Thermoresponsive Properties of PNIPAM-Based Hydrogels: Effect of

Article pubs.acs.org/Langmuir

Thermoresponsive Properties of PNIPAM-Based Hydrogels: Effect of Molecular Architecture and Embedded Gold Nanoparticles Hong Hanh Nguyen,† Bruno Payré,‡ Juliette Fitremann,† Nancy Lauth-de Viguerie,*,† and Jean-Daniel Marty*,† †

IMRCP, CNRS UMR 5623 and ‡CMEAB, IFR-BMT, Université de Toulouse, 31062 Toulouse Cedex 09, France S Supporting Information *

ABSTRACT: Thermoresponsive hydrogels were successfully prepared from poly(N-isopropylacrylamide)-based polymers with different architectures (linear, branched, or hyperbranched). The macromolecular architectures strongly influence the internal structure of the hydrogels, therefore modulating their thermoresponsive and rheological properties. These hydrogels were used for the in situ synthesis of gold nanoparticles. Significant changes in hydrogel microstructures and in average pore size due to the presence of gold nanoparticles were observed. Additionally, their presence significantly increases both the mechanical strength and the toughness of the hydrogel networks.

1. INTRODUCTION Response to external stimuli is central to how biological systems work, especially at the cellular level, with subsequent changes in properties and function from the molecular to the macroscopic level. The observation of various systems where such processes occur enables the design of stimuli-responsive hydrogels (i.e., temperature, pH, light,...) in a controllable and predictable fashion. In response to these external stimuli, these hydrogels have the ability to change their properties drastically (i.e., dimensions, structures, interactions, viscosity,...) and therefore have been applied as biomaterials, drug-delivery systems, or in catalysis.1,2 In most cases, physically or chemically cross-linked polymer chains are involved. Of special interest are thermoresponsive hydrogels: a small temperature change around a critical value induces the collapse or expansion of polymer chains as a response to adjustments in the interactions between the polymer chains and the aqueous media. When the phase diagrams of polymer solutions appear monophasic below a specific temperature and biphasic above, a lower critical solution temperature can be defined (LCST). Different families of thermosensitive polymers have been described in the literature such as poly(N-isopropylacrylamide) (PNIPAM), polyvinylcaprolactame,3 poly(oligo(ethylene glycol) methacrylate),4 poly(methyl vinyl ether),5 and poly(Nacryloyl-N′-propylpiperazine).6 Among them, PNIPAM, which presents an LCST of around 32 °C, has been by far the most commonly used for the formation of hydrogels.7−13 In the transition regime, water becomes a poor solvent as polymer− water H-bonds are disrupted and PNIPAM undergoes conformational changes involving both intrachain coil-toglobule transitions and interchain self-association. Those PNIPAM-based hydrogels have been used for the formation © 2015 American Chemical Society

of hybrid materials with tunable optical or catalytic properties, nanovectors for drug delivery, or smart hydrogels with striking mechanical properties.2,14 Depending on the preparation methods, hydrogels with bulk, micro-, and nanodimensions have been obtained. Whatever their overall dimensions, the properties of those hydrogels also strongly depend on their internal structure. Therefore, intensive efforts have been made to gain control over it at a nanoscale level. Indeed, finely tuning this structure yields access to materials with improved responsiveness in terms of kinetics, change of dimension,...2 Therefore, we first aim here to understand how the internal structure of the PNIPAM hydrogel can be modified by tuning the molecular architecture of the polymer. Three architectures were investigated: a linear PNIPAM polymer and two core−shell structures with PNIPAM chains grafted on either a branched or hyperbanched core. The transition rate and the rheological behavior of gels coming from the different architecture of polymers will be presented. Moreover, those hydrogels can act as reactors to mediate the formation of gold nanoparticles (NPs). Their influences on the internal structure and on the properties of bulk gels were evaluated.

2. EXPERIMENTAL SECTION 2.1. Materials. Tetrachloroauric acid trihydrate (HAuCl4, 3H2O), hydrochloride, 1,1′-carbonyldiimidazole, anhydrous DMSO, and carboxylic acid-terminated PNIPAM 2000, 5000, and 7000 g/mol were purchased from Aldrich and were used without

D-glucosamine

Received: January 5, 2015 Revised: February 14, 2015 Published: March 31, 2015 4761

