Sliding Friction of Zwitterionic Hydrogel and Its Electrostatic Origin

Polyzwitterionic materials, which have both cationic and anionic groups in each repeating unit of polymer, show excellent antibiofouling properties. I...
1 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/Macromolecules

Sliding Friction of Zwitterionic Hydrogel and Its Electrostatic Origin Jamil Ahmed,† Honglei Guo,† Tetsurou Yamamoto,† Takayuki Kurokawa,‡ Masakazu Takahata,§ Tasuku Nakajima,‡ and Jian Ping Gong*,‡ †

Graduate School of Life Science, Hokkaido University, Sapporo, 060-0810, Japan Faculty of Advanced Life Science, Hokkaido University, Sapporo, 060-0810, Japan § Faculty of Science, Hokkaido University, Sapporo, 060-0810, Japan ‡

S Supporting Information *

ABSTRACT: Polyzwitterionic materials, which have both cationic and anionic groups in each repeating unit of polymer, show excellent antibiofouling properties. In this study, the surface friction of carboxybetaine type zwitterionic hydrogels, poly(N-(carboxymethyl)-N,N-dimethyl-2-(methacryloyloxy)ethanaminium, inner salt) (PCDME), against glass substrates were investigated in aqueous solutions. The friction measurement was performed using a rheometer with parallel plate geometry and the sliding interface was monitored during the measurement. The frictional stress on glass was high in water and it showed weak dependence on pressure as long as the two sliding surfaces were in complete contact. The results performed in solutions with varied ionic strength revealed that the high friction on glass substrates has an electrostatic origin. The electrostatic potential measurement revealed that the PCDME gels have an isoelectric point at pH 8.5. Since the glass substrates carrying negative charges in pure water, the gel and the glass have electrostatic attraction in water. Study on the effect of pH has shown that below pH 8.5, attraction between the positively charged gels and negatively charged glass gives high friction, while above pH 8.5, the electrical double layer repulsion between two negatively charged surfaces gives low friction. From these results, it is concluded that although the PCDME gels behave like neutral gels in the bulk properties, their surface properties sensitively change with pH and ionic strength of the medium.

■

INTRODUCTION Polyzwitterionic materials, which have both cationic and anionic groups in each repeating unit of polymer, are drawing great attention as biomaterials due to their excellent antibiofouling properties and biocompatibility.1−10 The antibiofouling materials that prevent the attachment of bioorganisms in wet environment are in a great demand for advanced technologies. Recently a tough double network (DN) hydrogel has been developed by using poly(2-acrylamido-2methylpropanesulfonic acid) (PAMPS) as the first network and a carboxybetaine type zwitterion poly(N-(carboxymethyl)-N,Ndimethyl-2-(methacryloyloxy)ethanaminium, inner salt) (PCDME) as the second network.11 Comparing to the conventional double network (DN) hydrogels that use neutral polyacrylamide or poly(dimethyl acrylamide) as the second network,12 the DN hydrogel using the zwitterionic polymer as the second network exhibits excellent antibiofouling properties in addition to high mechanical strength and toughness.11 This promises a great potential of the zwitterion-based hydrogels as better candidate for bioapplication, for example, as coating materials of low-friction biomedical devices and implants. Understanding the surface frictional properties of the zwitterion hydrogels is indispensable for these potential applications. As the water-swollen hydrogels have many common features to that of internal organs in our body, the sliding friction of © 2014 American Chemical Society

hydrogels in aqueous solution has recently been drawing great scientific attention.13−24 It has been revealed that the friction of hydrogels is strongly dependent on the interfacial interaction between the hydrogels and the counter surfaces in liquid, and exhibits a wide range of frictional coefficients ranging from the order of 0.001 to 10.25−27 Previous studies have revealed that the friction of hydrogels from common polyelectrolyte (carrying the charges of same sign) in water is strongly dependent on the charge of the counter surfaces.28−30 When the counter surface has like charges with the hydrogel, the friction shows a monotonous increase with the sliding velocity. This friction is referred to the hydrated lubrication of a thin water layer formed at the interface due to osmotic repulsion of the overlapping electrical double layers of the charged surfaces. When the counter surface has opposite charges with that of the hydrogel, a very high friction is observed. The high friction is referred to the elastic deformation of the adsorbed polymer chains on the oppositely charged surface. The former is called as repulsive interaction, while the latter as attractive interaction. Received: February 20, 2014 Revised: April 14, 2014 Published: April 18, 2014 3101

