Article pubs.acs.org/Langmuir
Nanostructure of Poly(N‑isopropylacrylamide) Brush at the Air/Water Interface and Its Responsivity to Temperature and Salt Hideki Matsuoka* and Kyohei Uda Department of Polymer Chemistry, Kyoto University, Kyoto 615-8510, Japan S Supporting Information *
ABSTRACT: Nanostructure and transition of the poly(Nisopropylacrylamide (PNIPAm) brush at the air/water interface were investigated by π-A isotherm and X-ray reflectivity, and an interesting behavior was observed with the change in temperature and salt. The polymer monolayer of poly(n-butyl acrylate)(PnBA)-b-PNIPAm on the water surface showed a transition between carpet-only/carpet+brush structures as a function of brush density, which was controlled by compression/expansion, as was the case for ionic brush systems. The brush stretching factor was about 50%, which was slightly less than that for a strongly ionic brush. The number of water molecules inside the brush layer was estimated to be 11−13 per repeating unit of PNIPAm chain. This value is very close to the number of hydrated water molecules reported, which means that all the water molecules inside the brush layer were hydrated water. With elevating temperature, the PNIPAm brush shrank, and the number of water molecules in the brush layer was reduced to 3. These observations certainly indicated a dehydration process. Interestingly, a part of the PNIAPm chain formed a “hydrophobic PNIPAm layer” on the carpet layer under the PnBA hydrophobic layer. A similar transition was observed also by the addition of salt to the water subphase. Although the formation of “hydrophobic PNIPAm layer” was not observed in this case, shrinking of the brush was observed with increasing salt concentration, and finally it became a carpet-only structure, which contained no water molecules. This salt effect was found to be ion specific, and its effectiveness was in the order of F− > Cl− > Br−, which is in agreement with the Hofmeister series.
■
INTRODUCTION The polymer brush provides a novel technique for surface modification, which has potential application to biomaterials, friction control, etc.1−4 The science and technology of polymer brush made major progress by the development of the controlled living radical polymerization technique.5 Especially, an ionic polymer brush6 (polyelectrolyte brush) was found to be a very low friction system,7 and the zwitterionic brush has attracted keen attention because of its very high biocompatibility.8 We have been investigating systematically the nanostructure and the transition of the polyelectrolyte brush, including strongly anionic,9,10 weakly anionic,11 and cationic12 character in the spread polymer monolayer on the water surface. The advantage of the water surface monolayer system is a very easy and precise control of brush density, which is a very important factor for brush formation and its nanostructure, by compression and expansion of the monolayer utilizing the Langmuir−Blodgett trough. Hence, the evidence obtained from the water surface system is very important and useful not only from a fundamental aspect but also for development of novel surface materials having a polymer brush including solid surface materials. In other words, a study on polymer brush on the water surface should provide better understanding of polymer brush systems on the solid substrate by the fundamental point of view © 2016 American Chemical Society
We have found that the polyelectrolyte brush has two nanostructures; carpet-only and carpet+brush structures and a transition between these two structures occurs as a function of brush density and added salt concentration.9−12 A carpet layer always formed just beneath the hydrophobic layer on the water, which is thought to be formed to avoid the direct contact between the hydrophobic layer and water subphase. In addition, very interesting behavior was observed also for the effect of salt addition. For a strongly ionic brush, a critical salt concentration (csc) was found; below the csc, brush nanostructure was not affected by salt addition, while the brush suddenly shrunk above the csc.10 For a weakly ionic brush, with increasing salt concentration, the brush chain first extended and then shrunk-up by further addition of salt.11 This observation is a result of the competitive effect of an electrostatic screening effect and an increase of degree of dissociation.11 Recently, unique properties of the zwitterionic brush have also been reported.13 In this study, we have extended our brush study to poly(Nisopropylacrylamide) (PNIPAm) brush systems. PNIPAm is reported to be thermosensitive, and shows a lower critical Received: May 7, 2016 Revised: July 27, 2016 Published: July 28, 2016 8383
DOI: 10.1021/acs.langmuir.6b01752 Langmuir 2016, 32, 8383−8391
Article
Langmuir Scheme 1. Synthesis of PnBA-b-PNIPAm. Proton Symbols Are for 1H NMR in Figure 1
surface systems, and may provide a better understanding and development of novel surface materials and devices.