DOI: 10.1021/acs.langmuir.5b00008 Langmuir 2015, 31, 4761−4768

Article

Langmuir

spin−echo (PFGSE). Experiment was carried out by monitoring the attenuation of the NMR signals during a delay surrounded by two pulsed field gradients. In practice, a series of NMR diffusion spectra were acquired as a function of the gradient strength (g). The intensities of the resonances follow an exponential decay which depends on the self-diffusion coefficient (D).18,19 Their relationship is given by the Stejskal−Tanner relation I/I0 = exp[γ2g2δ2(Δ − δ/3)D], where I is the measured signal intensity, I0 is the signal intensity for a g value of 0 G/cm, γ is the gyromagnetic ratio for the 1H nucleus, g is the field gradient magnitude, δ is the gradient pulse length, and Δ is the time between the two gradients in the pulse sequence.20 D values are a function of temperature and viscosity according to the Stokes− Einstein equation: kT/6πηRh (where k is the Boltzmann constant, T is temperature, η is the viscosity of the solution, and Rh is the radius of the solvated species). The z pulsed field gradients were generated with a 10 A GRASP II/P gradient amplifier. Thus, the z-maximum gradient strength (g) was 53.5 G/cm. Experiments were performed by varying g and keeping all other timing parameters constant. Typically, the Δ and the δ durations were 100 and 1 ms, respectively, and g was varied from 2.675 (strength of 5%) to 50.825 (strength of 95%) G/cm in 3% steps. An NMR pulse sequence with a stimulated echo bipolar gradient pulse pair and one spoil gradient was used (pulse program named stebpgp1s in the Bruker library). The data were analyzed by maximum entropy with the DOSY module of NMR notebook software (NMRtec). The DOSY processing algorithm parameter was set to 3, and the data were processed with a diffusion window from 0.1 to 25 000 μm2/s. 2.3.4. Fourier Transform Infrared (FTIR). Spectra were recorded with a Nexus Thermonicolet spectrometer equipped with a DTGS detector in attenuated total reflection (ATR) mode with a diamond crystal in the spectral region of 600−4000 cm−1 with a resolution of 2 cm−1. To estimate the grafting ratio of the core−shell polymer, an indirect calibration method was established by analyzing the IR spectrum of homogeneous mixtures of H4 and P7. The homogenized mixtures of H4 and P7 were prepared as follows: to 1 mL of an aqueous solution of P7 (25 mg/mL) was added 278.9, 230.3, 172.7, or 95.9 μL of an aqueous solution of H4 (5 mg/mL) equivalent to 100, 80, 60, and 33.3% grafting ratios. A 100% grafting degre means that all of the NH2 groups of the core were functionalized with PNIPAM chains. The homogeneous solutions were then freeze dried, and the obtained solids were analyzed by ATR-FTIR. The ratio between the intensity of the band at 1386 cm−1 (characteristic of the C−H deformation of the CH3 group for the PNIPAM chain) and that at 1641 cm−1 (stretching vibration of the CO group) was measured for each mechanical mixture. From this, a calibration curve was established (Figure SI5 in SI). Grafting ratios for core−shell structures were evaluated thanks to this calibration curve. 2.3.5. Differential Scanning Calorimetry (DSC). The thermal properties of the polymer (in solution and in the bulk) were determined by DSC using a Mettler Toledo DSC 1 STARe system thermal analysis calorimeter equipped with a GC200 gas controller. Solution samples were sealed in impermeable crucibles of 120 μL. Transition temperatures were taken at the top of the peak on the thermogram on heating at 1 °C/min. The variation in enthalpy was measured as the temperature increased at a rate of 1 °C/min. The grafting ratio was deduced from the DSC thermogram of a 0.5 wt % polymer solution. For this, the variation in enthalpy involved at the transition temperature (cloud point) of the core−shell polymer was compared to that of the P7 solution. The PNIPAM weight proportion and therefore the grafting ratio were inferred from this. 2.3.6. Transmission Electron Microscopy (TEM). A drop of the aqueous dispersion was placed on a Formvar carbon-coated copper TEM grid (Ted Pella Inc.) and left to air dry. The samples were observed with a Hitachi HT7700 transmission electron microscope operating at a 80 kV acceleration voltage. Size-distribution histograms were determined by using magnified TEM images. The size distribution of the particles was determined by measuring a minimum of 200 particles of each sample using WCIF ImageJ software. The size distributions observed were analyzed in terms of Gaussian statistics (wc(σ)).