dx.doi.org/10.1021/ma500382y | Macromolecules 2014, 47, 3101−3107

Macromolecules

Article

Samples. Synthesis of Hydrogels. PCDME gels were prepared by radical polymerization initiated by UV irradiation.11 Gels with different water content and Young’s modulus were prepared by varying the cross-linker density CMBAA. Briefly, aqueous solutions containing 3 M monomer (CDME), 0.1 and 1.0 mol % cross-linker (MBAA) and 0.1 mol % initiator (2-oxoglutaric acid) were prepared (the mol % is in relative to the monomer concentration). The precursor solution was poured into a reaction cell consisted of two parallel glass plates separated by a silicone spacer of 2 mm in thickness. The reaction cell was irradiated by UV (365 nm) from both sides for 8 h under argon atmosphere. After polymerization, the glass plates were removed and the obtained sheet-shape gels were immersed in a large amount of water for sufficiently long time to reach swelling equilibrium and to wash away the residual chemicals. The sample thickness after swelling was 4.0 and 3.1 mm for 0.1 and 1.0 mol % cross-linker concentrations, respectively. Extreme care was taken during the synthesis to obtain a microscopically smooth surface of the gel. Counter Surface of Friction Test. Cover glasses (micro cover glass, C050701, Matsunami Glass Ind. Ltd., Japan) were used as counter surfaces for friction test. The water contact angle θw of the untreated cover glasses (droplet size =2 μL) was around 45°, as determined by contact angle meter (Drop Master 300, Kyowa Interface Science Co., Ltd., Japan). To increase the hydrophilicity of the cover glasses, they were soaked with alkali solution for 24 h, then washed and dried with air blow. After this treatment, the contact angle of glass became 25° due to the maximum exposure of SiO− groups on the glass surface. Measurements. The pH of the solutions that the gel samples were immersed was measured by pH meter (Orion 5-star, Thermo scientific). For the experiments to measure the electrostatic potential of the hydrogel surface and frictional stress at different pH, the ionic strengths of all the solutions were kept constant (0.1 M) with additional NaCl. The degree of swelling of the gels, q, defined as q = (swollen weight)/(dry weight), was calculated by obtaining the respective weights of the samples using moisture balance (MOC-120H, Shimadzu Co.). The dry samples were obtained by heating the gels at 120 °C until they reached a constant weight. The Young’s moduli E of gels were calculated from the slope of tensile stress−strain curves measured by using a tensile tester (Instron 5965, Instron Co.). The tensile tests were carried out with gel samples cut into a standard dumbbell shape (12 mm in length and 2 mm in width). The tensile strain rate was 10% length/min. The mesh size of the network ξ is estimated from E = 3kBT/ξ3.31 The hydrogel samples are summarized in Table 1.

A zwitterionic hydrogel contains both positive and negative charges on each repeating unit and the opposite charges form inner salt. As a result, the zwitterionic hydrogel usually behaves like a neutral gel, showing little dependence of their swelling property on ionic strength of the medium, in contrast to the behaviors of common polyelectrolyte hydrogels that their physical properties are strongly dependent on the ionic strength of the liquid medium. It has been observed that single network PCDME gels (Scheme 1) keep the same swelling degree q and Scheme 1. Chemical Structures of Poly(N-(carboxymethyl)N,N-dimethyl-2-methacryloyloxy)ethanaminium, inner salt) (PCDME) Hydrogel Cross-Linked by N,N′Methylenebis(acrylamide) (MBAA)

elastic modulus E over a wide range of ionic strength.11 Furthermore, the q and E of the PCDME hydrogel hardly change with pH until below pH 3.11 This suggests that, the carboxyl groups of this carboxybetaine type zwitterion are in fully ionic form above pH 3. In this paper, we investigate the frictional behavior of PCDME hydrogels against glass substrates in varied ionic strength and pH medium. We used a homemade optical system to in situ monitor the friction interface, which gives information on the evolution of gel contact to the glass during sliding motion. Taking the advantage that the PCDME gels do not change its swelling properties and elastic modulus over a wide range of ionic strength and pH, we can discuss the sliding friction mechanism only in terms of the interfacial molecular interaction between the gel and the counter surface in the varied medium. For this purpose, we performed the friction measurement at the condition of full contact of the gel to the glass, and no bulk liquid was trapped at the interface. This study on frictional behavior of biocompatible zwitterionic hydrogels might be helpful for understanding and finding promising approaches in designing of low-friction biomedical devices and implants.

■

Table 1. Synthesis Compositions, Swelling Ratio q, Elastic Modulus E, Network Mesh Size ξ, and Sample Thickness t of PCDME Hydrogels at Equilibrium Swelling State in Pure Water sample code

CCDME (M)

CMBAA (mol %)

q (w/w)

E (kPa)

mesh size ξ (nm)a

thickness t (mm)

S3-0.1 S3-1

3.0

0.1 1.0

20.1 7.1

5.4 120.0

13.2 4.6

4.0 3.1

a

Mesh size of the network was estimated from E = 3kBT/ξ 3.