solution temperature (LCST) behavior. Its cloud point is located around 30−34 °C, which is close to body temperature of human beings.14 Hence, PNIPAm is attracting keen attention as a potential polymer material, which has thermo-responsivity. For the air/water interface study of PNIPAm, some investigations for homopolymer systems can be found.15−17 Kawaguchi et al. studied the PNIPAm homopolymer adsorbed at the air/water interface for solution15 and spread monolayers16 by surface tension and ellipsometry measurements. The effect of temperature was examined and an increase of adsorbed amount above LCST was observed. Study on block copolymer systems is rare, but Liu et al.18 reported conformational changes of PNIPAm at the air/water interface by using block copolymers of polystyrene (PS) and PNIPAm. In this example, the block copolymer did not form a uniform monolayer due to a high Tg of the PS block, the PS formed a surface aggregate, and the PNIPAm chain adsorbed at the air/water interface. Liu et al. investigated the effect of temperature on surface pressure and found that the conformation of the adsorbed PNIPAm chain, such as train or loop structure, is a dominant factor. Xue et al.19 studied protein adsorption on the PNIPAm brush on the solid surface. They observed a decrease of brush thickness with elevating temperature for high brush density systems above 0.11 chains/nm2, but it was found to be constant at a low value for low brush density systems. We studied the nanostructure and transition of the PNIPAm brush by utilizing the advantage of the polymer monolayer system on the water surface, that is, an easy and precise control of the brush density for the samples of variety of brush chain length with narrow distribution. We carried out a surface pressure−area/molecule (π−A) isotherm and X-ray reflectivity (XR) measurements for the water surface monolayers of amphiphilic diblock copolymer, poly(n-butyl acrylate)-bPNIPAm, which was synthesized by reversible addition− fragmentation chain transfer (RAFT) polymerization technique, with precise control of its molecular weight and its distribution. The effects of the brush density, temperature, salt addition, and ion species were duly investigated. For the effect of brush density, a similar trend was observed for ionic brush systems studied so far, but very unique structural changes were found for the effect of temperature and added salt. Not only the brush structure, but also the number of water molecules in the brush layer was estimated, and the effect of temperature and salt were examined. The information obtained should be universal for PNIPAm brush, which can be applied to solid
■
EXPERIMENTAL SECTION
Materials. N-Isopropylacrylamide (NIPAm, 98%), the initiators, 2,2′-azobisisobuthylnitlyl (AIBN, 98%) and 4,4′-azobis(4-cyanovaleic acid) (ACVA, 98%), N,N-dimethylformamide (DMF), and tetrahydrofuran (THF), were purchased from Wako (Osaka, Japan) and used as received. n-Butyl acrylate (nBA) was a product of Nacalai Tesque (99%) and was used as received. Water used for solution preparation and dialysis was ultrapure water obtained by Milli-Q system (18.2 MΩcm). Dialysis for PNIPAm homopolymer purification was done with a dialysis tube made by Orange Scientific (MWCO:3500). Synthesis of CTA. The chain transfer agent (CTA) used for RAFT polymerization was 4-cyanopentanoic acid dithiobenzoate, which was synthesized by following the procedure reported by Mitsukami et al.20 The formation and purity of CTA was confirmed by 1H NMR with CDCl3 as a solvent. Synthesis of Block Copolymers. The synthesis method of block copolymers are shown in Scheme 1. First, poly(nBA) homopolymer was synthesized by RAFT technique. A typical recipe is as follows: nBA, CTA, and initiator (AIBN or ACVA) (150:1:1 (molar ratio)) were mixed and put into a Schlenk tube, and three cycles of a freeze− pump−thaw cycle were carried out for degassing. Polymerization was performed at 70 °C under Ar atmosphere for 2 h. Reaction was stopped by ice-cooling, and unreacted monomer was removed by pumping to low pressure. Polymer obtained was dissolved in a minimum amount of chloroform, and then concentrated by evaporation. Transparent redish-orange PnBA polymer soft-solid was obtained. The degree of polymerization of this example was 93 (by GPC) and the yield was 31%. Purity was confirmed by 1H NMR and IR spectra. PnBA with different degrees of polymerization was obtained by changing the amount of nBA monomer. Block copolymer was synthesized by using PnBA thus obtained as a macroinitiator. A typical recipe is as follows. PnBA homopolymer, AIBN, NIPAm (1:2:100, molar ratio) were dissolved in DMF and degassed by freeze−pump−thaw cycles four times. RAFT block copolymerization was carried out under Ar atmosphere for 24 h at 70 °C. After the reaction was stopped by ice-cooling, polymer precipitated by the addition of hexane and was washed by pure water and diethyl ether. The precipitant thus obtained was dried in a vacuum chamber to be colorless elastic solids. The block copolymer thus synthesized was identified by 1H NMR and IR spectra. The degree of polymerization of this example was 93:93 as will be shown later. Gel Permeation Chromatography (GPC). The molecular weight and its distribution were evaluated by a JASCO GPC system composed with RI-965 RI detector, UV2075 Plus UV detector, PU980 pump, DG-980-50 degasser, and CO-965 column oven. The column used was Shodex KF804L with THF as an eluent and polystyrene standard. 8384