further purification. A Pur-A-Lyzer mega dialysis kit of MWCO 3.5, 6− 8, and 12−14 kDa was purchased from Aldrich. Tris(2-aminoethyl)amine (Aldrich) was distilled under reduced pressure and stored under an argon atmosphere before use. Ultrapure water (ρ = 18 MΩ cm−1) was obtained from an Aquadem apparatus. 2.2. Synthesis. 2.2.1. Branched and Hyperbranched Polymer Synthesis. The synthesis of the hyperbranched polyamidoamine core (noted as H4) was carried out following previously published work by our group.15,16 A hyperbranched structure with a weight-average molar mass of 13 000 g·mol−1, a polydispersity equal to 2.0, and an amount of primary amine of 6.2 mmol per gram of polymer was obtained. The synthesis of core−shell polymers was performed by grafting polymer chain PNIPAM (7000 g/mol, noted as P7) using either three-branch commercial molecule tris(2-aminoethyl)amine (TREN) or H4 as the core. One gram of carboxylic acid-terminated P7 (1.43 × 10−4 mol, 1 equiv) and 25.5 mg of 1,1′-carbonyldiimidazole (2.14 × 10−4 mol, 1.1 equiv) were dissolved in 10 mL of dry DMSO. The mixture was stirred overnight under argon at room temperature (25 °C). Then, for H4P7, 1.04 mL of a solution of H4 in DMSO at 20 mg/mL (thus 20.7 mg of H4) was added slowly to the previous mixture. For the branched TP7 structure, 6.33 mg of TREN was used. The reaction was continuously stirred for 24 h under argon at room temperature. After the reaction, the mixture was dialyzed in ultrapure water for 3 days, and final core− shell polymers TP7 and H4P7 were collected after freeze drying. H4P7: 1H NMR, 300 MHz, D2O: δ 1.01 (CO−NH−CH−CH3), 1.44 (CH−CH2), 1.87 (CH−CH2, main-chain PNIPAM), 2.28 (CO− CH2−CH2−S), 2.37 (N−CH2−CH2−CO), 2.45 (N−CH2−CH2−N), 2.53 (CO−NH−CH2−CH2), 2.55 (CO−CH2−CH2−S), 2.63(N− CH2−CH2−CO), 3.23 (CO−NH−CH2−CH2), 3.75 (CO−NH− CH−CH3). 13C NMR, 75 MHz, D2O: δ 21.7 (CO−NH−CH− CH3), 27.8 (CO−CH2−CH2−S), 35.0 (CH−CH2), 35.1 (N−CH2− CH2−CO), 36.5 (CO−NH−CH2−CH2), 37.1 (CO−CH2−CH2−S), 41.8 (CO−NH−CH−CH3), 42.7 (CH−CH2, main-chain PNIPAM), 51.1 (N−CH2−CH2−CO), 52.0 (CO−NH−CH2−CH2), 53.6 (N− CH2−CH2−N), 174 (CO−NH). IR: υ̅ 3434, 3288, 3073, 2972, 2933, 2857, 1641, 1542, 1458, 1386, 1366, 1268, 1172, 1130, 1026, 975, 927, 882, 838 cm−1. 2.2.2. In Situ Synthesis of Gold Nanoparticles. Polymer (50 mg) was dissolved in 184 μL of water to form a 20 wt % homogeneous solution. Then 10 μL of a solution of 0.01 mol·L−1 HAuCl4 in water was added to the solution. Five microliters of a solution of 0.2 mol·L−1 glucosamine hydrochloride in water was added along with 1 μL of aqueous 1 mol·L−1 NaOH, and the mixture was quickly vortex mixed. Then the flask was immediately placed in the oven at 40 °C for 24 h. The solution was cooled to room temperature and analyzed by IR. To analyze the morphology of the formed gold nanoparticles, the solution has to be diluted 10-fold and deposited on a TEM grid. 2.3. Apparatus. 2.3.1. Nuclear Magnetic Resonance (NMR). To determine the structural characteristics of the polymers, NMR experiments were performed at 298 K in D2O on a Bruker Avance 300 or 500 MHz spectrometer equipped with a 5 mm z-gradient TCI cryogenic probe. The 90° pulse duration was 9 μs, the sweep width was 10 kHz, and the acquisition time was 3.5 s. The scan number was adjusted to obtain a sufficient signal-to-noise ratio, and the relaxation delay between transients was 3 s. For 1D 1H experiments, a 30° pulse was used. Attribution of the signals was deduced from COSY, HSQC, and HMBC experiments. 2.3.2. Size Exclusion Chromatography (SEC). Average number molecular weights (Mn) and polydispersity indexes (Đ) were determined by SEC on an apparatus equipped with a Waters 2140 refractive index (RI) detector using a Waters Styragel HR 4E column (eluent, THF, flow rate, 1 mL·min−1). Typically, samples at a concentration of 5 mg·mL−1 in THF were injected. Alternatively, samples were analyzed with an SEC apparatus comprising a Varian ProStar 325 UV detector (dual-wavelength analysis) and a Waters 410 refractive index detector using two Shodex K-805 L columns (8 mm, 300 mm, and 13 μm) and DMF−LiCl (1 g·L−1) as the eluent at 40 °C (flow rate 1 mL·min−1). 2.3.3. Diffusion-Ordered Spectroscopy (NMR-DOSY PFGSE).17 Coefficients of diffusion were determined by pulse field gradient 4762

