Chain Length Dependence of Non-Surface Activity and Micellization

Mar 10, 2014 - The cationic and anionic amphiphilic diblock copolymers with a critical chain length and block ratio do not adsorb at the air/water int...
0 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/Langmuir

Chain Length Dependence of Non-Surface Activity and Micellization Behavior of Cationic Amphiphilic Diblock Copolymers Arjun Ghosh,† Shin-ichi Yusa,‡ Hideki Matsuoka,*,† and Yoshiyuki Saruwatari§ †

Department of Polymer Chemistry, Kyoto University, Katsura, Nishikyo-ku, Kyoto 615-8510, Japan Graduate School of Engineering, University of Hyogo, 2167 Shosha, Himeji, Hyogo 671-2280, Japan § Osaka Organic Chemical Industries Ltd., 7-20 Azuchi-Machi 1-Chome, Chuo-Ku, Osaka 541-0052, Japan ‡

ABSTRACT: The cationic and anionic amphiphilic diblock copolymers with a critical chain length and block ratio do not adsorb at the air/water interface but form micelles in solution, which is a phenomenon called “non-surface activity”. This is primarily due to the high charge density of the block copolymer, which creates a strong image charge effect at the air/water interface preventing adsorption. Very stable micelle formation in bulk solution could also play an important role in the non-surface activity. To further confirm these unique properties, we studied the adsorption and micellization behavior of cationic amphiphilic diblock copolymers of poly(n-butyl acrylate)-b-poly(3-(methacryloyloxy)ethyl)trimethylammonium chloride) (PBA-b-PDMC) with different molecular weights of hydrophobic blocks but with the same ionic block length. These block copolymers were successfully prepared via consecutive reversible addition−fragmentation chain transfer (RAFT) polymerization. The block copolymer with the shortest hydrophobic block length was surface-active; the solution showed surface tension reduction and foam formation. However, above the critical block ratio, the surface tension of the solution did not decrease with increasing polymer concentration, and there was no foam formation, indicating lack of surface activity. After addition of 0.1 M NaCl, stable foam formation and slight reduction of surface tension were observed, which is reminiscent of the electrostatic nature of the non-surface activity. Fluorescence and dynamic and static light scattering measurements showed that the copolymer with the shortest hydrophobic block did not form micelles, while the block copolymers formed spherical micelles having radii of 25−30 nm. These observations indicate that micelle formation is also important for non-surface activity. Upon addition of NaCl, cmc did not decrease but rather increased as observed for nonsurface-active block copolymers previously studied. The micelles formed were very stable, and their size decreased by only ∼5 nm after addition of 0.1 M NaCl.

■

neutralized carboxylic acid10 and strongly ionized quaternary pyridinium,11 sulfonic acid,12−14 and quaternary ammonium15 groups have also been studied. No reduction of the surface tension of the solution was observed, but micelle formation in aqueous solution was confirmed for all of these systems. Almost no adsorption was confirmed by foam formation observation and also directly by X-ray reflectivity. Before our first report on the finding of non-surface activity, no reduction of surface tension of the solutions of the ionic block copolymers was reported by Wagner et al., but they did not make a detailed discussion because their aim was emulsion polymerization.16 Amiel et al. reported that a strongly ionizable poly(tertbuty1styrene)-b-sodium poly(styrenesulfonate) (PtBS-b-PSS) block copolymer forms micelles in aqueous solution without adsorption, but after addition of salt, adsorption occurred at the silica surface.17 After our report on the detailed observation and examination on the non-surface activity, Müller et al. also demonstrated that the PBA-b-PAA block copolymers, which have a low glass transition temperature, were not adsorbed at