DOI: 10.1021/acs.langmuir.6b01752 Langmuir 2016, 32, 8383−8391
Article
Langmuir
Figure 1. (a) 1H NMR spectra of PnBA and PnBA-b-PNIPAm (solvent: CDCl3), (b) IR spectra of PnBA-b-PNIPA, m:n = 93:93. Proton symbol is shown in Scheme 1. 1 H Nuclear Magnetic Resonance (NMR). 1H NMR spectra were obtained by JEOL JNM AL-400 and JNM EX-400 spectrometer. The polymer concentration was 1 wt %, and the solvent was CDCl3 (EURISO-TOP 99.8%). Infrared Spectroscopy (IR). IR measurements for confirmation of polymer structure and purity were carried out with a FT-IR-8400 (Shimadzu, Kyoto, Japan) by the KBr method. X-Ray Reflectivity (XR). The XR instrument for monolayer nanostructure analyses was RINT-TTR-MA (Rigaku, Tokyo, Japan) with a specially designed Langmuir−Blodgett (LB) trough (USI System, Fukuoka, Japan). The details of the instrument and data analysis were fully described in our previous papers.9−11,21 The temperature of water subphase in LB trough was controlled with a water circulator (Thermoelectron RTE-7DI). The temperature of the surface of water subphase was directly measured with a surface thermometer CUSTOM IR-303 (CUSTOM Corp., Tokyo, Japan). 30 °C was the upper limit of the sample temperature due to instrumental limitation. The polymer monolayer was prepared by spreading chloroform solution on the water surface. The solution was left to stand 30 min to 1 h for solvent evaporation, the π−A (surface pressure−area/molecule) isotherm was measured with a 0.030 mm/ sec barrier moving speed (1.08 cm2/min). Brewster Angle Microscope (BAM). BAM observation of the monolayer on the water surface was performed with a Multiskope System22 (Optrel, Sinzing, Germany) with a specially designed Langmuir−Blodgett (LB) trough (USI System, Fukuoka, Japan). Turbidity. The cloud point of PNIPAm aqueous solutions was determined with a UV−vis spectrometer U-3310 (Hitachi) for 1 mg/ mL solution. We assigned the temperature, at which the transmittance of the solution was 90% at a wavelength of 550 nm, as a cloud point.
Table 2. Characteristics of Block Copolymers Synthesized PnBA-b-PNIPAm1 PnBA-b-PNIPAm2 PnBA-b-PNIPAm3 b
■
PnBA1 PnBA2 PnBA3 a
12300 29500 26500
1.35 1.25 1.35
22800 38300 29800
93:93 228:76 205:29
Determined by 1H NMR spectra (peak d:f).
Figure 2. π−A isotherms of homopolymer and block copolymer monolayers on the water surface.
surface and homopolymers, i.e., PnBA and PNIPAm, as references. The isotherm for PNIPAM increased continuously with compression, while that for PnBA also increased but with a plateau region in the middle range, which might be typical for poly(acrylate) monolayer systems since other examples exist.11−13 The isotherms for block copolymer monolayers also increase smoothly with a plateau region, which might be due to the PnBA block, and a second increase at around A = 5− 6 nm2/molecule, depending on the n value, was observed. This might be attributed to the repulsive interaction between the PNIPAm blocks. Figure 3panels a−c show XR profiles (left) and density profiles normal to the water surface for block copolymer monolayers at the air/water interface at different A (area/ molecule), i.e., at a different brush density when a brush exists. In these cases, a three-layer model (3box model) was used for
Table 1. Characteristics of PnBA homopolymers synthesized Mw/Mna
m:nb
Brush Density Dependence at 25 °C. Figure 2 shows π− A isotherms of three block copolymer monolayers on the water
RESULTS AND DISCUSSION Characterization of Block Copolymers. 1H NMR and IR spectra are shown in Figure 1 All peaks could be assigned as shown in the figures. The block ratio, m:n was estimated by the area of peaks d and f. Characteristics of homo and block copolymers are summarized in Table 1 and 2, respectively. GPC estimation of the molecular weight distribution (Mw/Mn) of the block copolymer was unsuccessful, unfortunately (see Supporting Information for the details).
Mna
Mnb
Determined by GPC with THF eluent (PSt standard). 8385
DOI: 10.1021/acs.langmuir.6b01752 Langmuir 2016, 32, 8383−8391
Article
Langmuir
Figure 3. (a) XR profiles (left) and density profiles of (nBA)93-b-(NIPAm)93 monolayer on water at 25 °C (right), (b) XR profiles (left) and density profiles of (nBA)228-b-(NIPAm)76 monolayer on water at 25 °C (right), (c) XR profiles (left) and density profiles of (nBA)205-b-(NIPAm)29 monolayer on water at 25 °C (right). (d) Layer thickness of the (nBA)m-b-(NIPAm)n monolayer on water at 25 °C. Surface pressure variation is also shown. Dashed vertical line shows c.b.d. m:n = 93:93, 228:76, 205:29. For XR, q is the scattering vector: q = 4π sin θ/λ, where θ is the reflection angle and λ the wavelength of X-ray (1.5406 Å).
