1590
J. Phys. Chem. B 2007, 111, 1590-1596
Voltammetric Characterization on the Hydrophobic Interaction in Polysaccharide Hydrogels Yimei Yin,† Hongbin Zhang,*,† and Katsuyoshi Nishinari‡ Department of Polymer Science and Engineering, Shanghai Jiao Tong UniVersity, Shanghai 200240, People’s Republic of China, and Graduate School of Human Life Science, Osaka City UniVersity, 3-3-138 Sugimoto, Sumiyoshi-ku, Osaka 558-8585, Japan ReceiVed: September 15, 2006; In Final Form: December 8, 2006
Cyclic voltammetric (CV) investigations on the properties of microdomains in polysaccharide hydrogels, methyl cellulose (MC) and κ-carrageenan (CAR), coated on glassy carbon electrodes were reported in which methylene blue (MB), tris(1,10-phenanthroline)cobalt(III) (Co(phen)33+/2+) cations, and ferricyanide/ferrocyanide (Fe(CN)63-/4-) anions were used as electroactive probes. Information on the patterns and strength of intermolecular interactions in these polysaccharide hydrogels can be inferred from the net shift of normal potentials (E°′), the change of peak currents (ip), the ratio of binding constants (Kred/Kox) for reduced and oxidized forms of bound species, and the apparent diffusion coefficients (Dapp) of probe in hydrogels. The transition of hydrophobic interaction in MC hydrogel with temperature was manifested by the CV method, which is in agreement with the evolution of the storage modulus (G′) during gelation. It was also found that, in addition to inducing the change of E°′ and ip of these probes used, the hydrophobic-hydrophilic nature of the microenvironment in hydrogels coated on the substrate electrodes greatly influenced the peak-peak separation (∆Ep) of MB and the redox reversibility of Fe(CN)63-/4- via modulation of both the heterogeneous electron-transfer process at the gel-substrate interface and the charge-transfer process in hydrogels. The results imply that the CV method is of significant benefit to the understanding of the gelation driving forces in the polysaccharide hydrogels at a molecular level.
Introduction Polysaccharide hydrogels play a prominent role in modern industry, especially in food, medical, pharmaceutical and cosmetic industries as, e.g., water retainers or delivery matrixes.1 They have recently attracted much interest for more applications, such as biocompatible matrixes for biosensors,2 conductive solid medium,3 smart gel,4 cell culturing,5 and scaffolding materials for tissue engineering.6 Moreover, some of the polysaccharides (e.g., 1,3-β-D-glucans) were found to function as strong anticancer or anti-HIV reagents by physically interacting with target biomolecules of DNA, RNA, and HIV virus concomitant with gelation.7 Polysaccharide hydrogels can be regarded as solid water3 because of a large amount of water (>90%) contained in a small quantity of polysaccharide network structures. The nature of the microdomains composed of polysaccharide macromolecules and locked water in the hydrogels is significantly related to the composition, conformation, and junction of the polysaccharides. In other words, it is closely related to the intra- and intermolecular interactions induced by the external stimuli such as temperature, pH, light, electric/magnetic field, salt, and solvent composition. Therefore, the detection of the nature of the microdomains can provide valuable information on the interactions occurring among polysaccharide macromolecules during gelation. The gelation of polysaccharides is the subject of numerous studies8,9,10 for both practical application and scientific interest * To whom correspondence should be addressed. Fax: +86-2154741297. E-mail:
[email protected]. † Shanghai Jiao Tong University. ‡ Osaka City University.
as well, and they are usually driven by cooperative noncovalent interactions which generally fall into four categories: electrostatic, hydrophobic, van der waals, and hydrogen bonding forces. To understand the interactions is not only pivotal for elucidating the mechanisms of gelation and the origin of bioactivities of polysaccharides but also a central concern in structural biochemistry and biomaterial science.11 The characterization of these interactions is a challenge for their low-energy effect, while the hydrophobic interaction is especially difficult to investigate because of its entropy-controlling nature without obvious heat, electric, and photic effects. Furthermore, the strength of hydrophobic interaction does not rely on the attraction between nonpolar groups but is determined by the minimum potential energetic status of the system. Many attempts have been made to investigate the hydrophobic interaction in polysaccharide solutions and clear gels by means of various techniques, such as light scattering,12 fluorescence spectroscopy,13,14 13C NMR,7,14 and UV-vis spectroscopy,15 etc. These spectroscopic methods have provided valuable insight into the hydrophobic interaction that occurs during aggregation of macromolecules and the physicochemical properties of the polysaccharide solutions. However, of equal importance is the need to understand the hydrophobic interactions in the hydrogel state. Some limitations may exist in many of the above methods in studying hydrogels; for example, NMR requires very often isotopic enriched samples, and its signal may become very complicated during sol-gel transition. Although pyrene fluorescence probe measurement is widely used for studying hydrophobic microdomains in solution systems13 or in clear gels,14 it might not be applicable for turbid gels which are usually the case for many polysaccharide hydrogels. We
10.1021/jp0660334 CCC: $37.00 © 2007 American Chemical Society Published on Web 01/31/2007
Polysaccharide Hydrogel Hydrophobic Interaction
Figure 1. Repeating monomer units of polysaccharides: (a) methylcelullose and (b) κ-carrageenan.