INTRODUCTION Amphiphilic molecules containing both hydrophobic and hydrophilic groups like common surfactants first adsorb at the air/water interface and form a Gibbs monolayer when dissolved in water.1 The adsorption induces the surface excess, which consequently lowers the surface tension. The surfactants in the bulk and the adsorbed molecules are in dynamic equilibrium. Ionic amphiphilic block copolymers should behave as a “macrosurfactant” since they contain both hydrophilic and hydrophobic blocks. Macrosurfactants have several applications including nanoreactors, separation science, vehicles for drug delivery, DNA carrier systems, emulsion stabilizer, and surface modifications (coating, paints, etc.).2−8 However, it has been reported that they do not adsorb at the air/water interface (hence, no reduction of surface tension), but they form micelles in bulk solution under suitable conditions that depend on the length and length ratio of the hydrophobic and hydrophilic blocks, the ionic strength, and the total molecular weight.9 This unique property has been called “non-surface activity”.10−16 The lack of surface adsorption of the anionic as well as cationic amphiphilic molecules with micellization is a completely new observation in the area of physical chemistry of surface and interfaces. Block copolymers with weakly ionized © 2014 American Chemical Society

Received: August 19, 2013 Revised: March 3, 2014 Published: March 10, 2014 3319

dx.doi.org/10.1021/la403042p | Langmuir 2014, 30, 3319−3328

Langmuir

Article

Scheme 1. Synthetic Scheme of the PBA-b-PDMC Diblock Copolymer

synthesized block copolymers having the same cationic (DMC) length but different nBA lengths. We studied systematically the fundamental properties of this novel cationic amphiphilic diblock copolymer, such as surface activity, foam formation ability, and also self-assembling behavior in solution. As a result, we clarified that these polymers also show non-surface activity when the hydrophobic block is long enough to form micelles in solution. Non-surface-active polymers became surface active after addition of salt. Furthermore, the addition of salt did not decrease the critical micelle concentration (cmc), which is a phenomenon observed with non-surface-active anionic block copolymers. These observations strongly support our view that the primary origin of the non-surface activity is the image charge and a stable micelle formation in bulk.

the air/water interface when the degree of dissociation increased, but micelles were formed in aqueous solution.18 Laschewsky et al. reported that ionic amphiphilic block copolymers showed transition from non-surface active to surface active in aqueous solution with decreasing hydrophobicity of the hydrophobic block. By increasing hydrophilic chain length and keeping a fixed hydrophobic chain length, the non-surface activity increases, due to an increase in charge density, which resulted in stronger image charge repulsion.19 Armes et al. showed a decrease in surface activity with the decrease of pH of the PDMAEMA-b-PMMA block copolymer due to an increase of the degree of ionization of the poly(DMAEMA) block.20−22 In our previous studies, transition from non-surface active to surface active was observed after addition of salt. Therefore, the primary origin of non-surface activity of the ionic block copolymer is thought to be an image charge repulsion23−26 at the air/water interface.10−15 Because of its hydrophobicity, the block copolymer should be adsorbed at the air/water interface. However, a strong electrostatic repulsion from the interface is induced by the image charge effect. In basic electromagnetism, an electric field generated by an electric charge near the interface between two media, which have a different dielectric constant, induces the formation of image charge effect.23−26 Because the hydrophilic chain consists of polyions, this electrostatic repulsion is so strong that the adsorbed polymers are largely destabilized. So far, we have investigated the origin of non-surface activity, which is a novel and very unique character, and we have examined its difference from normal, low molecular weight surfactant behavior with respect to the dependence of the critical micelle concentration (cmc) on the salt concentration and the hydrophobic chain length.10−15 A systematic investigation using other polymers having various molecular architecture would be needed to confirm the universality of this concept. The block copolymer has been observed to change from non-surface active to surface active when the hydrophobic block of the copolymer is elongated while maintaining a constant hydrophilic block length. Here, we also report the effect of shortening of hydrophobic block. In the past decade, the reversible addition−fragmentation chain transfer (RAFT) polymerization technique has been utilized to synthesize simple and complex block copolymers with a very narrow molecular weight distribution.27−29 Using the RAFT technique, we synthesized a poly(3(methacryloyloxy)ethyl)trimethylammonium chloride) (DMC) homopolymer with a well-defined molecular weight. Utilizing this polymer as a macro-chain-transfer agent (CTA), we extended the chain to a diblock copolymer having a hydrophobic n-butyl acrylate (nBA) segment. Poly(nBA) has a very low glass transition temperature Tg (−56 °C). We