The surface roughness of cover glasses was several nm.30 The surface roughness of hydrogels prepared on smooth glass surface was less than 30 nm.32 The surface of gel samples was also examined by 3D LASER microscope (Keyence, VK-9710), which performs laser scanning in the plane parallel to sample surface at various sample depth in an area 1 mm ×1 mm. No roughness at a level of 1.0 μm could be identified over the scanned area, confirming that the whole gel was very flat (Supporting Information Figure S1a,b). The electrostatic potential of the hydrogels was measured by applying the neurophysiological recording techniques.33 Measurements were performed using a microelectrode that consisted of a reversible silver/silver chloride electrode (Ag|AgCl) inserted in a glass

EXPERIMENTAL SECTION

Materials. N-(Carboxymethyl)-N,N-dimethyl-2(methacryloyloxy)ethanaminium, inner salt (CDME), a courtesy from Osaka Organic Chemical Industry, Ltd., Japan, was used as received. 2-Oxoglutaric acid (Wako Pure Chemical Industries, Ltd.), Sodium hydroxide (NaOH, Wako Pure Chemical Industries, Ltd.), Hydrochloric acid (HCl, Wako Pure Chemical Industries, Ltd.) and Sodium chloride (NaCl, Wako Pure Chemical Industries, Ltd.) were used as received. N,N′-Methylenebis(acrylamide) (MBAA, Wako Pure Chemical Industries, Ltd., Japan) was recrystallized from ethanol. 3102

dx.doi.org/10.1021/ma500382y | Macromolecules 2014, 47, 3101−3107

Macromolecules

Article

tube with a tip diameter θcwater. So, from an angle θr between these two critical angles, that is, θcwater < θr < θcgel, one can observe the contact interface. A bright image of gel is observed when it is in contact with the substrate, while a black image is observed when a water film exists at the interface. Using this principle, a prism was attached to the upper plate of the rheometer, and below the prism, the cover glass as counter surface of friction was attached. Immersion oil was used to fill the air gap between the prism and glass substrate and to keep the refractive index of the whole system nearly constant. The images of the frictional interface were recorded by a digital video camera (HDR CX550, Sony). The video camera was located 1.5 m away from the friction specimen so that the whole specimen could be observed at the angles θr that satisfy the condition θcwater < θr < θcgel. The frictional interface was irradiated with white light to obtain clear photographic images. The angle θr was experimentally determined by gradually increasing the observation angle up to a position that the water image in the front region of the sample disappeared and gel image was still observable. The zoom function of camera was used to get close images. Adjustments of focus and exposure were set in automatic mode when recording the real time video and the raw images were extracted from the video later on. The disk-shaped specimen was observed as an ellipse shape in the raw images due to observation from a large angle (about 62°) to the sample normal surface. The ellipse shape was converted back to the circle shape by processing the raw images with photo editing software, Photoshop CS 8.0. Our observation experience tells that this optical system could only identify the existence of water film of thickness larger than decades of nanometers.

Scheme 2. Graphical Representation of Stepwise Velocity Increase and the Frictional Torque Measured by Rheometer

■

rheometer rotated the gel at an angular velocity (ω) that was increased stepwise from 101 to 103 rad/s, each lasted for 40 s, without separation of the two surfaces. At each velocity, the average torque (T) of the last 20 s was adopted as friction torque. In this measurement geometry, the sliding velocity varied along the radius of the sample. For simplicity, we assume that the frictional stress is linearly proportional to the sliding velocity, then the overall frictional force F is related to the torque by F = 4T/3R,27 where R (=7.5 mm) is the radius of the diskshaped gel. The average frictional force per unit area, that is, the average frictional stress, σ, is defined as σ = F/πR2.27 In the case of positive dependence of the frictional stress on the linear velocity, the overall frictional torque was dominated by the region of large radius. So we adopted the linear velocity as the value at the outermost part of the disk-shaped gel, that is, v = ωR. In Situ Observation of Frictional Interface. The contact of the hydrogel on the glass surface was observed during the sliding friction using a homemade system recently developed.34 The observation was based on the critical refraction principle as shown in Scheme 3. As the glass has a higher refractive index than that of a hydrogel, the light from the gel side refracts at angles less than the critical refraction angle