DOI: 10.1021/acs.langmuir.5b00008 Langmuir 2015, 31, 4761−4768

Article

Langmuir 2.3.7. Cryo-Scanning Electron Microscopy (Cryo-SEM). Cryo-SEM images were realized with an FEG FEI Quanta 250 microscope (Japan). One drop of the sample was preheated to undergo the gel transition just before being frozen in nitrogen slush at −220 °C. The frozen sample was transferred under vacuum in the cryo-fracture apparatus (Quorum PP3000T cryo transfer system) chamber where it was fractured at −145 °C. The temperature was then decreased to −95 °C and maintained at this temperature for 30 min for sublimation. It was then metalized with Pd for 60 s and introduced into the microscope chamber, where it was maintained at −145 °C during the observation, operating at a 5 kV acceleration voltage. 2.3.8. UV−Vis Spectroscopy. Au-embedded hydrogels were cooled to room temperature prior to performing UV measurements using a BMG Labtech Spectrostar device. The wavelength range of this device was 300 to 1000 nm. For time-dependent turbidity measurements, hydrogels and Au-embedded hydrogels preformed at 40 °C were placed inside the spectrophotometer, where the temperature was set to 25 °C. The UV spectra were recorded every 30 s for 250 min. 2.3.9. Rheological Measurements. Measurements were made with an AR1000 rheometer from TA Instruments in the cone−plate configuration (diameter 2 cm, angle 2°) equipped with a Peltier plate. A stress of 1 Pa was set for viscoelastic measurements. A frequency sweep from 10 to 0.01 Hz provided the viscoelastic “spectrum” of the gel. The variation of the elastic and viscous moduli with temperature was then monitored at 1 Hz (frequency) and 1 Pa (stress) over a continuous linear temperature ramp at 1 °C·min−1. The inflection point was selected as the transition temperature.

Scheme 1. Synthesis of TREN- and HYPAM-Based Hyperbranched Polymers with a Thermoresponsive Shell

3. RESULTS AND DISCUSSION 3.1. Synthesis and Characterization of Core−Shell Copolymers. Two different branched cores were used within this study, namely, tris(2-aminoethyl)amine (TREN) and a polyamido-amine hyperbranched polymer (HYPAM). HYPAM has a structure that resembles poly(acrylamidoamine) PAMAM dendrimers. It was obtained by a one-step reaction of a hexaester (tris(2-di(methyl acrylate)-aminoethyl)amine) with TREN.15,16 A ratio between these two reactants close to 10:1 mol/mol led to a hyperbranched polymer with a molecular weight close to that of a fourth-generation PAMAM and is noted as H4. The attachment of a carboxylic acid-terminated PNIPAM (with an average molecular weight of 7000 g·mol−1 noted P7) on the two cores was carried out by using 1,1′carbonyldiimidazole (CDI) as a coupling agent in DMSO (Scheme 1). Obtained compounds TP7 and H4P7 were purified by dialysis. The disappearance of the carbonyl stretching band of the carboxylic function of P7 grafted onto the branched cores was monitored by FTIR up to completion (Figure SI4 in SI). With 1H NMR, the effectiveness of the coupling in the case of TP7 was evidenced by the appearance of an additional peak at 3.2 ppm assigned to methylene protons in the α position of the newly formed amide bond (Figure 1). For H4P7, this signal is superimposed by a proton signal of H4 (Figure SI2 in SI). Moreover, PFGSE NMR spectroscopy (Figure SI3 in SI) yields a self-diffusion coefficient for hyperbranched polymer (D = 4 × 10−11 m2/s) that is half of that of P7 (D = 8 × 10−11 m2/ s). Finally, size exclusion chromatography in DMF with 1 g·L−1 LiCl as the eluent (Figure SI1 in SI) showed a lower retention time for the two core−shell structures compared to that for P7. Additionally, no residual peaks of P7 are seen (Figure SI1 in SI). The grafting ratio was further determined by FTIR and DSC measurements (Figures SI4−SI6 in SI) and is reported in Table 1. From both techniques, the number of PNIPAM chains grafted per TREN and per H4 core was found to be around ∼2.5 and ∼25.6, respectively. Average molecular weights of

TP7 and of H4P7 were then calculated from these grafting ratios and from the molar masses of the cores. 3.2. Formation and Characterization of Thermosensitive Hydrogels. Aqueous solutions of the different PNIPAMbased polymers were studied at different concentrations ranging from 0.5 to 20%. Cloud-point temperatures (Tc) and corresponding enthalpy variations were measured by DSC (Table 1, Figure 2, and Figure SI7). For a given polymer concentration, Tc increased slightly with the degree of branching (from 29.3 to 32.4 °C at 20 wt %). As already observed in the case of block copolymers, this might result from the incorporation of a hydrophilic core within the polymer.21 The related enthalpy variation normalized with respect to PNIPAM content showed a strong decrease when the degree of branching increased. Because such a decrease is not observed under dilute conditions, this tends to demonstrate a clear difference between the different polymers concerning the polymer/polymer and polymer/solvent interactions in gel solution. Below the transition temperatures, aqueous solutions of P7, TP7, and H4P7 look like free-flowing liquids independent of the concentration. As the solutions were heated above the transition temperature, they rapidly became white and turbid. In the case of P7, the formation of a gel was observed only for the highest concentration of polymer used (i.e., 20 wt %). For the two other polymers, gel formation occurred at a lower concentration (15%). In all cases, those gelation concentrations were found to be much higher than the entanglement concentration, which can be evaluated to be around 3 wt % in the case of P7.22 Therefore, this allows us to discard the effect of molar mass on gelation properties. To confirm this, PNIPAM with a molar mass similar to that of TP7 was studied, i.e., 20000 g·mol−1. As expected, its behavior in solution is similar to that of P7 with gelation occurring at only 20 wt % (Figure SI8 in SI). Only the highest concentration (i.e., 20 wt %) will be considered in the following text. Whereas the P7 sample presents a pronounced syneresis effect after only a few minutes, (hyper-)branched structures TP7 and H4P7 4763

DOI: 10.1021/acs.langmuir.5b00008 Langmuir 2015, 31, 4761−4768

Article

Langmuir

Figure 1. 1H NMR (300 Hz, 298 K, D2O) spectra of TREN, P7, and TP7.