■

EXPERIMENTAL SECTION

Materials. 4,4′-Azocyanovaleric acid (ACVA) 2,2′-azobis(isobutyronitrile) (AIBN), dimethylformamide (DMF), N-methylformamide (NMF), tetrahydrofuran (THF), n-butyl acrylate (BA), fluorescence probe (pyrene), and NaCl were purchased from Nacalai Tesque (Kyoto, Japan) and Wako Pure Chemicals (Osaka, Japan). 3(Methacryloyloxy)ethyl)trimethylammonium chloride (DMC) (Scheme 1 showing molecular structure) was a kind gift from Osaka Organic Chemicals Co. Ltd. AIBN, n-butyl acrylate, and pyrene were purified by standard methods. (4-Cyanopentanoic acid)-4-dithiobenzoate was used as chain transfer agent (CTA), and it was synthesized as reported.30,31 D2O was a product of Cambridge Isotope Laboratory, and CD3OD and CDCl3 were products of Euriso-top (France). Ultrapure water was used for sample preparation and dialysis, and it was obtained from a Milli-Q System (Millipore, Bedford, MA) with a resistivity of 18 MΩ cm. 1 H Nuclear Magnetic Resonance (1H NMR). A JEOL 400WS (JEOL, Japan) NMR spectrometer was used to obtain 1H NMR spectra. Deuterated water was used as a solvent for the macro-DMC and a deuterated methanol and chloroform mixture (CD3OD + CDCl3) for the block copolymer. Gel Permeation Chromatography. A JASCO system LC-2000 with a UV-2075 UV detector and a RI-2031 refractive index detector was used for GPC experiments with aqueous buffer as eluent. The GPC column used was Shodex OH pack (SB-804 HQ). The eluent was 0.3 M Na2SO4 and 0.5 M CH3COOH, pH 3, buffer solutions for PDMC, and the concentration of sample solution injected was ca. 2 mg/mL and the flow rate was 0.5 mL/min. The number-averaged molecular weight (Mn) and the polydispersity index (Mw/Mn) were determined for macro-DMC. Five standard poly(2-vinylpyridine) specimens were used for calibration. Surface Tension Measurements. The surface tension measurements for each block copolymer aqueous solution was carried out with a FACE CBVP-Z surface tensiometer (Kyowa Interface Science Co., Ltd. (Tokyo, Japan)), using a Pt plate in full automatic mode. The solution with the highest polymer concentration was prepared and measured first, and then the other concentration solutions were measured with dilution by Milli-Q water or salt solution. The surface 3320

dx.doi.org/10.1021/la403042p | Langmuir 2014, 30, 3319−3328

Langmuir

Article

tension was measured about 12 h after the solution was put into the glass cell without disturbance. Formation and Stability of Foam. Pure water and 0.1 M NaCl aqueous solutions containing 1 mg/mL polymer were shaken violently for 1 min in a glass cuvette. Five minutes later, the state of the air/ water interface was recorded by a digital camera. Fluorescence Measurement. The critical micellar concentration of these polymers was determined by fluorescence techniques using pyrene as a fluorescence probe. The fluorescence spectra were measured with a F-2500 fluorescence spectrometer, Hitachi, Tokyo, Japan, equipped with a polarizer and analyzer that uses the L-format configuration. Pyrene ((2−10) × 10−7 M) was used as fluorescence probe. The samples were excited at 335 nm, and the emission spectra were recorded in the range from 350 to 450 nm using excitation and emission slits with a band-pass of 2.5 and 2.5−5.0 nm, respectively. The temperature of the sample in the water-jacketed cell holder was controlled to be 25 ± 0.1 °C by water circulation with a Thermo Neslab RTE-7 circulating water bath. A 0.5 mM stock solution of these probes in methanol was prepared because fluorescence probes are insoluble in water. By addition of an appropriate amount of the stock solution, the final probe concentration was adjusted to 0.2−1.0 μM. All fluorescence measurements were started 24 h after sample preparation. Light Scattering (SLS and DLS). A Photal SLS-7000DL light scattering spectrometer (Otsuka Electronic, Osaka, Japan) equipped with a He−Ne laser (632.8 nm, 15 mW) was used for static light scattering (SLS) and dynamic light scattering (DLS) measurements. All the light scattering experiments were performed at 25 °C. A GC1000 correlator (Otsuka) was attached to the SLS-7000DL system to carry out DLS experiments. The typical accumulation time was 30 min for all scattering angles and concentrations. The time correlation functions thus obtained were analyzed by the double-exponential or Cumulant methods depending on the condition.32 The measurements for at least four scattering angles were performed to confirm an observation of the translational diffusion mode, and the diffusion coefficient D was calculated from the slope of the straight line in the decay rate Γ versus q2 plot, where q is the scattering vector.

