, J. Phys. Chem. 98 (1994) 11527. [15] T.W.N. Bieze, A.C. Barnes, C.J.M. Huige, J.C. Leyte, J. Phys. Chem. 98 (1994) 6568. [16] N. Shinyashiki, Y. Matsumura, N. Miura, S. Yagihara, S. Mashimo, J. Phys. Chem. 98 (1994) 13169. [17] T. Terada, Y. Maeda, H. Kitano, J. Phys. Chem. 97 (1993) 3619. [18] Y. Maeda, N. Tsukida, H. Kitano, T. Terada, J. Yamanaka, J. Phys. Chem. 97 (1993) 13903.
467
[19] T. Terada, T. Inaba, H. Kitano, Y. Maeda, N. Tsukida, Macromol. Chem. Phys. 195 (1994) 3261. [20] N. Tsukida, Y. Maeda, H. Kitano, Macromol. Chem. Phys. 197 (1996) 1681. [21] N. Tsukida, H. Muranaka, M. Ide, Y. Maeda, H. Kitano, J. Phys. Chem. B 101 (1997) 6676. [22] H. Kitano, K. Sudo, K. Ichikawa, M. Ide, K. Ishihara, J. Phys. Chem. B 104 (2000) 11425. [23] H. Kitano, M. Imai, K. Sudo, M. Ide, J. Phys. Chem. B 106 (2002) 11391. [24] H. Kitano, M. Imai, T. Mori, M. Gemmei-Ide, Y. Yokoyama, K. Ishihara, Langmuir 19 (2003) 10260. [25] H. Kitano, S. Tada, T. Mori, K. Takaha, M. Gememei-Ide, M. Tanaka, M. Fukuda, Y. Yokoyama, Langmuir 21 (2005) 11932. [26] H. Kitano, K. Takaha, M. Gemmei-Ide, Phys. Chem. Chem. Phys. 8 (2006) 1178. [27] Y. Maeda, H. Kitano, Spectrochim. Acta A 51 (1995) 2433. [28] Y. Maeda, H. Kitano, Trends Phys. Chem. 6 (1997) 269. [29] Y. Maeda, M. Ide, H. Kitano, J. Mol. Liq. 80 (1999) 149. [30] H. Morawetz, Macromolecules in Solution, Interscience Publishers, New York, 1965. [31] S.E. Kudaibergenov, Polyampholytes: Synthesis, Characterization and Application, Kluwer Academic/Plenum Publishers, New York, 2002. [32] A.B. Lowe, C.L. McCormick, Chem. Rev. 102 (2002) 4177. [33] S. Kudaibergenov, W. Jaeger, A. Laschewsky, Adv. Polym. Sci. 201 (2006) 157. [34] T. Alfrey, H. Morawetz, J. Am. Chem. Soc. 74 (1952) 436. [35] G. Ehrlich, P. Doty, J. Am. Chem. Soc. 76 (1954) 3764. [36] H. Ladenheim, H. Morawetz, J. Polym. Sci. 26 (1957) 251. [37] K. Rosenheck, A. Katchalsky, J. Polym. Sci. 32 (1958) 511. [38] R. Hart, D. Timmerman, J. Polym. Sci. 28 (1958) 638. [39] W.-F. Lee, C.-C. Tsai, Polymer 35 (1994) 2210. [40] W.-F. Lee, Y.-M. Tu, J. Appl. Polym. Sci. 72 (1999) 1221. [41] E.E. Kathmann, L.A. White, C.L. McCormick, Polymer 38 (1997) 871. [42] R.S. Armentrout, C.L. McCormick, Macromolecules 33 (2000) 2944. [43] S.A. Ali, A. Rasheed, M.I.M. Wazeer, Polymer 40 (1999) 2439. [44] I.V. Berlinova, I.V. Dimitrov, R.G. Kalinova, N.G. Vladimirov, Polymer 41 (1999) 831. [45] S.F. Lascelles, F. Malet, R. Mayada, N.C. Billingham, S.P. Armes, Macromolecules 32 (1999) 2462. [46] C. Viklund, K. Irgum, Macromolecules 33 (2000) 2539. [47] J. Virtanen, M. Arotçaréna, B. Heise, S. Ishaya, A. Laschewsky, H. Tenhu, Langmuir 18 (2002) 5360. [48] B. Jönsson, B. Lindman, K. Holmberg, B. Kronberg, Surfactants and Polymers in Aqueous Solution, Wiley, Chichester, 1998, pp. 21–23. [49] D. Saint-Leger, G. Hussler, G. Kaba, Eur. Pat. Appl. EP 728473. [50] T. Murayama, Japan Patent 2000-219670A. [51] K. Takano, K. Ide, Japan Patent 11012596A. [52] T. Nakaya, Y.J. Li, Prog. Polym. Sci. 24 (1999) 143. [53] A.L. Lewis, Colloids Surf. B Biointerfaces 18 (2000) 261. [54] Y.L. Yuan, F. Ai, J. Zhang, X.B. Zang, J. Shen, S.C. Lin, J. Biomater. Sci. Polym. Ed. 13 (2002) 1081. [55] Y. Yuan, X. Zang, F. Ai, J. Zhou, J. Shen, S. Lin, Polym. Int. 53 (2004) 