XR profile fitting. Then the density profiles shown at the right were obtained. Small fringes were observed in XR profiles as
indicated by small arrows in the figure, which were well reproduced by a fitting line (solid lines in the figures). By this 8386
DOI: 10.1021/acs.langmuir.6b01752 Langmuir 2016, 32, 8383−8391
Article
Langmuir
Figure 4. Number of water molecule per PNIPAm monomer unit in the carpet and brush layer: (a) (nBA)93-b-(NIPAm)93, (b) (nBA)228-b(NIPAm)76, and (c) (nBA)205-b-(NIPAm)29.
fitting, the density profiles were obtained, and the thickness of each layer and its variation with the brush density is shown in Figure 3d in a histogram manner. At a small A, there exist only two layers; hydrophobic PnBA layer on the water surface and PNIPAm carpet layer (dense hydrophilic polymer layer) under water. By compression, the thickness of the hydrophobic layer increased, which is quite a natural trend, while that for the carpet layer is not largely affected. At a certain A (i.e., brush density), the third layer is formed under the carpet layer for (nBA)93-b-(NIPAm)93 and (nBA)228-b-(NIPAm)76. The thickness of this third layer increased with compression and the density inside is lower than that of the carpet layer (its density is largely close to the PNIPAm homopolymer); hence, this layer should be a brush layer. This kind of three-layered structure and transition between carpet-only structure and carpet-and-brush structure were also observed for anionic and cationic brush systems which we had studied previously.9−13 For (nBA)205-b-(NIPAm)29, the brush layer was not formed even after a large compression, which is reasonably understood when we consider its very short NIPAM chain. The fitting parameters for XR analysis are shown in Tables S1−S3 in Supporting Information. Now, the critical brush density (c.b.d.) for the PNIPAm brush, which is the brush density where the brush is formed; in other words, the transition between carpet-only and carpet-andbrush occurs, can be evaluated. The c.b.d. position was shown in Figure 3d by vertical dashed line. The c.b.d for (nBA)93-b(NIPAm)93 and (nBA)228-b-(NIPAm)76 were evaluated to be 0.2 nm−2 (between 0.16 and 0.23).9 This value is larger than that for the poly(styrenesulfonate) (PSS) brush (0.12 nm−2). PSS is a strongly acidic brush, and the main factor of its brush formation is an electrostatic repulsion; hence, the c.b.d. for the PSS brush is very low. This c.b.d value for PNIPAm brush is close to other brush systems, in which steric repulsion plays an important role for brush formation. The c.b.d. for the poly(methacrylic acid) (protonated form) brush was reported to be 0.2−0.4 nm−2 depending on its chain length. c.b.d. for the poly(carboxybetaine) brush, which is zwitterionic, was also reported recently to be about 0.2 nm−2, and the steric repulsion was assigned to be an important factor.13 Hence, in the present PNIPAm brush case, the steric repulsion is judged to be an important factor for brush formation. Since we now know the density inside the brush (which should be an average of those for PNIAPm and water), and brush thickness in addition to bulk density of PNIPAm (1.13 g/ cm3) and water, we can calculate the number of water molecules inside the brush layer, n. Be sure that n is the value
for one repeating unit of hydrophilic chain (its length is 2.54 Å). A similar calculation is also possible for the carpet layer. (The details of calculation was described in the Supporting Information) The n values were plotted as a function of brush density in Figure 4, which was only about 2 in the carpet layer, and this value is not largely affected by the change of brush density. In our studies for ionic brush systems, the carpet layer was composed of the hydrophilic block alone and its density was almost the same as that for the bulk hydrophilic polymer. In this PNIPAm case, the fact that only about 2 water molecules are inside, that is, the carpet layer consists of mostly hydrophilic polymer, is consistent with previous examples studied so far. For inside the brush layer, n values of 11−13 were found. This value is quite interesting since a similar number can be found in other studies. Ono and Shikata23 investigated the number of hydrated water molecules around the PNIAPm chain in water solution by the dielectric relaxation technique, and reported a hydration number of 11−12 water molecules per NIPAm unit. Shibayama et al.,24 reported a hydration number of 11−15 molecules in the PNIPAm gel. Our n value, 11−13, is very close to these values. This means that all water molecules inside the brush layer are hydrated water; no free water molecules inside the brush layer. Response to Temperature. Figure 5a shows π−A isotherms of the block copolymer monolayer at various temperatures between 25 and 34 °C. With elevating temperature, the isotherm shifted toward larger A values, while almost no change was found above 30 °C. This might be due to an increase of lateral steric repulsion by shrinking of the PNIPAm chain caused by dehydration. The reversibility of the response to temperature was confirmed by a constant-A experiment. Figure 5b shows the change of surface pressure by a temperature change at constant A. The surface pressure reversibly responded to temperature change. Uniform and smooth monolayer formation both at 25 and 30 °C was confirmed by BAM observation (Figure 5c). Needless to say, a block copolymer can form a stable monolayer, but a polymer blend even with the same composition cannot. The PNIPAm and PnBA mixture forms a unique monolayer with microphase separation as shown in Figure S1 in the Supporting Information, which is, however, out of the scope of this study. Figure 6 panel a shows XR profiles at A = 4.5 nm2 with a two cycle change of temperature between 25 and 30 °C, and panel b is the density profile normal to the surface obtained by model fitting of XR profiles in panel a. At 25 °C, the monolayer consists of a hydrophobic layer on the water surface, and a PNIPAm carpet layer and a long brush layer under water. By 8387