RESULTS AND DISCUSSION Effect of Initial Contact. Our previous studies have revealed that when a soft hydrogel contacts on a solid surface in liquid, some liquid drop may trapped at the interface to form partial contact in the macroscopic scale. This macroscopic contact condition substantially influences the frictional behavior of the gel.34,35 So, first of all, the contact between PCDME hydrogels and cover glass was observed prior to and during frictional measurement in water (pH 6.8). The irregular bright images shown in the first column of Figure 1a indicates that at low normal pressure, the disk-shape gel (sample S3-1, Young’s modulus E = 120 kPa) does not form full contact with the counter surface and in some region water film exists at the interface prior to sliding motion. The initial contact area increases with the increase of the normal pressure. Considering the partial macro contact of the gel to the counter surface in water, the total friction stress σ comes from 3103

dx.doi.org/10.1021/ma500382y | Macromolecules 2014, 47, 3101−3107

Macromolecules

Article

change in the initial contact area and the amount of water entrapped at the interface prior to the sliding motion. In order to justify that the gel is practically in contact with the glass at the low velocity end (v = 1.7 × 10−4 m/s) where bright images are observed in Figure 1a, we discuss the order of magnitudes of the frictional stress. When the gel is adhered on the counter surface, the frictional stress σE is from two contributions; the elastic term σel from elastic stretching of the adsorbed polymer chain, and the viscous term σvis from the lubricating resistance of the hydrated fluidic phase of the hydrogel,26 σE = ϕmicroσel + (1 − ϕmicro)σvis

Figure 1. Effect of partial contact on the sliding friction of PCDME hydrogel (S3-1) on untreated glass (θw = 45°) in water (pH 6.8): (a) contact images; (b) frictional stress. Dotted circles in part a represent the overall area of the disk-shaped gel. The bright regions are in contact with the glass substrate and the dark regions are trapped with water.

where ϕmicro is the microscopic contact area ratio and it is roughly the surface volume fraction of polymer. The term σvis has the lubricating layer thickness in the order of mesh size ξ of the polymer network. For the friction geometry used in the present study, σvis = 2ηv/3ξ. To justify that the first adsorption term plays the predominant role in the low velocity end, we estimate the magnitude of the second term at v = 1.7 × 10−4 m/s. Taking ξ = 5 nm, ϕmicro = 1/q = 0.14 from Table 1, and a fluid viscosity η = 0.001 Pa·s, we find a stress (1 − ϕmicro)σvis = 20 Pa by using eq 3. This stress is about 1−2 orders lower in magnitude than the observation of Figure 1b. Thus, it is justified that the gel in bright region is in contact with the glass and the friction is from the elastic deformation of the adsorbing polymer chains. In order to discuss the molecular mechanism of friction, we need to avoid the macroscopic contact effect caused by water entrapment, and perform friction measurement with a full initial contact. For this purpose, we use a more soft gel (sample S3-0.1, E = 5.4 kPa) that can form complete contact by more large deformation at the same normal pressure range with that in Figure 1. Starting from full contact (ϕmacro = 1), the contact area does not decrease with the velocity increase (Figure 2a).

two contributions, the elastic friction from contact region σE and the hydrodynamic lubrication σH from the noncontact region. σ = ϕmacroσE + (1 − ϕmacro)σH

(1)

Here, ϕmacro is the macro contact area ratio. As shown in Figure 1a, ϕmacro is strongly velocity-dependent. At a constant normal pressure, the contact area decreases as the sliding velocity increases (Figure 1a), and this leads to a decrease in the friction stress (Figure 1b). This shearweakening results clearly indicates an elasto-hydrodynamic transition, in consistent with our previous observation on the polyacrylamide (PAAm) hydrogels sliding on glass substrate.34,35 The water film pre-existed at the interface spreads above a certain velocity by the forced wetting mechanism, and induces this elasto-hydrodynamic transition.36 Figure 1b shows that the shear-weakening effect due to loss of contact is especially prominent in the case of low pressure. At the high velocity end of 10−2 m/s, the gel no longer is in contact with glass for low pressures (P = 0.55 kPa, 1.1 kPa), and a macroscopic continuum water film separates the gel from the glass. In this case, ϕmacro = 0, and the frictional stress is only from the hydrodynamic lubrication, σ = σH, giving very low values, as shown in Figure 1b. We can estimate the water film thickness h when the gel lost the whole contact. By assuming a homogeneous film thickness h over the sample, the frictional force F due to hydrodynamic lubrication is F=

∫0

R

⎛ 2πηωR3 ⎞ ⎛ ηωr ⎞ ⎜ ⎟2πr dr = ⎜ ⎟ ⎝ h ⎠ ⎝ 3h ⎠

(4)

Figure 2. Effect of full contact on the sliding friction of PCDME hydrogel (S3-0.1) on untreated glass (θw = 45°) in water (pH 6.8): (a) contact images; (b) frictional stress. The black dashed line in part b represents the microscopic viscous resistance (1 − ϕmicro)σvis in eq 4

(2)

Here, η is the viscosity of water. As the frictional stress σ is estimated from σ = F/πR2, and the velocity v from ν = ωR in Figure 1b, h is given by 2ηv h= (3) 3σ