Table 1. Main Characteristics of the Hydrogels and of Corresponding Nanocomposites DSCb ΔH [J/g]

compounds P7 TP7 H4P7 P7@Au TP7@Au H4P7@Au

numberaverage molar mass [g/mol]a 7000 17 000 192 000

grafting ratio [%] by IR/DSC 81/83 66/71

gel pore sizes [μm] by cryo-SEM 2.8 ± 0.9 1.4 ± 0.5 0.5 ± 0.3 n.d. 0.6 ± 0.4 0.11 ± 0.08

c

rheologyb Tc (°C)

AuNP diameter [nm] by TEM

heating

cooling

heating

13.7 ± 5.5 10.5 ± 2.4 5.9 ± 2.1

30.9 14.6 8.5 25.7 11.4 10.6

2.4 3.6 8.0 11 2.6 9.2

29.3 31 32.4 30.8 31.5 32.3

Tc (°C)

cooling

G′[Pa] at 40 °C, on heating

G″[Pa] at 40 °C, on heating

heating

cooling

28.7 28.8 30.9 29.9 28.7 31.0

15 55 80 40 80 1050

30 88 135 117 150 1595

32 33 30 31 33 33

31 31 29 31 29 31

a

Estimated from the grafting ratio combined to average the molar mass of the core HYPAM obtained from SEC. bMeasurements of temperature transitions at a heating or cooling rate of 1 °C/min for polymer solution at 20 wt % (Figure SI7 in SI). cNormalized on PNIPAM content.

To gain insight into the network structure of the hydrogels, freeze-dried samples were analyzed by scanning electron microscopy at cryogenic temperature (cryo-SEM, Figure 3). The freeze-dried hydrogels show honeycomb-like microstructures with slightly dense cell walls. Nevertheless, their average size differs strongly, decreasing from 2.8 μm down to 1.4 and 0.5 μm for P7, TP7, and H4P7, respectively. Moreover, as depicted in Figure 3 some dendritic structures appear in the case of the TP7 intermediate branched structure. Therefore, a clear relationship between molecular architecture and microstructure is evidenced: for a similar chemical composition, increasing the level of branching of the polymer enables access to hydrogels with different morphologies and with smaller average pore sizes. Rheological measurements provided a quantitative characterization of the mechanical changes during the sol−gel transition. The elastic modulus G′ and the viscous modulus G″ were monitored from 25 to 40 °C in 1 °C/min steps. The oscillatory stress amplitude was set at 1 Pa, and the frequency was set at 1

Figure 2. DSC thermogram for a 20 wt % solution of H4P7 recorded at 1 °C·min−1 on heating (red line) and cooling (blue line).

hydrogels are stable for hours even if on a longer time scale (day) this syneresis effect is also observed. 4764

DOI: 10.1021/acs.langmuir.5b00008 Langmuir 2015, 31, 4761−4768

Article

Langmuir

Figure 4. Temperature-dependent elastic (empty marks) and viscous moduli (filled marks) for (A) P7, TP7, and H4P7 (20 wt % aqueous solutions) and (B) H4P7 (20 wt % aqueous solution) with and without in-situ-synthesized gold nanoparticles at a heating rate of 1 °C/min.

Figure 3. Images obtained from cryo-SEM of gel structures at 40 °C for P7, TP7, and H4P7 and for gels with in situ-synthesized gold nanoparticles. Scale bar: 2 μm.

P7, TP7, and H4P7). Both the rheological and the transitionrate results correlate with the density of gels observed by cryoSEM: gels become stiffer with a slower sol−gel transition as the degree of branching increases. 3.3. Formation of Nanocomposites. These hydrogels can act as nanoreactors to mediate the in situ formation of NPs. Hyperbranched structures based on polyamidoamine compounds (PAMAM and HYPAM) have demonstrated their ability to interact in solution with AuCl4− ions to mediate the synthesis of gold NPs.23,24 Interactions between HAuCl4 and the H4 core were evidenced by 1H NMR experiments in D2O solution: the addition of HAuCl4 to a neutral polymer induced a shift in chemical shifts only for protons of primary and tertiary amines (an upfield shift for NCH2CH2NH2 and a downfield shift for CH2NH2; Figure SI13 in SI). Thus, with H4P7, metal ions can be coordinated to the polymer tertiary amine and free primary amine functions. This interaction was exploited to obtain a homogeneous solution of gold ions within the hydrogels prior to reduction to gold NPs. The high viscosity of the gel medium discards the possibility of using a fast reducing agent such as sodium borohydride. Glucosamine was therefore selected to act as a slow reducing agent. It is mixed to the hydrogel below Tc, and then the temperature is increased to reach the gel state.25 Successful reduction was first evidenced by a change in color of the solutions from yellow to red resulting from the characteristic plasmon band of Au NPs (Figure SI14 in SI). TEM images showed particles whose sizes decreased with the level of branching. Thus P7 led to the formation of NPs with an average diameter of 13.7 ± 5.5 nm, whereas TP7 and H4P7 induced the formation of NPs with average diameters of 10.5 ± 2.4 and 5.9 ± 2.1 nm, respectively (Figure 6; see corresponding size distributions in Figure SI15). The same reduction performed in the fluid state only, below T c , led to the formation of slightly brown solutions