q = (4πn/λ) sin(θ /2)

stirrer. To this solution, 11.1 g (50.0 mmol) of DMC and 40 mL of water were added to the mixture. This step was followed by the degassing of the mixture with three freeze−thaw cycles and filled the flask with Ar gas. The reaction was continued for 2 h at 70 °C; the crude product was purified by dialysis for 3 days in Milli-Q water, and the conductivity of the dialyzed water was checked. To get the solid product, the dialyzed solution was lyophilized. GPC was used to get the number-averaged degree of polymerization (DP) and molecular weight distribution Mw/Mn of the homopolymer yielding values of 60 and 1.13, respectively. PDMC (0.6 g, 0.048 mmol, Mn 12 600) was dissolved in 4 mL of NMF in a 25 mL Schlenk flask, followed by addition of 0.4 mL (2.84 mmol) of n-BA and 8.0 mg (0.048 mmol) of AIBN. After degassing this mixture with three freeze−thaw cycles, the mixture was heated to 70 °C under an Ar atmosphere. Copolymerization was monitored by 1 H NMR measurement by taking a small amount of reaction mixture. The copolymer was purified from the homopolymer of BA by precipitation from THF, and the solid product was finally dialyzed with a 12 000 MWCO tube for 7 days to remove unreacted PDMC. The final polymer was lyophilized. By varying the mole ratio of the above mixture, we synthesized the different block copolymers. The block ratio was calculated by 1H NMR peak integration of the block copolymer in a deuterated methanol/chloroform solvent as shown in Figure 1 with peak assignments. Molecular parameters thus obtained were tabulated in Table 1.

(1) Figure 1. 1H NMR spectrum of block copolymer in mixture of solvent (CD3OD and CDCl3).

where n is the refractive index of solvent, λ is the wavelength of laser incident beam, and θ is the scattering angle. Hydrodynamic radii, Rh, were calculated by the Stokes−Einstein equation

R h = kBT /6πηD

Table 1. Characteristics of the PDMC and PBA-b-PDMC Polymers and Micelles of Block Copolymer in Aqueous Solutiona

(2)

where kB is the Boltzmann constant, T the absolute temperature, and η the viscosity of the solvent. In SLS, the excess Rayleigh ratio, ΔRθ, which is the angular dependence of the excess absolute time-averaged scattered intensity, of a dilute polymer solution at concentration C (mg/mL) at a relatively low scattering angle θ can be related to the weight-average molecular weight Mw as33 KC /ΔR θ = (1 +

1 2 2 R g q )/M w + 2A 2 C 3

cmc (×103 mg/mL)

(3)

2

m:n

Mn

PDI

60 8:60 25:60 50:60 75:60

12 600 14 500 15 800 19 000 22 200

1.13

H2O

0.1 M NaCl

2.60 1.46 1.28

6.93 2.80 1.92

a m,n: the degree of polymerization of hydrophobic and hydrophilic blocks, respectively; Mn: the number-averaged molecular weight; PDI: polydispersity index (Mw/Mn); cmc: the critical micelle concentration. m was calculated from 1H NMR spectra, and n was estimated from GPC measurement.

with NA, dn/dC, n, and λ0 being where K = 4π n (dn/dC) Avogadro’s number, the specific refractive index increment of the solution, the solvent refractive index, and the wavelength of light in vacuo, respectively. A2 is the second virial coefficient, and Rg is the radius of gyration. By measuring Rθ at a set of C and q, Mw, Rg, and A2 were evaluated from the Zimm plot which incorporates the extrapolations of q → 0 and C → 0 on a single grid. Synthesis of PBA-b-PDMC Copolymer. Poly(n-butyl acrylate)-bpoly(3-(methacryloyloxy)ethyl)trimethylammonium chloride) (PBAb-PDMC) block copolymer was synthesized according to Scheme 1. In this process, the poly-DMC homopolymer was first synthesized followed by block copolymerization using the DMC homopolymer as a macro-charge-transfer agent (macro-CTA). In brief, 0.14 g (0.5 mmol) of ACVA and 0.26 g (0.89 mmol) of CTA and 10 mL of DMF were taken in a 100 mL Schlenk flask and stirred with a magnetic 2 2

polymer PDMC PBA-b-PDMC1 PBA-b-PDMC2 PBA-b-PDMC3 PBA-b-PDMC4

/(NAλ04)