This confirms that there is no bulk water invasion to the sliding interface from outside. As the hydrogel and glass substrate are flat enough, the friction of hydrogels at full contact comes from the elastic resistance, that is, a collective strength of deformation of adsorbed polymer chains. Figure 2b shows that the friction modestly decreases and then increases with the increase of the velocity. This velocity dependence should be related to the change of the adsorption dynamics. The shear-weakening in friction is apparently due to the decrease of collective adsorption, while the shear-hardening

From Figure 1b, we found that h is 0.67 and 6.7 μm at v = 10−2 m/s for pressure 1.1 and 0.55 kPa, respectively. This result indicates that an abundant of bulk water is trapped at the soft interface at low pressure. As shown by Figure 1b, the frictional stress increases with increasing normal pressure, and the shearweakening effect becomes less prominent at high pressure. The pressure effect on the friction stress is apparently due to the 3104

dx.doi.org/10.1021/ma500382y | Macromolecules 2014, 47, 3101−3107

Macromolecules

Article

may come from two effects. One is the increased stretching of adsorbing chain with the velocity increase. Another is the increased viscous resistance with velocity increase, as the viscous term is proportional to velocity. In order to clarify these two effects, we estimated the magnitude of the viscous term of eq 4 using ξ = 13 nm, ϕmicro = 0.05 (Table 1), and the result is shown as dashed line in Figure 2b. At the high velocity end, the viscous term is not negligible but still less than the observation. This result indicates that the chain stretching effect is responsible for the shear-hardening in the observed velocity range. Regardless of the change in the frictional stress with the sliding velocity in Figure 2b, the bright images in Figure 2a hardly change. This indicates that the present optical method could not identify changes in the nanoscale contact. In contrast to Figure 1b, which shows strong pressure dependence of friction, the friction is only weakly dependent on the normal pressure, as shown in Figure 2b. So the strength of the polymer adsorption is only weakly influenced by the pressure applied. These results are also in agreement with our observation for PAAm−glass system.34 Thus, the normal pressure strongly influences the friction by affecting the initial contact and the amount of liquid entrapping at the interface. In the case of full contact with no bulk water trapped at the interface, friction is due to molecular interaction between the gel and the counter surface. In the following part of this study, we performed all the friction measurements at the condition of full initial macro contact (ϕmacro = 1) using sample S3-0.1. Origin of Interfacial Interaction. What is the origin of the molecular interaction between the PCDME hydrogel and the glass substrate? Our previous study has shown that the swelling degree q and elastic modulus E of PCDME hydrogels remain unchanged over a wide range of ionic strength from 10−4 M to 3 M.11 Furthermore, we confirmed in this study that q and E of the PCDME hydrogels, modulated by the cross-linker density CMBAA (mol %), following a scaling relationship of E ∼ q−2.5, which is characteristic for neutral hydrogels.37 These results demonstrate that the bulk PCDME hydrogels behave like typical neutral gels, and the quaternary ammonium moiety and the carboxylate moiety on the PCDME are balanced to give zero net charge. However, when the PCDME gels are in contact with the glass surface that is negatively charged in water,38 the positively charged quaternary ammonium moiety may have attractive interaction with glass surface, especially when we take consider the weak acidic nature of the carboxylate moiety. If the ionic attraction between the gel and the glass exist, it should be shielded at high ionic strength. So the friction of gels swelled in different concentrations of NaCl solutions was measured, and the results are shown in Figure 3a. In this study, the hydrophilic glass with water contact angle θw = 25° was used. Figure 3a reveals that the friction of PCDME gel decreases with the increase of ionic strength. Since the bulk properties of the gel, q and E, maintain the same values with the change in the ionic strength, the decrease in friction is due to the change in the interfacial interaction between the gel and the substrate, that is, the shielding of the electrostatic attraction at high ionic strength. Thus, the adsorption of PCDME to glass has the electrostatic origin. This means that although the bulk gels show the neutral behavior, the surface of the gel has net positive charges, due to protonation of carboxylic group. When the gel approaches the glass surface, the negative charges on the glass

Figure 3. (a) Velocity dependence of frictional stress for PCDME (S3−0.1) hydrogel sliding on hydrophilic glass (θw = 25°) in solution of various NaCl concentrations. (b) Frictional stress vs concentrations of NaCl solution at sliding velocity of 2.6 × 10−4 m/s. Normal pressure: 2.75 kPa. The friction tests were performed at full interfacial contact.