DOI: 10.1021/acs.langmuir.6b01752 Langmuir 2016, 32, 8383−8391
Article
Langmuir
brush layer still existed. In fact, the 3-box model could not well reproduce the XR profiles at 30 °C, so the 4-box model was employed and satisfactory fitting results were obtained. As shown in the inset of Figure 6a, the 3-box model could not well reproduce the experimental XR profile at a small angle region (around q = 0.1 Å−1) as shown in the inset of Figure 6a by the arrow. By addition of one more layer to the fitting model, that is, by using a 4-box model, satisfactory agreement was obtained as shown in the inset although the fitting curve cannot be notified because the line is hidden by data points (i.e., perfect fitting). The fourth layer introduced is a “PNIPAm hydrophobic layer” on the water surface. At 25 °C, the thickness of the PnBA−hydrophobic, carpet and brush layers were estimated to be 42, 22, and 53−62 nm, respectively. At 30 °C, these values were changed to be 42, 17−19, and 20 nm, respectively, with an additional layer of thickness of 7 nm as the “PNIPAm hydrophobic layer” which exists between PnBA hydrophobic and carpet layers. (Figure 6c) This change may indicate that part of the PNIPAm chain goes into the hydrophobic layer on the water as a hydrophobic chain, and part of the PNIPAm chain immersed into the carpet layer as the brush chain shrunk. We have defined the value of the “brush stretch” as a ratio of brush thickness and counter length of brush chain, which is included in the brush layer (i.e., the length of the chain in carpet layer is not included). The brush stretch value for 25 °C was 55.8% and 48.9% for the first and second cycle, respectively. This value means that the brush chain stretched fairly well, although not so much as the case of a strongly ionic brush such as poly(styrenesulfonate) brush (about 80%).9,10 This value changed to 26.2% and 24.5% for each, clearly indicating the shrinking of brush chain due to the hydrophobic nature of PNIPAm chain under this temperature. The fitting parameters including those for other block copolymer monolayers were tabulated in Tables S4−S6. The number of water molecules in the brush layer can be estimated also for this system by volume requirement. It was calculated to be about 13 and 3 molecules at 25 and 30 °C, respectively. It is obvious that dehydration occurred in the brush layer, which causes shrinking of the PNIPAm brush chains.
Figure 5. (a) p-A isotherms of nBA93-b-NIPAm93 on water at various temperatures; (b) reversible switching of surface pressure at 4.5 nm2 between 25 and 30 °C; (c) BAM observation of a monolayer at 25 and 30 °C.
the temperature change from 25 to 30 °C, it is certainly observed that the PNIAPm brush chains shrunk. This behavior is reasonably explained by the fact that PNIPAm chains changed from hydrophilic to hydrophobic by the temperature change above the LCST, although it may not be perfectly hydrophobic. What is interesting is the change of the hydrophobic/carpet layer interface, indicated by an arrow in Figure 6b. The density of this region increased; a part of PNIPAm in the carpet layer appeared to immerse into the hydrophobic layer. This observation is also explained by the nature of the PNIPAm chain; since the PNIPAm chain is now hydrophobic, it should form a “hydrophobic layer on the water surface”. This phenomena was partly but probably not perfectly established since the PNIPAm carpet layer and the “shrunken”
Figure 6. (a) XR profiles and (b) density profiles, and thickness of each layer of (nBA)93-b-(NIPAm)93 monolayer on water at 4.5 nm2 at 25 and 30 °C. The inset of panel a is the magnification of a small angle region. This region is well reproduced by the 4box model but not by the 3box model. 8388
DOI: 10.1021/acs.langmuir.6b01752 Langmuir 2016, 32, 8383−8391
Article
Langmuir
Figure 7. (a) XR profiles and (b) density profiles of nBA93-b-NIPAm93 monolayer on the water at 30 °C as a function of the brush density.
Figure 8. Layer thickness of nBA93-b-NIPAm93 monolayer on aqueous salt solutions of different concentrations at 25 °C: red, hydrophobic layer; blue, carpet layer; green, brush layer.
Here, it may be fair to mention about the limitation of our experimental condition. We performed XR measurements at 30 °C, which might be a little bit lower temperature than the LCST of PNIPAm generally accepted (about 34 °C). This is due to our instrument limitation. Hence, there is a possibility that the structure which we found at 30 °C might not be the nanostructure after transition, but maybe the one during transition. We found three conditions of PNIPAm chain in the monolayer at 30 °C, that is, shrunk brush, carpet layer, and hydrophobic PNIPAm layer. This existence of multiconditions of PNIPAm chain might mean that this situation is not in equilibrium but in a transition process (during transition). Brush Density Dependence at 30 °C. Figure 7a shows the XR profiles of the (nBA)93-b-(NIPAm)93 block copolymer monolayer on the water surface at various A (i..e., brush density varies when brush exists) at 30 °C, and Figure 7b shows the density profiles evaluated by XR model fitting. As mentioned in the section above, the XR profiles were well reproduced by the 4-layer (4box) model, which consists of a hydrophobic PnBA layer, a hydrophobic PNIPAm layer on the water surface, and carpet and brush layers of PNIPAm under water. As can be seen in Figure 7b, under a low brush density condition, no brush layer is formed; only a carpet layer exists under water (fitting parameters are given in the Supporting Information).