■

RESULTS AND DISCUSSION Non-Surface Activity of PBA-b-PDMC. Ionic amphiphilic block copolymers are universally non-surface active when the requirements related to molecular weight, block ratio, and salt concentration are satisfied as previously reported.10−15 When all the requirements are satisfied, the polymers are not adsorbed at the air/water interface but form micelles in solution. In this work, we further investigated the effect of hydrophobic chain 3321

dx.doi.org/10.1021/la403042p | Langmuir 2014, 30, 3319−3328

Langmuir

Article

Figure 2. Surface tension (γ) as a function of the concentration of PBA-b-PDMC polymer. (a), (b), (c), and (d) are BA8-b-DMC60, BA25-b-DMC60, BA50-b-DMC60, and BA75-b-DMC60, respectively. Filled square is in water, and open square is in 0.1 M NaCl.

also observed for other polymers. Weakly ionic amphiphilic diblock copolymer systems such as those having carboxylic acid (neutralized) groups,10 strongly anionic sulfonic acid groups at a very low concentration,12−14,34,35 and strongly cationic systems having quaternary ammonium ions15 did not show surface tension reduction. It was shown that the block copolymer with a strong hydrophobic block (PSt-b-PAA) forms micelles without adsorption at the air/water interface, whereas the polymer (PDEGA-b-PAA) with a weak hydrophobic block adsorbed at the air/water interface.36 We previously reported that a strongly ionic PBA-b-PMAPTAC block copolymer was completely non-surface active.15 The present work showed that the shorter hydrophobic block is a normal surfactant and that the non-surface activity increases with the increase of hydrophobic block length. This suggested that the hydrophobicity and/or hydrophobic block length also played an important role in the surface activity/non-activity of their solution in addition to the image charge effect. This point will be discussed further in the next section. The surface tension decreased by the addition of salt, which is a somewhat peculiar phenomenon for high polymer concentration regions. This means that the electrostatic effect by image charge is the primary origin of the non-surface activity. Figure 3 shows photos for the foam formation and foam stability of four polymer solutions in pure water and 0.1 M NaCl at 1 mg/mL polymer concentrations. These aqueous solutions did not show any foam formation in the case of the longer PBA blocks with m = 25, 50, and 75 even after vigorous shaking, but shorter chain length of BA with m = 8 showed highly stable foam formation, all of which are consistent with the surface tension results shown in Figure 2. In 0.1 M NaCl, all cases showed foam formation. With an increase in PBA chain length, the foam height decreased but the foam was stable.

length of strongly cationic amphiphilic block copolymers on non-surface activity/surface activity by surface tension measurement and foam formation observation. Figure 2 shows the polymer concentration dependence of the surface tension (γ) of PBA-b-PDMC in H2O and 0.1 M NaCl for four polymer samples with different block ratios, that is, m:n = 8:60, 25:60, 50:60, and 75:60, where m and n are the degree of polymerization of hydrophobic and hydrophilic block, respectively. Two longer BA polymers with m:n = 50:60 and 75:60 (Figure 2c,d) showed almost no decrease of surface tension in the absence of salt with increasing polymer concentration, and the cmc could not be estimated from the surface tension data alone, which is a common method used to determine low molecular weight surfactants and nonionic polymer surfactants from the break point in the γ− concentration curve. Furthermore, the absolute value of γ of the two polymer solutions was almost equal to that for pure water (73 mN/m at 25 °C), which indicates that almost no surface excess exists and also that no surface depletion occurs. In 0.1 M NaCl, the surface tension slightly decreased. No surface tension reduction in pure water and a slight decrease in the presence of salt indicate that non-surface activity has an electrostatic origin. Shorter BA block copolymers with m:n = 25:60 showed a slight decrease in surface tension, which was enhanced by salt addition. Furthermore, the shortest BA polymer, BA8-b-DMC60, showed a behavior similar to normal surfactants as shown in Figure 2a; γ decreased with an increase in polymer concentration and reached a constant value. These surface tension measurements indicated that there is a critical length of hydrophobic block for the PBA-b-PDMC block copolymer to become non-surface active. It is non-surface active when m ≥ 25 but surface active when m = 8. Previously, a similar phenomenon, i.e., no decrease of surface tension, was 3322