surface may induce this protonation and thus stabilize the electrostatic attraction between the negative charges of glass and the positive charges of the gel. As shown in Figure 3a, the friction curve shifts approximately to low values with the increase in the NaCl concentration over the whole velocity range. This phenomenon is in agreement with our explanation on the shear-weakening and shear-hardening phenomenon in Figure 2b. That is, the friction in the whole velocity range is dominated by the elastic mechanism, which decreases by salt screening. As shown in Figure 3b, we observed a power law relation between the frictional stress σ and the salt concentration CNaCl, σ ∼ CNaCl−0.2. The exponent of this power law relation is smaller than the value of 0.5 that is known for screening of electrostatic interaction. As the protonation equilibrium of carboxylic group is also influenced by the ionic strength of the medium, the screening effect may be weakened by this effect. Surface Potential of PCDME Gels. Being weak acid in nature, the equilibrium of the proton transfer of the carboxyl group shown in Scheme 4 depends on the pH of solution. So Scheme 4. Protonation Association/Dissociation Equilibrium of PCDME Depending on pH

the behavior of zwitterionic PCDME gel should dependent on pH of the solution with which the gel is swelled and the friction is performed. As the elastic modulus and swelling degree q of the bulk PCDME gels do not change significantly over a wide range of pH from 3 to 12, while q dramatically increases below pH 2, the pKa of the carboxyl group in bulk PCDME gels is around pH 2, above which the protonation is almost suppressed. In order to investigate the pH dependence of the PCDME gels, we measured the electrostatic potential ψgel of the gels swelled in different pH with respect to the bulk solution. When there are net fixed charges on the polymer network, a concentration difference in mobile ions will be built inside and outside of the gel by the Donnan equilibrium, which gives to a finite electrostatic potential ψgel. Figure 4 shows that at low pH, the electrostatic potential of the gel has positive value and it gradually approaches to zero at pH 8.5 and then to a little negative value as the pH is increased. 3105

dx.doi.org/10.1021/ma500382y | Macromolecules 2014, 47, 3101−3107

Macromolecules

Article

Figure 5. Effect of pH of the medium on friction of PCDME (S3-0.1) gel against untreated glass surface (θw = 45°). Ionic strength of the medium: 0.1 M. Normal pressure: 2.75 kPa.

Figure 4. Effect of pH on the surface potential of PCDME gel (S30.1). Ionic strength of the medium: 0.1 M. Temperature: 25 °C. The isoelectric point of the gel was found at around pH = 8.5. For comparison, the volume change V/V0 of the PCDME gel is also shown. Here, V is the volume of the gel at the respective pH and V0 is the volume at pH 6.8. The volume change result was reproduced with the permission from ref 11.

Above the isoelectric point, shear-hardening is observed, and this behavior is characteristic for hydrated lubrication, due to electric double layer repulsion between the two like charged surfaces.28 At 0.1 M ionic strength, the Debye length is about 1 nm. A thin diffuse double layer of 1 nm gives a lubricating resistance in the order of 102 Pa at the low velocity limit of 1.7 × 10−4 m/s, which is in agreement with the observation for pH 12.7. To explain these phenomena more quantitatively, we consider the pH dependence of electrostatic potential of glass surface. Lameiras et al. showed that lowering the pH of the solution pushes the ζ potential of glass toward zero (Figure 6a).38 The electrostatic interaction between gel and glass

This result indicates that the gel has an isoelectric point at around pH = 8.5, and it has net positive charges below pH 8.5. This is associated with the weak acid nature of the carboxylate groups that slightly take the protons from the solution to give net positive charge of polymer. But at high pH all the carboxylate groups are in dissociated state, and some of the quaternary ammonium ions are associated with the OH− to give net negative charges (Scheme 4). As shown in Figure 4, the relative volume V/V0 of the gel increases as the absolute value of ψgel, either toward positive or toward negative, increases due to polyelectrolyte effect. The result in Figure 4 directly confirmed that the PCDME gel has net positive charges in water (pH 6.8). So its high frictional stress against glass is due to the adhesion of the gel on the glass surface that is negatively charged (Scheme 5). As Scheme 5. Molecular Interaction between the Zwitterion Hydrogel and the Glass Surface

Figure 6. (a) Variation of potential of glass (ψglass) and PCDME (S30.1) hydrogel (ψgel) with different pH. (b) Product of potentials of these two surfaces and it is correlation with frictional stress over different pH. Frictional stress at low velocity limit (1.7 × 10−4 m/s) was adopted. ζ potential of glass (ψglass) at different pH was reproduced with permission from ref 38. Copyright 2008 Materials Research.