addressed this topic in the present work with electrochemical cyclic voltammetric (CV) method, which is commonly considered as a fast, informative, and relatively inexpensive technique. The detecting principle of the electrochemical method relies on probe-microdomain interactions which alter the rate of electron transfer (ET) and diffusion behavior of probe species. The electrochemical methods such as cyclic voltammetry, steady-state voltammetry, and chronoamperometry are commonly used for characterizing DNA structures and cyclodextrin inclusion complexes, yielding information about noncovalent (electrostatic or hydrophobic) interactions occurring between the small molecules and their microenviroments.16,17 These methods are also carried out to clarify the hydrophilic/hydrophobic nature of surfactant micelles.18 Recently, Ciszkowska et al. showed the advantages of these techniques in the study of some hydrogels,19,20 such as ι-carrageenan, agarose, and poly(N-isopropylacrylamide), to characterize the effect of structural changes or volume phase transitions on the diffusion behavior of electroactive probes. But, as far as we know, electrochemical methods have been scarcely employed to study the hydrophobic interactions between probe molecules with microdomains in polysaccharide hydrogels stemming from intraand intermolecular interactions between hydrophobic groups. In the present work, based on our primary study,21 we employed a hydrophobic substituted cellulose, methylcellulose (MC, shown in Figure 1a), as a model polysaccharide and three kinds of molecules, methylene blue (MB), tris(1,10-phenanthroline)cobalt(III) cation, and ferricyanide, as probes in CV studies. In comparison with MC, the electrochemistry of MB and ferricyanide at hydrophilic anionic polysaccharide, κ-carrageenan (CAR, shown in Figure 1b), modified electrodes has also been investigated. Experimental Section 1. Materials. MC was kindly provided by Shin-Estu Chemical Co. Ltd., Japan, with the commercial name of SM4000. According to the manufacturer, MC has an average degree of substitution (DS) of 1.8 and a viscosity of 4.69 Pa‚s at 20 °C for a 2 wt % aqueous solution. And, its weight-average molecular weight (Mw) is ca. 3.8 × 105 g/mol.22,23 κ-Carrageenan was kindly provided by Danisco (China) in Kunshan, China, with an intrinsic viscosity of 44 cm3/g measured by an Ubbelohde viscometer in 0.1 mol/L NaCl at 25 °C and a Mw of 4.8 × 106 g/mol determined by gel
J. Phys. Chem. B, Vol. 111, No. 7, 2007 1591 permeation chromatography (GPC). The cations contained in the sample measured by inductively coupled plasma-atomic emission spectroscopy (ICP-AES) are as follows: K+, 53.8 µg/ g; Na+, 10.5 µg/g; Al3+, 1.5 µg/g; Mg, 3.2 µg/g; Ca, 2.7 µg/g. The content of the sulfur element is 491.8 µg/g. Tris(1,10-phenanthroline)cobalt(III) cations, Co(phen)3+/2+, were prepared according to previously reported procedures,24 being recrystallized twice. All other reagents were of analytical grade, purchased from China Medical Reagent Co. and used as received. Distilled deionized water was used throughout. 2. Preparation of Polysaccharide-Coated Electrodes. Glassy carbon disc (GC) working electrodes were used. The geometry areas of GCs were 0.049 or 0.070 cm2, respectively. The GCs were first polished to mirror and then ultrasonically cleaned in absolute ethanol and distilled water for 2-5 min, respectively. The GCs were not used until reversible cyclic voltammetrical curves were observed in a 0.1 M phosphate saline buffer solution (PBS, pH 7.0) containing 1 mmol/L Fe(CN)63-/ 4-. MC coated electrodes (GC/MC) were prepared from a 0.5% (w/v) MC solution. MC powders were dispersed with distilled water at 70 °C, stirred well, and then kept in a refrigerator (4 °C) for at least 24 h to obtain a clear solution. A volume of 6 µL of such a MC solution was dropped onto the surface of a clean GC, and then the electrode was kept at 60 °C in an oven for 30 min. The obtained GC/MC was immediately transferred to a 45 °C test solution containing 5.0 × 10-5 mol/L methylene blue (MB) and 0.1 mol/L PBS and then kept for 15 min during which equilibrium of MB accumulation could be established between the coated layer and the bulk