dx.doi.org/10.1021/la403042p | Langmuir 2014, 30, 3319−3328

Langmuir

Article

polarity (hydrophobocity) of the environment of pyrene. The intensity ratio of the first (at 372 nm) and third (at 384 nm) vibronic peaks is sensitive to the microenvironment of the probe position, which is called the micropolarity index (I1/ I3).39,40 Figure 4 shows that the micropolarity index (I1/I3) decreases above a certain concentration of polymer with the increase of polymer concentration and reaches a constant value. The cmc values could be evaluated from the intersection point of these plots, which were also tabulated in Table 1. The micropolarity index does not decrease with the increase of polymer concentration for BA8-b-DMC60, indicating that the micelle does not form even at a concentration of 1 mg/mL, which is also supported by the strong adsorption at the air/ water interface elucidated by surface tension experiments. On the other hand, other block copolymers with a longer BA chain showed a clear bending point, from which the cmc value of the copolymers was evaluated to be 0.0026, 0.0014, and 0.0012 mg/mL for BA25-b-DMC60, BA50-b-DMC60, and BA75-bDMC60, respectively. The cmc value of the block copolymer decreased with increasing hydrophobic chain length. The low polarity of the local environment of the probe suggests that the fluorophore is located in the core of the micelle to be stabilized by hydrophobic block chains in it. Since the value of I1/I3 above cmc is very low compared to that in water (1.82), the microenvironment of the probe is thought to be highly nonpolar. These observations are consistent with the results for other polymers reported previously.41−43 The cmc value in 0.1 M NaCl was larger than that in water. This observation contradicts the well-known Corin−Harkins law44 for ionic low molecular weight surfactants, which predicts a decrease of cmc by salt addition. Zana et al. reported that the cmc value decreased with the increase of size of the headgroup of the cationic surfactant.45 The Corin−Harkins law is explained by the electrostatic screening effect between ionic head groups in the micelle shell, which makes micelle formation easier. However, in the present case, cmc increased with salt addition, which means that the micelles are harder to form with salt addition. What we should note is that the polymers are surface active to some extent in 0.1 M NaCl condition and that some polymer adsorption occurs under this condition. This adsorption is a unfavorable factor for micelle formation since all three components; i.e., unimers in solution, micelle, and adsorbed unimers are in equilibrium. This cmc increase is more pronounced for BA25-b-DMC60, implying that the cmc

Figure 3. Photo observation of foam formation behavior of 1 mg/mL of PBA-b-PDMC polymer in water (A) and 0.1 M NaCl (B) (5 min after shaking): (a) BA8-b-DMC60; (b) BA25-b-DMC60; (c) BA50-bDMC60; (d) BA75-b-DMC60.

Form formation can be explained from the following fundamental principle.1 The excess surface free energy generated by the mechanical shaking permits an increase of surface energy by an increase in surface area due to foam formation. Foam stability is largely influenced by adsorbed molecules at the foam surface. In general, the water film of foam becomes thinner due to the water draining effect at the Plateau border. However, if surfactant molecules are adsorbed at both sides of the water film of the foam, the water thin film was stabilized by steric and electrostatic repulsion between them.1 Hence, no foam formation in water solution indicates almost no polymer adsorption at the air/water interface. The absence of surface tension change for pure water systems, i.e., no added salt, is consistent with the observation of no foam formation. Almost no polymer adsorption at the air/water interface for anionic block copolymer systems was confirmed directly by X-ray reflectivity measurements in our previous works.12−14 Critical Micelle Concentration of the Polymers. The fluorescence probe studies were performed to examine the formation of the micelles with a core−shell structure using pyrene as a probe molecule. Fluorescence experiments with pyrene as a probe are widely used to determine the microenvironment of pyrene and the critical micelle concentration (cmc) of the micelles formed by synthetic surfactants36 or amphiphilic polymers.38 The vibronic peak intensity of the pyrene fluorescence spectrum is strongly influenced by the