discussed in the previous section, the protonation of the carboxylate groups may be enhanced by the ionic bond formation between the quaternary ammonium ion of the gel and the −SO− ion of the glass. So the net charge density of the gel may higher than that measured in water. pH Dependence of Friction. It could be interesting to observe the frictional behavior of PCDME gel at different pH and correlate with the surface potential. Figure 5 shows the velocity dependence of friction at 7 different pH values with a constant ionic strength of 0.1 M, by adding required amount of NaCl. There was no change in frictional stress as the pH was increased from 1.9 to 6.63, but the frictional stress tends to decrease with further increase of pH. Furthermore, the velocity dependences of friction below and above the isoelectric point (pH 8.3) of the gels are quite different. Below the isoelectric point, shear-weakening is observed, and this behavior is characteristic for elastic friction, due to the attraction between the oppositely charged surfaces.34

surface can be approximated as the product of their surface potential ψgelψglass. Using the ζ potential of glass from literature as ψglass and the measured ψgel of PCDME hydrogel, Figure 6b shows the correlation between ψgelψglass and frictional stress as a function of pH. At low pH, the value of ψgelψglass is negative, which means there is an attraction between the sliding surfaces, and the friction is high. As the pH increases, ψgelψglass starts to increase and crosses zero at pH 8.5. Beyond this pH, the two surfaces are in repulsion and the friction starts to decrease. This result confirms again that the friction of zwitterionic PCDME gels on glass substrates has the electrostatic origin, as long as the surfaces remain microscopically smooth and are in complete contact. Previous work has shown that the electrostatic interaction, and thereby the friction, between a gel and a substrate can be tuned by applying electric field.39 Owing to the weak acid nature of the zwitterionic PCDME hydrogel, it is possible to switch the friction between high value of attraction and low 3106

dx.doi.org/10.1021/ma500382y | Macromolecules 2014, 47, 3101−3107

Macromolecules

Article

(9) Yamauchi, J.; Masuhara, E.; Kadoma, Y.,; Nakabayashi, N. Japanese Patent 54063025, 1977. (10) Kadoma, Y.; Nakabayashi, N.; Masuhara, E.; Yamauchi, J. Kobunshi Ronbunshu 1978, 35, 423. (11) Yin, H.; Akasaki, T.; Sun, T. L.; Nakajima, T.; Kurokawa, T.; Nonoyama, T.; Taira, T.; Saruwatari, Y.; Gong, J. P. J. Mater. Chem. B 2013, 1, 3685. (12) Gong, J. P.; Katsuyama, Y.; Kurokawa, T.; Osada, Y. Adv. Mater. 2003, 15, 1155. (13) Sokoloff, J. B. Soft Matter 2012, 8, 8164. (14) Nakano, Y.; Kurokawa, T.; Du, M.; Liu, J.; Tominaga, T.; Osada, Y.; Gong, J. P. Macromolecules 2011, 44 (22), 8908. (15) Kamada, K.; Furukawa, H.; Kurokawa, T.; Tada, T.; Tominaga, T.; Nakano, Y.; Gong, J. P. Special Issue: International Symposium on Non-Equilibrium, Soft Matter. J. Phys.: Condens. Matter 2010, 23, 284107. (16) Gong, J. P. Soft Matter 2006, 2, 544. (17) Gong, J. P.; Iwasaki, Y.; Osada, Y.; Kurihara, K.; Hamai, H. J. Phys. Chem. B 1999, 103, 6001. (18) Gong, J. P.; Kurokawa, T.; Narita, T.; Kagata, G.; Nishimura, G.; Kinjo, M.; Osada, Y. J. Am. Chem. Soc. 2001, 123, 5582. (19) Kagata, G.; Gong, J. P.; Osada, Y. J. Phys. Chem. B 2002, 106, 4596. (20) Baumberger, T.; Caroli, C.; Ronsin, O. Phys. Rev. Lett. 2002, 88, 75509. (21) Baumberger, T.; Caroli, C.; Ronsin, O. Eur. Phys. J. E 2003, 11, 85. (22) Nitta, T.; Kato, H.; Haga, H.; Nemoto, K.; Kawabata, K. J. Phys. Soc. Jpn. 2005, 74, 2875. (23) Tominaga, T.; Takedomi, N.; Biederman, H.; Furukawa, H.; Osada, Y.; Gong, J. P. Soft Matter 2008, 4, 1033. (24) Yamaguchi, T.; Ohmata, S.; Doi, M. J. Phys.: Condens. Matter 2009, 21, 205105. (25) Gong, J. P.; Higa, M.; Iwasaki, Y.; Katsuyama, Y.; Osada, Y. J. Phys. Chem. B 1997, 101, 5487. (26) Gong, J. P.; Osada, Y. J. Chem. Phys. 1998, 109, 8062. (27) Gong, J. P.; Kagata, G.; Osada, G. J. Phys. Chem. B 1999, 103, 6007. (28) Oogaki, S.; Kagata, G.; Kurokawa, T.; Osada, Y.; Gong, J. P. Soft Matter 2009, 5, 1879. (29) Gong, J. P.; Iwasaki, Y.; Osada, Y.; Kurihara, K.; Hamai, H. J. Phys. Chem. B 1999, 103, 6001. (30) Tominaga, T.; Kurokawa, T.; Furukawa, H.; Osada, Y.; Gong, J. P. Soft Matter 2008, 4, 1645. (31) Flory, P. J. Principles of Polymer Chemistry; Cornell University Press: Ithaca, NY, 1953. (32) Kurokawa, T.; Gong, J. P.; Osada, Y. Macromolecules 2002, 35, 8161. (33) Purves, R. D. Microelectrode Methods for Intracellular Recording and Ionophoresis; Academic Press: London, 1981. (34) Yamamoto, T.; Kurokawa, T.; Ahmed, J.; Kamita, G.; Yashima, S.; Furukawa, Y.; Ota, Y.; Furukawa, H.; Gong, J. P. Soft Matter 2014, under revision. (35) Yashima, S.; Takase, N.; Kurokawa, T.; Gong, J. P. Soft Matter 2014, DOI: 10.1039/C3SM52883A. (36) Gennes, P. G. D.; Wyart, F. B.; Quere, D. Capillarity and Wetting Phenomena: Drops, Bubbles, Pearls, Waves; Springer: New York, 2003. (37) Obukhov, S. P.; Rubinstein, M.; Colby, R. H. Macromolecules 1994, 27, 3191. (38) Lameiras, F. S.; Souza, A. L. D.; Melo, V. A. R. D.; Nunes, E. H. M.; Braga, I. D. Mater. Res. 2008, 11 (2), 217. (39) Takata, M.; Yamaguchi, T.; Gong, J. P.; Doi, M. J. Phys. Soc. Jpn. 2009, 78 (8), 084602.