solution. The MC coated layer swelled in the above solution and resulted in a MC hydrogel of ca. 4% MC. Because of the thermal hysteresis of de-gelation (occurred at about 30 °C in this case), the MC hydrogel would not melt at the testing temperature range of 45-70 °C, which was controlled within (0.2 °C by a water circulator. CV measurements were performed then, and the first scans were recorded. CAR-coated electrodes (GC/CAR) were prepared by dipping a polished glassy carbon electrode into a 1% (w/v) CAR aqueous solution containing 0.05 mol/L KCl in which CAR was dissolved by heating the solution above 70 °C. Then the electrode was cooled and dried in the air at room temperature (20 °C) to get a GC/CAR electrode. 3. CV Measurements. CV measurements were carried out with a CHI 602 electrochemical analyzer (CHI Instrument, Austin, TX), and corresponding software was used for manipulation and data storage. The voltammetric experiments were performed with a typical three-electrodes system, bare GC or coated-GC as working electrode, saturated calomel electrode (SCE) reference, and Pt wire counter electrode. All solutions were deaerated with highly pure nitrogen prior to measurements. Supporting electrolytes were 0.1 mol/L phosphate saline buffer (adjusted to pH7.0 with NaOH solution). Prior to CV measurements, the coated electrodes were immersed in the test solutions for 15 min to ensure equilibration, and then the voltammogram was recorded during the first CV scan. All the measurements, unless specified otherwise, are the average of at least three replicate measurements. The redox formal potential, E°′, was taken as the average between the anodic and cathodic peak potentials, (Epa + Epc)/ 2. The peak-peak separation, ∆Ep, was taken as (Epa - Epc). Plots of peak currents (ip) vs the square root of the scan rate (V1/2) were found to be linear with R g 0.990 in all of our
1592 J. Phys. Chem. B, Vol. 111, No. 7, 2007 measurements, as expected for processes controlled by diffusion in the hydrogel system, which is consistent with the results reported by Crumbliss et al.25 for CAR hydrogel coated platinum electrodes. The experimentally observed apparent diffusion coefficients (Dapp) of MB were estimated from the slopes of the plots of ipa vs V1/2.25-27 As for the concentration of MB in MC or CAR hydrogels, it was determined as follows. The concentration of MB in the κ-CAR hydrogel was determined according to the reported method.25 And the same method was modified to obtain the concentration of MB in the MC gel. The MC coated GC electrode was weighed after being swelled in the test solution at 45 °C. The variation in weight between the coated and bare electrodes represents the weight of the hydrogel; thus, the volume of the contained water was known. To obtain the concentration of MB in the hydrogel, the electrode was then immersed in cold water in a refrigerator to completely dissolve the coated hydrogel, and then the obtained solution was diluted to a certain volume. On the basis of the absorbance of this solution that was measured at a wavelength of 664 nm for MB by UV-vis spectrometry, the concentration of MB in the hydrogel was estimated. 4. Rheological and Calorimeric Measurements. Rheological measurements were carried out on a Haake Rheo Stress RS600 rheometer (Germany). Cone-plate geometry of C60/2Ti was used with a 60 mm diameter, 2° angle, and a fixed gap of 0.105 mm. The shear storage modulus G′ and loss molulus G′′ of 1 wt % MC were measured as a function of temperature at an angular frequency of 1 rad/s within the linear viscoelastic regime. To prevent dehydration, a thin layer of low-viscosity silicone oil was placed on the periphery surface of the solution during measurements. A Setaram micro-differential scanning calorimeter (Caluire, France) was used to determine the thermal properties of 1 wt % MC solution during heating at a rate of 1 °C/min. Results and Discussion 1. Electrochemistry of MB at GC/MC. MC is a soluble cellulose derivative substituted by methyl groups, as shown in Figure 1a. Concentrated and semidilute MC aqueous solutions can form thermoreversible hydrogels upon heating.22,28 It is widely accepted that MC dissolved in water at low temperatures (