Hz (Figures SI9 and SI10). Figure 4A shows the variation of G′ and G″ with temperature for 20 wt % P7, TP7, and H4P7 solutions in water. A first observation is that P7 and TP7 display quite similar behavior, both exhibiting a sol−gel transition at the PNIPAM transition at 32.5 °C. The introduction of the TREN core tends to increase the stiffness of the gel, as illustrated by the change in G′ from 15 Pa in P7 to 55 Pa in TP7 at 40 °C. A more important change is displayed by H4P7, for which the transition temperature decreases from 32.5 to 30 °C. Interestingly, in the specific case of H4P7 a maximal elastic modulus G′ is observed at 32 °C at 1000 Pa. This value decreased significantly for higher temperature to reach 80 Pa at 40 °C. This might be ascribed to kinetic effects. Indeed, because of its high molar mass of H4P7, it might reorganize more slowly than P7 or TP7 to reach an equilibrium state. This was confirmed because such a phenomenon was not observed when cooling back to 25 °C (Figure S19 in SI). A good correlation is observed between this increasing G′ modulus and the density of entanglements in the gels, as can be seen on the SEM images. The thermoreversibility of these hydrogels was studied by monitoring the transition to the liquid state. As shown in Table 1 from DSC and rheological measurements, branched structures TP7 and H4P7 showed a more pronounced hysteresis effect than P7 (i.e., a significant decrease in Tc during the cooling process, also Figures SI7 and SI9 in SI). The kinetics of this return were investigated by registering the turbidity versus time for gels formed at 40 °C and then placed at 25 °C (Figures SI11 and SI12 in SI). As shown in Figure 5, branched structures required a longer time to return to the initial clear hydrated system (8, 11, and 22 min, respectively, for 4765

DOI: 10.1021/acs.langmuir.5b00008 Langmuir 2015, 31, 4761−4768

Article

Langmuir

Figure 5. (A) Time-dependent turbidity at 25 °C of TP7 20 wt % gels preformed at 40 °C with and without in-situ-synthesized gold nanoparticles. (B) Time required to return to the initial clear hydrated system for gels formed at 40 °C and then placed at 25 °C evaluated from turbidity measurements.

Figure 6. TEM images of gold nanoparticles formed in gel structures of P7, TP7, and H4P7 at 20 wt %; bar scale 200 nm, values given represent the average diamter of Au NPs.

gel structure to a filamentous-like structure. For TP7 and H4P7, the average pore size decreased from 1.4 to 0.6 and from 0.5 to 0.1 μm, respectively. This observation is consistent with rheological measurements: higher moduli were measured for gels with NP (Table 1, Figures 4B and SI9). The increase is quite low for TP7 and much stronger for H4P7, which shows a much more pronounced increase in density of the network. In the case of P7@Au, the moduli also increased a little despite the transition to the filamentous structure. Whereas TP7 and H4P7 hydrogels were white, TP7@Au and H4P7@Au were homogeneously colored hydrogels. Several cooling/heating cycles did not modify the homogeneity of these hydrogels (Figure SI16 in SI). As depicted in Figure 7, cryo-SEM images of the polymer-NPs hydrogels confirm that gold nanoparticles are well dispersed within the hydrogel. Sol−gel transition temperatures of these polymer−nanoparticle composites, extracted either from DSC or rheology measurements (Table 1 and Figures SI7, SI9, and SI10 in SI) are found to be very similar by both techniques. More pronounced differences have been previously observed for hydrogels without embedded NPs. Improved heat transfer thanks to the inclusion of NPs might be responsible for the lower discrepancy between the two techniques. Nevertheless, strong hysteresis between the heating and cooling transition (especially for TP7@Au, see Table 1) was still observed. Timedependent turbidity measurements show a noticeable difference in the rehydration process between the gels and the Au-doped gels for all polymers (Figure 5B and Figures SI11 and SI12 in SI). The latter took more time to return to the solution state,

corresponding to Au NPs with characteristic sizes below 5 nm. Those differences may first result from the interactions between PNIPAM-based structures and the surface of Au NPs through nonspecific interactions.26 These interactions result in the controlled growth of in-situ-formed NPs. Thus, the in situ formation of Au NPs in an aqueous solution of P7, TP7, and H4P7 at 0.01 wt % induced the formation of NPs with average diameters of 3.8 ± 1.2, 3.3 ± 1.9, and 4.4 ± 2.3 nm. Interestingly, the average diameters observed by reduction within the gel were significantly different. This suggests that an additional effect arising from the gel microstructure might be involved during the growth of nanoparticles. Conversely, the introduction of Au NPs induces strong modifications of microstructures as observed by cryo-SEM (Figures 3 and 7). P7@Au gel switched from a compartmented