value of repulsion with the same pair of samples by adjusting the pH and ionic strength of the swelling solution while keeping the bulk properties of the hydrogel unchanged. From the viewpoint of fundamental study on friction, this is a great advantage of PCDME gel over common polyelectrolyte hydrogels, because, the latter dramatically change their bulk properties (q, E) with ionic strength as well as pH, which makes it difficult for revealing the effect of electrostatic interaction on friction of polyelectrolyte gels. The adjustable tribological property along with the exceptional mechanical properties and the biocompatibility, the zwitterionic PCDME hydrogels are promising candidate for biological application.

■

CONCLUSIONS In summary, the carboxybetaine type zwitterionic hydrogel PCDME has an isoelectric point (IP) at around pH 8.5. The electrostatic potential of the gel depends on the relative amount of −COO− and −COOH, which is determined by the pH of the medium. Over a wide range from pH3 to pH 12, the gel is only slightly charged, being positive below IP and negative above IP. These changes in charged state hardly influence the bulk properties of the gels but dramatically change the sliding friction of the gels on negatively charged counter surface. The electrostatic attraction below IP leads to elastic friction while the electrostatic repulsion above IP gives lubrication. The obtained results give insight into understand the frictional properties of biological tissues that consist of biopolymers with both positive charges and negative charges, and the study is also informative for designing low-friction biomaterials.

■

ASSOCIATED CONTENT

S Supporting Information *

Figures containing 3D LASER microscopic images of hydrogel and glass surface. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*(J.P.G.) Telephone/Fax: +81-11-706-2635. E-mail: gong@ mail.sci.hokudai.ac.jp. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research was financially supported by a Grant-in-Aid for Scientific Research (S) (No. 124225006) from the Japan Society for the Promotion of Science (JSPS).

■

REFERENCES

(1) Zhou, J.; Shen, J.; Lin, S. C. Chem. J. Chin. Univ. 2002, 23, 2393. (2) Ai, F.; Yuan, Y. L.; Wu, Y. D.; Zhou, J.; Shen, J.; Lin, S. C. Acta Polym. Sin. 2002, 4, 535. (3) Yuan, Y. L.; Ai, F.; Zhang, J.; Zang, X. B.; Shen, J.; Lin, S. C. J. Biomat. Sci., Polym. Ed. 2002, 13, 1081. (4) Zhang, J.; Yuan, Y. L.; Wu, K. H.; Shen, J.; Lin, S. C. Colloids Surf., B: Biointerfaces 2003, 28, 1. (5) Yuan, J.; Mao, C.; Zhou, J.; Shen, J.; Lin, S. C.; Zhu, W.; Fang, J. L. Polym. Int. 2003, 52 (12), 1869. (6) Yuan, J.; Yuan, Y. L.; Shen, J.; Lin, S. C. Chem. J. Chin. Univ. 2003, 24, 916. (7) Yuan, J.; Yuan, Y. L.; Shen, J.; Lin, S. C.; Zhu, W.; Fang, J. L. Chin. J. Polym. Sci. 2003, 21, 419. (8) Lin, S. C. Chin. Polym. Bull. 1998, 1, 1. 3107

dx.doi.org/10.1021/ma500382y | Macromolecules 2014, 47, 3101−3107