By compression, the thickness of the PnBA hydrophobic layer increased from 30 to 53 Å, accordingly. On the other hand, the thickness of the hydrophobic PNIPAm layer is almost unchanged at 9 Å. Under water, the PNIPAm brush layer was formed by compression at the brush density of 0.25 nm−2, and its thickness increased by compression. The thickness of the carpet layer was almost unaffected by compression as was the case for other ionic brushes. Hence, the carpet-only/carpetand-brush transition was also observed at 30 °C. A similar trend was observed for (nBA)228-b-(NIPAm)76, while no brush was found for (nBA)205-b-(NIPAm)29 short PNIPAm block polymer, which are shown in Figures S2 and S3. The results are summarized in histogram manner in Figure S4 and fitting parameters are summarized in Tables S7−S9. Another interesting feature for this short PNIPAm polymer is that no hydrophobic PNIPAm layer was formed at a low surface pressure (25 and 27 mN/m) even at 30 °C. As discussed previously, a carpet layer is formed to stabilize the monolayer; it is formed to avoid direct contact between the hydrophobic PnBA layer and water subphase. In other words, it is formed to minimize the interfacial energy between the monolayer and water. Since this block copolymer has a short PNIPAm chain, all of the ions are used to form the carpet layer; no brush and no hydrophobic PNIPAm layers could be formed. 8389
DOI: 10.1021/acs.langmuir.6b01752 Langmuir 2016, 32, 8383−8391
Article
Langmuir
Figure 9. Number of water molecules in brush layers of nBA93-b-NIPAm93 monolayer as a function of added salt concentration.
Salt Concentration Dependence and Ion-Specific Effect. The LCST behavior of PNIPAm is affected by salt addition.25 We have also confirmed that the cloud point of the PNIPAMm homopolymer (the degree of polymerization of 123) solution decreased with increasing added salt concentration and is also affected by ion species of salts (F−, Cl−, and Br−, in this case) as shown in Figure S5, which shows a trend similar to that reported by Cremer et al.25 We examined the nanostructure and its change by salt addition for three diblock copolymer monolayers on the water surface with NaF, NaCl, and NaBr as an added salt at 25 °C. XR profiles and density profiles normal to the surface at different salt concentrations (0.01 M−2.0 M) are shown in Figures S6−S8 and fitting parameters are tabulated in Tables S10−S18. Figure 8 shows the thickness of each layer as a function of salt concentration of three different kinds of salt and at constant A (=3.7−4.0 nm2) for (nBA)93-b-(NIPAm)93 in a histogram manner. The thickness of the brush layer decreased with salt concentration, and then disappeared. This is a transition from a carpet+brush structure to carpet-only structure by salt addition. As shown in Figure S5, the cloud point decreased with increasing salt concentration; it is reasonable that PNIPAm changed from hydrophilic to hydrophobic at a certain salt concentration even at 25 °C. The salt concentration where the brush layer disappeared was 0.5 M for NaF, 1.0 M for NaCl, and 2.0 M for NaBr. This trend is similar to the observation in Figure S5. From Figure 8, it can be said that F− is the most effective ion for the transition (in the sense that a small amount of ion can cause transition), Cl− is the medium, and Br− has the weakest effect. It may be interesting to note that this order is just the same as the Hofmeister series concept;26−28 a kosmotropic ion has a stronger effect on the transition. This trend may be related to our observation that all the water molecules in the brush layer are hydrated water. The Hofmeister series was originally proposed as an order of efficient salting-out effect for protein,26,27 but another new interpretation of this series has been proposed and is now under discussion.29 The transition at the same salt concentration and similar trend were observed for (nBA)228-b-(NIPAm)76. For (nBA)205-b-(NIPAm)29, which has only a carpet layer, but no brush layer, no change of carpet layer thickness was observed by salt addition. This may occur because the carpet layer is a dense PNIPAm layer with few
water molecules (about only one per repeating unit), so ions cannot penetrate into the carpet layer. Also as seen in Figure 8, the thickness of the carpet layer is not affected by salt addition, although a slight increase of thickness was observed after transition to the carpet-only structure. As was in the temperature study, the number of water molecules in each layer can be estimated from the average density of the layers, which is plotted as a function of salt concentration in Figure 9. The absolute number of water molecules is different for the kind of salt used due to ionspecific effect discussed above, but the general trend can be seen. In the brush layer of the 93:93 polymer, 11 water molecules exist at no salt condition. As mentioned, all of these water molecules are hydrated. With increasing salt concentration, the number of water molecules decreased and finally becomes almost zero at around 1 M; that is, the layer changed from brush layer to carpet layer. The dehydration process and the transition from the brush layer to carpet layer by salt addition were clearly and quantitatively observed.