Figure 7. Images at higher magnification of TP7@Au and H4P7@Au hydrogels formed at 40 °C. 4766

DOI: 10.1021/acs.langmuir.5b00008 Langmuir 2015, 31, 4761−4768

Langmuir

■

which is in good agreement with rheological and cryo-SEM results. As for the controlled growth of NPs, this effect might arise from interactions between the surface of Au NPs and the PNIPAM-based structures.26 Such interactions, as already observed in the literature, enable NPs to act as physical cross-linking agents that increased the stiffness of the gels (increased values of G′ and G″ in Table 1).27,28

ASSOCIATED CONTENT

S Supporting Information *

Additional UV−visible, rheological, and TEM data. This material is available free of charge via the Internet at http:// pubs.acs.org.

■

REFERENCES

(1) Kawamura, A.; Miyata, T. In Biologically Stimuli-Responsive Hydrogels; Li, Q., Ed.; Intelligent Stimuli-Responsive Materials; John Wiley & Sons, 2013; pp 335−362. (2) Sahiner, N. Soft and flexible hydrogel templates of different sizes and various functionalities for metal nanoparticle preparation and their use in catalysis. Prog. Polym. Sci. 2013, 38, 1329−1356. (3) Beija, M.; Marty, J.-D; Destarac, M. Thermoresponsive poly(Nvinyl caprolactam)-coated gold nanoparticles: sharp reversible response and easy tunability. Chem. Commun. 2011, 47, 2826−2828. (4) Lutz, J.-F.; Akdemir, Ö .; Hoth, A. Point by Point Comparison of Two Thermosensitive Polymers Exhibiting a Similar LCST: Is the Age of Poly(NIPAM) Over? J. Am. Chem. Soc. 2006, 128, 13046−13047. (5) Verdonck, B.; Goethals, E. J.; Du Prez, F. E. Block Copolymers of Methyl Vinyl Ether and Isobutyl Vinyl Ether With Thermo-Adjustable Amphiphilic Properties. Macromol. Chem. Phys. 2003, 204, 2090− 2098. (6) Gan, L. H.; Gan, Y. Y.; Deen, G. R. Poly(N-acryloyl-N′propylpiperazine): A New Stimuli-Responsive Polymer. Macromolecules 2000, 33, 7893−7897. (7) Hu, L.; Sarker, A. K.; Islam, M. R.; Li, X.; Lu, Z.; Serpe, M. J. Poly (N-isopropylacrylamide) microgel-based assemblies. J. Polym. Sci., Part Polym. Chem. 2013, 51, 3004−3020. (8) Snowden, M. J.; Chowdhry, B. Z.; Vincent, B.; Morris, G. E. Colloidal copolymer microgels of N-isopropylacrylamide and acrylic acid: pH, ionic strength and temperature effects. J. Chem. Soc., Faraday Trans. 1996, 92, 5013−5016. (9) Wu, X.; Pelton, R. H.; Hamielec, A. E.; Woods, D. R.; McPhee, W. The kinetics of poly(N-isopropylacrylamide) microgel latex formation. Colloid Polym. Sci. 1994, 272, 467−477. (10) Morris, G. E.; Vincent, B.; Snowden, M. J. Adsorption of Lead Ions ontoN-Isopropylacrylamide and Acrylic Acid Copolymer Microgels. J. Colloid Interface Sci. 1997, 190, 198−205. (11) Dai, Z.; Ngai, T. Microgel particles: The structure-property relationships and their biomedical applications. J. Polym. Sci., Part A: Polym. Chem. 2013, 51, 2995−3003. (12) Saunders, B. R.; Vincent, B. Microgel particles as model colloids: theory, properties and applications. Adv. Colloid Interface Sci. 1999, 80, 1−25. (13) Gan, D.; Lyon, L. A. Tunable Swelling Kinetics in Core−Shell Hydrogel Nanoparticles. J. Am. Chem. Soc. 2001, 123, 7511−7517. (14) Xia, L.-W.; Ju, X.-J.; Wang, W.; Chen, Q.; Chu, L.-Y. Nanostructured smart hydrogels with rapid response and high elasticity. Nat. Commun. 2013, 4, 2226. (15) Pérignon, N.; Mingotaud, A.-F.; Marty, J.-D.; Rico-Lattes, I.; Mingotaud, C. Formation and Stabilization in Water of Metal Nanoparticles by a Hyperbranched Polymer Chemically Analogous to PAMAM Dendrimers. Chem. Mater. 2004, 16, 4856−4858. (16) Saliba, S.; Kahn, M. L.; Mingotaud, C.; Haag, R.; Marty, J.-D. Hyperbranched Polymers for the Formation and Stabilization of ZnO Nanoparticles. Chem. Mater. 2010, 22, 6301−6309. (17) Johnson, C. S. Diffusion ordered nuclear magnetic resonance spectroscopy: principles and applications. Prog. Nucl. Magn. Reson. Spectrosc. 1999, 34, 203−256. (18) Price, W. S. Pulsed-field gradient nuclear magnetic resonance as a tool for studying translational diffusion: Part 1. Basic theory. Concepts Magn. Reson. 1997, 9, 299−336. (19) Price, W. S. Pulsed-field gradient nuclear magnetic resonance as a tool for studying translational diffusion: Part II. Experimental aspects. Concepts Magn. Reson. 1998, 10, 197−237. (20) Stejskal, E. O.; Tanner, J. E. Spin Diffusion Measurements: Spin Echoes in the Presence of a Time Dependent Field Gradient. J. Chem. Phys. 1965, 42, 288−282. (21) Beija, M.; Marty, J.-D.; Destarac, M. RAFT/MADIX polymers for the preparation of polymer/inorganic nanohybrids. Prog. Polym. Sci. 2011, 36, 845−886. (22) Füllbrandt, M.; von Klitzing, R.; Schönhals, A. Probing the phase transition of aqueous solutions of linear low molecular weight