Figure 4. Variation of micropolarity parameter (I1/I3) with PBA-b-PDMC polymer at 25 °C. 3323

dx.doi.org/10.1021/la403042p | Langmuir 2014, 30, 3319−3328

Langmuir

Article

Figure 5. Change of hydrodynamic radius (Rh) with the concentration of PBA-b-PDMC polymers in aqueous solutions. (a), (b), and (c) are BA25-bDMC60, BA50-b-DMC60, and BA75-b-DMC60, respectively. Filled squares show the values in water and open squares those in 0.1 M NaCl.

they were converted into hydrodynamic radius (Rh). The hydrodynamic radius (Rh) of BA8-b-DMC60 is around 7 nm, indicating that this polymer is present as a unimer in solution, which was also suggested from the fluorescence results, and explaining why this polymer is not non-surface active. Stable micelle formation is required for non-surface activity; micelles should be so stable that they do not show adsorption at the air/ water interface, which is largely destabilized by image charge repulsion. For the BA25-b-DMC60, BA50-b-DMC60, and BA75-bDMC60 block copolymers, the Rh values were found to be almost 24−34 nm with a polydispersity of less than 20%, which should correspond to micelle size. Figure 5a−c shows the concentration dependence of hydrodynamic radii in water and 0.1 M NaCl(aq). The concentration independence of the Rh value in water as well as NaCl(aq) was confirmed, as was the case for other non-surface active polymer micelles. The end-toend distance (Lc) of fully stretched BA25-b-DMC60, BA50-bDMC60, and BA75-b-DMC60 is estimated to be 21.2, 27.5, and 34.3 nm, respectively.46 The slightly larger or similar value of the Rh of the polymer micelles compared to the length of the stretched chains and the concentration-independent R h indicated spherical micelle formation and that there is no further intermicellar aggregation because the high charge density at the corona region prevents aggregation. As expected, the hydrodynamic radii hardly increased with increasing hydrophobic chain length. This observation can be interpreted by the critical packing parameter for micelles proposed by Israellativili47 and is consistent with our previous findings. The hydrodynamic radii of these polymers decreased (∼5 nm) upon addition of 0.1 M NaCl. This decrease is due to the shrinking of the charged corona. Ballauff et al. showed that, upon increasing salt concentration, shrinking of 2-acryloylethyl-

increases with decreasing the hydrophobic chain length. This result indicates that micelles with a shorter hydrophobic block are not as stable as those prepared with longer hydrophobic chain polymers. Thus, the adsorption−micelle equilibrium easily shifted toward adsorption with salt addition, which is consistent with the foam formation result. For the polymer with the shortest PBA, i.e., m:n = 8:60 polymer, the I1/I3 value did not change with the polymer concentrationjust flatand no bending point was found. This means that there is no cmc for this polymer; in other words, this polymer cannot form micelles. This might be due to the low hydrophobicity and also to the high water solubility due to longer hydrophilic ionic block. It is noteworthy to recall that this polymer was surface active, while the other three polymers were non-surface active, to some extent. Hence, to become non-surface active, stable micelle formation in bulk solution is also an important factor in addition to image charge repulsion at the air/water interface. If an image charge repulsion were the only unique factor, m:n = 8:60 polymer should be mostly non-surface active because this molecule has the lowest hydrophobicity while image charge effect should be equivalent for all four block copolymers. Size of Copolymer Micelle. Further confirmation of micellization in the aqueous and 0.1 M NaCl aqueous solutions of the polymer were carried out by DLS measurements. The relaxation rates (Γ) were first measured at various scattering angles in the range of 60°−105° for a given polymer concentration. The plots of Γ as a function of the square of the scattering vector (q) pass through the origin, and their linearity clearly suggests that translational diffusion of the scattering particles is being observed. From the slope of the straight line, the translational diffusion coefficients were calculated, and by assuming the Stokes−Einstein equation, 3324

dx.doi.org/10.1021/la403042p | Langmuir 2014, 30, 3319−3328

Langmuir

Article

Figure 6. Typical Zimm plot for the three block copolymers at 25 °C at angles of every 10° from 30° to 150°. The polymer concentration was varied from 2.0 to 10.0 mg/mL in H2O for BA25-b-DMC60 (a), 1.0 to 5.0 mg/mL in H2O for BA50-b-DMC60 (b), 1.0 to 4.0 mg/mL in H2O for BA75-bDMC60 (c), and 1.0 to 5.0 mg/mL in 0.1 M NaCl for BA25-b-DMC60 (d).