■
CONCLUSIONS The nanostructure and the transition of PNIPAm brush in the PnBA-b-PNIPAm block copolymer monolayer at the air/water interface were duly investigated by π−A isotherm and XR technique with the temperature change and salt addition as external stimuli. The PNIPAm brush showed a carpet-only/ carpet+brush nanostructure transition as a function of brush density as was the case for ionic polymer brushes studied so far. The brush stretch factor was about 50%, which was slightly smaller than that of strongly ionic brushes. With elevating temperature, since the PNIPAm chain becomes hydrophobic, the PNIPAm brush shrunk-up, some PNIPAm brush chains immersed into the carpet layer, and some part of the PNIPAm chain in the carpet layer moved to the water surface to form a “hydrophobic PNIPAm layer” between the hydrophobic PnBA and carpet layers. There were estimated to be 11−13 water molecules in the brush layer per repeating unit, which is very close to the number of hydrated water molecules for PNIPAm in solution and gel. This means that all the water molecules in the brush layer are hydrated water. The transition from carpet +brush to carpet-only structures was also observed by salt addition. This was a result of the effect of salt to lowered LCST 8390
DOI: 10.1021/acs.langmuir.6b01752 Langmuir 2016, 32, 8383−8391
Article
Langmuir
(9) Kaewsaiha, P.; Matsumoto, K.; Matsuoka, H. Synthesis and Nanostructure of Strong Polyelectrolyte Brush in Amphiphilic Diblock Copolymer Monolayers on Water Surface. Langmuir 2004, 20 (16), 6754−6761. (10) Kaewsaiha, P.; Matsumoto, K.; Matsuoka, H. Salt Effect on the Nanostructure of Strong Polyelectrolyte Brushes in Amphiphilic Diblock Copolymer Monolayers on the Water Surface. Langmuir 2007, 23 (13), 7065−7071. (11) Matsuoka, H.; Suetomi, Y.; Kaewsaiha, P.; Matsumoto, K. Nanostructure of Poly(Acrylic Acid) Brush and its Transition in the Amphiphilic Diblock Copolymer Monolayer on the Water Surface. Langmuir 2009, 25 (24), 13752−13762. (12) Matsuoka, H.; Fujita, S.; Ghosh, A.; Nakayama, S.; Yamakawa, Y.; Yusa, S.; Saruwatari, Y. Nanostructure of Cationic Polymer Brush at the Air/Water Interface. MATEC Web Conf. 2013, 4, 1−4. (13) Matsuoka, H.; Yamakawa, Y.; Ghosh, A.; Saruwatari, Y. Nanostructure and Salt Effect of Zwitterionic Carboxybetaine Brush at the Air/Water Interface. Langmuir 2015, 31, 4827−4836. (14) Schild, H. Poly(N-isopropylacrylamide): Experiment, Theory and Application. Prog. Polym. Sci. 1992, 17, 163−249. (15) Kawaguchi, M.; Hirose, Y.; Kato, T. Effects of Temperature and Concentration on Surface Tension of Poly(N-isopropylacrylamide) Solutions. Langmuir 1996, 12, 3523−3526. (16) Kawaguchi, M.; Saito, W.; Kato, T. Poly(N-isopropylacrylamide) Films at the Air/Water Interface. Macromolecules 1994, 27, 5882−5884. (17) Zhang, J.; Pelton, R. Poly(N-isopropylacrylamide) at the Air/ Water Interface. Langmuir 1996, 12, 2611−2612. (18) Liu, G.; Yang, S.; Zhang, G. Conformational Changes of Poly(N-isopropylacrylamide) Chains at Air/Water Interface: Effect of Temperature, Compression, and Packing Density. J. Phys. Chem. B 2007, 111, 3633−3639. (19) Xue, C.; Yonet-Tanyeri, N.; Brouette, N.; Sferrazza, M.; Braun, P. V.; Leckband, D. E. Protein Adsorption on Poly(N-isopropylacrylamide) Brushes: Dependence on Grafting Density and Chain Collapse. Langmuir 2011, 27 (14), 8810−8818. (20) Mitsukami, Y.; Donovan, M. S.; Lowe, A. B.; McCormick, C. L. Water-Soluble Polymers. 81. Direct Synthesis of Hydrophilic StyrenicBased Homopolymers and Block Copolymers in Aqueous Solution via RAFT. Macromolecules 2001, 34, 2248−2256. (21) Mouri, E.; Matsumoto, K.; Matsuoka, H. Effect of pH on the Nanostructure of Acidic Amphiphilic Carbosilane/Methacrylic Acid Block Copolymer Monolayers at Air/Water Interface. J. Appl. Crystallogr. 2003, 36 (3), 722−726. (22) Harke, M.; Teppner, R.; Schulz, O.; Motschmann, H.; Orendi, H. Description of a Single Modular Optical Setup for Ellipsometry, Surface Plasmons, Waveguide Modes, and Their Corresponding Imaging Techniques Including Brewster Angle Microscopy. Rev. Sci. Instrum. 