4. CONCLUSIONS Thermoresponsive hydrogels have been successfully synthesized from core−shell hyperbranched polymers and PNIPAM. The molecular architecture of the polymer modulates the microscopic structure of the hydrogels and therefore its thermoresponsivity and rheological properties. These hydrogels were utilized for the in situ synthesis of well-defined gold nanoparticles. Significant changes in hydrogel microstructures and in average pore size due to the presence of gold nanoparticles were observed. Additionally, their presence significantly increases both the mechanical strength and the toughness of the hydrogel networks. Such control of the hydrogel microstructure will be taken into account in future studies to modulate the properties of the materials on the macroscopic level (catalytical, responsiveness...). Moreover, the conjugates of gels and metal nanoparticles should provide opportunities for the development of nanostructured advanced materials which may uncover applications in the promising field of supramolecular devices.

■

Article

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Author Contributions

The manuscript was written through the contributions of all authors. All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We acknowledge the EU for financial support in acquiring the DSC apparatus (FEDER-35477) and Corinne Routaboul and Pierre Lavedan from ICT FR2599 for FTIR and NMR measurements.

■

ABBREVIATIONS HYPAM, hyperbranched polyamidoamine; tris(2-aminoethyl)amine, TREN; poly(N-isopropylacrylamide), PNIPAM; cloudpoint temperature, Tc; lower critical solution temperature, LCST; nanoparticles, NPs; attenuated total reflection-Fourier transform infrared, ATR-FTIR; differential scanning calorimetry, DSC; cryo-scanning electron microscopy, cryo-SEM; transmission electron microscopy, TEM 4767

DOI: 10.1021/acs.langmuir.5b00008 Langmuir 2015, 31, 4761−4768

Article

Langmuir poly(N-isopropylacrylamide) by dielectric spectroscopy. Soft Matter 2012, 8, 12116−12123. (23) Pérignon, N.; Marty, J.-D.; Mingotaud, A.-F.; Dumont, M.; Rico-Lattes, I.; Mingotaud, C. Hyperbranched Polymers Analogous to PAMAM Dendrimers for the Formation and Stabilization of Gold Nanoparticles. Macromolecules 2007, 40, 3034−3041. (24) Zhao, M.; Crooks, R. M. Intradendrimer Exchange of Metal Nanoparticles. Chem. Mater. 1999, 11, 3379−3385. (25) Moukarzel, W.; Fitremann, J.; Marty, J.-D. Seed-less aminosugar mediated synthesis of gold nanostars. Nanoscale 2011, 3, 3285− 3290. (26) Glaria, A.; Beija, M.; Bordes, R.; Destarac, M.; Marty, J. D. Understanding the Role of ω-End Groups and Molecular Weight in the Interaction of PNIPAM with Gold Surfaces. Chem. Mater. 2013, 25, 1868−1876. (27) Kar, T.; Dutta, S.; Das, P. K. pH-Triggered conversion of soft nanocomposites: in situ synthesized AuNP-hydrogel to AuNPorganogel. Soft Matter 2010, 6, 4777−4787. (28) Appel, E. A.; Scherman, O. A. Gluing gels: A nanoparticle solution. Nat. Mater. 2014, 13, 231−232.

4768

DOI: 10.1021/acs.langmuir.5b00008 Langmuir 2015, 31, 4761−4768