trimethylammonium chloride (ATAC) charged corona grafted on polystyrene increased and reached a maximum for 0.1 M monovalent salt.48 In the case of block copolymers, the grafting density (0.18 chians/nm2 for BA75-b-DMC60 block polymer) is high compared to the grafted one (∼0.05 chain/nm2), and its shrinking is less compared to the grafted one. A similar tendency was observed for a polyelectrolyte corona of corecross-linked polymer micelles.49 Hence, it can be said that the extremely high stability of polymer micelles having polyelectrolyte corona is a universal property. The origin of the high stability of polymer micelles structure is that the charge density inside the corona is so high in the bulk solution that added salt ions cannot go into the corona region. Static light scattering measurements were carried out at four polymer concentrations and scattering angles in the range of 30° ≤ θ ≤ 150° with 10° interval. The data obtained were analyzed by the Zimm plot method. The excess inverse Raleigh ratio (KC/ΔRθ) were plotted against sin2(θ/2) + kC. The corresponding Zimm plots for BA75-b-MAPTAC60 in water and 0.1 M NaCl are shown in Figure 6. The molecular weight of the scatterer and the second virial coefficient A2 were determined from the intersection and the slope of (KC/ΔRθ)θ=0 versus concentration, respectively, and Rg (radius of gyration) was calculated from the slope of (KC/ΔRθ)θ=0 versus sin2(θ/2). The molecular characteristics obtained from the Zimm plot are given in Table 2 for all three polymer solutions. The ratio of Rg/Rh is useful for characterizing the shape of the scatterer, i.e., molecular assemblies in this case. The theoretical value of Rg/ Rh for a homogeneous hard sphere is 0.778 (assuming no hydration), and it substantially increases for anisotropic particles for both prolate and oblate; for example, Rg/Rh = 1.5−1.7 for flexible linear chains in a good solvent, whereas Rg/ Rh ≥ 2 for a rigid rod.50,51 As shown in Table 2, the Rg/Rh ratio for all three polymers here was less than 0.778. In all cases, the

Table 2. Dynamic and Static Light Scattering Data for Three Polymers in Water at 25 °Ca polymer BA25-bDMC60 BA50-bDMC60 BA75-bDMC60

Mw × 10−5

A2 × 104

Rh

Rg

(nm)

(nm)

Nagg

Rg/Rh

Lc

24.5

11.0

0.86

3.91

6

0.45

21.6

29.5

14.4

3.29

0.68

17

0.48

28.0

34.0

19.0

4.26

2.26

20

0.56

34.3

(nm)

a

Rh: hydrodynamic radius; Rg: radius of gyration; Mw: molecular weight; A2: the second virial coefficient; Nagg: aggregation number; Lc: fully stretched chain length of the polymer.

Rg/Rh value was low, and this is consistent with inhomogeneous spherical micelle formation such as a sphere with a nonnegligible core.52,53 The value of Rg = 19.0 nm found for the BA75-b-DMC60 polymer micelle (Table 2) is smaller than that expected from the fully extended length of the BA75-b-DMC60 chains (i.e., ca. 34.3 nm). To further elucidate the reason for the smaller value of Rg/Rh for the BA25-b-DMC60 micelles compared with the others, we calculated the average particle density (ρd) of BAm-b-DMCn micelle from the Mw,agg and Rh values according to the equation ρd = 3M w,agg /(4πNAR h 3)

(4)

The average particle densities were found to be 0.0022, 0.0046, and 0.0073 g/cm3 for BA25-b-DMC60, BA50-b-DMC60, and BA75-b-DMC60 block copolymer micelles, respectively, which are lower than that of the bulk polymer (