121. [56] A.B. Lowe, M. Vamvakaki, W.A. Wasall, L. Wong, N.C. Billingham, S.P. Armes, A.W. Lloyd, J. Biomed. Mater. Res. 52 (2000) 88. [57] S.L. West, J.P. Salvage, E.J. Lobb, S.P. Armes, N.C. Billingham, A.L. Lewis, J.G.W. Hanlon, A.W. Llyod, Biomaterials 25 (2004) 1195. [58] J.L. Green, A.R. Lacey, M.G. Sceats, J. Phys. Chem. 90 (1986) 3958. [59] J.L. Green, A.R. Lacey, M.G. Sceats, S.J. Henderson, R.J. Speedy, J. Phys. Chem. 91 (1987) 1684. [60] J.L. Green, A.R. Lacey, M.G. Sceats, J. Chem. Phys. 86 (1987) 1841. [61] J.L. Green, A.R. Lacey, M.G. Sceats, Chem. Phys. Lett. 134 (1987) 385. [62] J.L. Green, A.R. Lacey, M.G. Sceats, Chem. Phys. Lett. 137 (1987) 527. [63] J.L. Green, M.G. Sceats, A.R. Lacey, J. Chem. Phys. 87 (1987) 3603. [64] N. Karger, H.-G. Lüdemann, M.G. Sceats, Ber. Bunsenges. Phys. Chem. 99 (1995) 1104. [65] S.A. Rice, M.G. Sceats, J. Phys. Chem. 85 (1981) 1108. [66] M.G. Sceats, S.A. Rice, in: F. Franks (Ed.), Water—A Comprehensive Treatise, vol. 7, Plenum Press, New York, 1983, chap. 2.
468
[67] [68] [69] [70]
[71] [72] [73] [74] [75] [76] [77] [78] [79] [80] [81] [82]
H. Kitano et al. / Journal of Colloid and Interface Science 313 (2007) 461–468
F.H. Stillinger, Science 209 (1980) 451. D.E. Hare, C.M. Sorensen, J. Chem. Phys. 93 (1990) 25. R.M. Fuoss, U.P. Strauss, J. Polym. Sci. 246 (1948) 602. H. Dautzenberg, W. Jaeger, J. Kotz, B. Philipp, Ch. Seidel, D. Stcherbina, Polyelectrolytes: Formation, Characterization and Application, Hanser, Munich, 1994. H. Vink, Polymer 33 (1992) 3771. J. Yamanaka, H. Matsuoka, H. Kitano, M. Hasegawa, N. Ise, J. Am. Chem. Soc. 112 (1990) 587. E.V. Anufrieva, T.M. Birshtein, T.N. Nekrasova, O.B. Ptitsyn, T.V. Shveleva, J. Polym. Sci. C 16 (1968) 3519. A. Katchalsky, P. Spitnik, J. Polym. Sci. 2 (1947) 432. A. Katchalsky, J. Polym. Sci. 13 (1957) 69. Y. Kantor, M. Kardar, Europhys. Lett. 27 (1994) 643. Y. Kantor, M. Kardar, Phys. Rev. E 51 (1995) 1299. Y. Kantor, M. Kardar, Phys. Rev. E 52 (1995) 835. D.B. Thomas, Y.A. Vasilieva, R.S. Armentrout, C.L. McCormick, Macromolecules 36 (2003) 9710. A. Kauzmann, Adv. Protein Chem. 14 (1959) 39. P.-G. de Gennes, Scaling Concepts in Polymer Physics, Cornell Univ. Press, Ithaca, 1979. W.W. Graessley, Adv. Polym. Sci. 16 (1974) 1.
[83] N.J. Tao, S.M. Lindsay, A. Rupprecht, Biopolymers 28 (1989) 1019. [84] T. Sato, H. Niwa, A. Chiba, J. Chem. Phys. 108 (1998) 4138. [85] O.A. El Seoud, R.C. Bazito, G.K. Barlow, Prog. Colloid Polym. Sci. 111 (1998) 151. [86] H. Kitano, M. Imai, M. Gemmei-Ide, K. Takaha, J. Colloid Interface Sci. 269 (2004) 459. [87] K. Fontell, B. Halle, Liq. Cryst. 16 (1994) 235. [88] S. Wiesner, E. Kurian, F.G. Prendergast, B. Halle, J. Mol. Biol. 286 (1999) 233. [89] M. Ide, Y. Maeda, H. Kitano, J. Phys. Chem. B 101 (1997) 7022. [90] R.H. Garrett, C.M. Grisham, Biochemistry, Saunders College Publishing, Fort Worth, 1995, p. 36. [91] H. Kitano, K. Takaha, M. Gemmei-Ide, J. Colloid Interface Sci. 283 (2005) 452. [92] H. Høiland, J. Chem. Soc. Faraday I 71 (1975) 797. [93] R.E. Holmlin, X. Chen, R.G. Chapman, S. Takayama, G.M. Whitesides, Langmuir 17 (2001) 2841. [94] H. Kitano, A. Kawasaki, H. Kawasaki, S. Morokoshi, J. Colloid Interface Sci. 282 (2005) 340. [95] S. Kobayashi, L. Gross, G. Muacevic, H. Ringsdorf, Makromol. Chem. 184 (1983) 793.