1997, 68 (8), 3130−3134. (23) Ono, Y.; Shikata, T. Contrary hydration behavior of Nisopropylacrylamide to its polymer, P(NIPAm), with a lower critical solution temperature. J. Phys. Chem. B 2007, 111, 1511−1513. (24) Shibayama, M.; Mizutani, S.; Nomura, S. Thermal properties of copolymer gels containing N-isopropylacrylamide. Macromolecules 1996, 29, 2019−2024. (25) Zhang, Y.; Furyk, S.; Bergbreiter, D. E.; Cremer, P. S. J. Am. Chem. Soc. 2005, 127, 14505−14510. (26) Hofmeister, F. Zur Lehre von der Wirkung der Salze’ (about the science of the effect of salts. Naunyn-Schmiedeberg's Arch. Pharmacol. 1888, 24, 247−260. (27) Kunz, W.; Henle, J.; Ninham, B. W. ‘Zur Lehre von der Wirkung der Salze’ (about the science of the effect of salts): Franz Hofmeister’s historical papers. Curr. Opin. Colloid Interface Sci. 2004, 9, 19−37. (28) Wilson, E. K. A Renaissance For Hofmeister. Chem. Eng. News 2007, 85 (48), 47−49. (29) Zhang, Y.; Cremer, P. S. Interactions between macromolecules and ions: the Hofmeister series. Curr. Opin. Chem. Biol. 2006, 10, 658−663.
of PNIPAm. A specific ion effect was found for the salt concentration in which transition occurs, and its order was F− > Cl− > Br−, which is consistent with the Hofmeister series. A thermoresponsive polymer brush has now been established, and its structure can be also tuned by brush length, brush density, and salt concentration, which has a potential application to novel surface materials with responsivity.
■
ASSOCIATED CONTENT
* Supporting Information S
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.6b01752. Fitting parameters for the structures studied; calculations for the numbers of water molecules; estimation of molecular weight distribution (Mw/Mn) of P(nBA)-bPNIPAm block copolymer; additional figures (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by a grant-in-aid for Scientific Research on Innovative Areas “Molecular Soft-Interface Science” (20106006) from the Ministry of Education, Culture, Sports, Science and Technology of Japan, to which our sincere gratitude is due. We express sincere thanks to Professor Ken-ici Iimura (Utsunomiya University, Japan) for his kind support in the BAM observation, and to Professor Shin-ichi Yusa (Hyogo Prefecture University, Japan) for his kind suggestions for RAFT synthesis. Our great appreciation also goes to Dr. Arjun Ghosh, Dr. Murugaboopathy Sivanantham, and Dr. Saurabh Shrivastava for their fruitful suggestions and discussions as postdoctoral fellows in our research group.
■
REFERENCES
(1) Advincula, R. C., Brittain, W. J., Caster, K. C., Rühe, J. Eds. Polymer Brushes; Wiley: 2004. (2) Hamely, I. Block Copolymers in Solution: Fundamentals and Applications: Wiley: 2005. (3) Halperin, A.; Tirrell, M.; Lodge, T. P. Thetherd Chains in Polymer Microsructures. Adv. Polym. Sci. 1992, 100, 31−71. (4) Milner, S. T. Polymer Brushes. Science 1991, 251, 905−914. (5) Mathers, R. T.; Magenau, A. J. D.; Schröder, K.; Matyjaszewski, K. Overvew of Controlled/Living Radical Polymerizations of Vinyl Monomers.In Monitoring Polymerization Reactions: From Fundamentals to Application, 1st ed.; Reed, W. F., Alb, A. M. Eds.; John Wiley & Sons, 2014; Chapter 2. (6) Rühe, J.; Ballauff, M.; Biesalski, M.; Dziezok, P.; Gröhn, F.; Johannsmann, D.; Houbenov, N.; Hugenberg, N.; Konradi, R.; Minko, S.; Motornov, M.; Netz, R. R.; Schmidt, M.; Seidel, C.; Stamm, M.; Stephan, T.; Usov, D.; Zhang, H. Polyelectrolyte Brushes. Adv. Polym. Sci. 2004, 165, 79−150. (7) Raviv, U.; Giasson, S.; Kampf, N.; Gohy, J. F.; Jerome, R.; Klein, J. Lubrication by Charged Polymers. Nature 2003, 425 (6954), 163− 165. (8) Hiraguchi, Y.; Nagahashi, K.; Shibayama, T.; Hayashi, T.; Yano, T.; Kushiro, K.; Takai, M. Effect of the Distribution of Adsorbed Proteins on Cellular Adhesion Behaviors Using Surfaces of Nanoscale Phase-Reversed Amphiphilic Block Copolymers. Acta Biomater. 2014, 10, 2988−2995. 8391
DOI: 10.1021/acs.langmuir.6b01752 Langmuir 2016, 32, 8383−8391
