Tunable Nanoparticle Stability in Concentrated ... - ACS Publications

Article pubs.acs.org/Macromolecules

Tunable Nanoparticle Stability in Concentrated Polymer Solutions On the Basis of the Temperature Dependent Solvent Quality Na Kyung Kwon,† Chang Seo Park,† Chae Han Lee,† Yung Sam Kim,‡ Charles F. Zukoski,§ and So Youn Kim*,† †

School of Energy and Chemical Engineering and ‡Department of Chemistry, Ulsan National Institute of Science and Technology (UNIST), 50 UNIST-gil, Ulsan 44919, Republic of Korea § Department of Chemical and Biological Engineering, University of Buffalo, Buffalo, New York United States S Supporting Information *

ABSTRACT: The ability to control the degree of particle dispersion in polymer solutions has been a long-standing subject in colloidal science. While a generally accepted principle is that nonadsorbing polymers can induce depletion attraction, which is mostly temperature independent, the effects of adding adsorbing polymers are still poorly understood. In this study, we investigated the effects of adsorbing polymers on the temperature-dependent stability of nanoparticles. The model systems consisted of silica nanoparticles in low-molecularweight poly(ethylene glycol) solutions. The detailed microstructures were determined with small-angle X-ray and neutron scattering measurements, while the dynamics of the temperature-dependent microstructures of the nanoparticles and polymers were probed with diffusing-wave spectroscopy. It was found that a poor solvent for polymer could drive adsorbed polymers to leave the particle substrate and return to the bulk solution due to a complicated interaction with surface, while the loss of the steric layer causes the nanoparticles to aggregate at elevated temperatures.

■

unadsorbed polymer chains.23,24 Therefore, the particles can experience both entropic repulsions and entropic depletion attractions, and the net interactions will determine the particle stability of the dispersion. Recently, Feng et al.25 reported that the dispersion of particles in 3-methacryloxypropyl trimethoxysilane- (TPM-) based colloids is temperature dependent. TPM colloids can be crystallized by the entropic depletions caused by nonadsorbing polymers at room temperature, while bridging flocculation is caused by adsorbing polymers because of the decreased polymer solubility at elevated temperatures. In between, they could control the particle stability, and a dispersed phase could be experimentally observed at the crossover of the polymer−colloid sticking energy, where the two frustrated phases do not form. On the other hand, particles may undergo more complicated depletion interactions when the depletion is only effective at short ranges, resulting in the formation of clusters that remain stable, which can be controlled by the polymer concentration and degree of polymer adsorption.24,26 Jouault et al.17 discussed the role of solvent on nanoparticle dispersion in polymer nanocomposites. They systemically studied particle dispersion in different solvents where polymers may or may not adsorb onto particle surface

INTRODUCTION Colloidal dispersions have been widely studied for several decades because of the critical roles they play in a wide range of applications.1−8 The state of particle or colloidal dispersions can vary from stable particles to flocculated particles that form a gel, glass, or crystal depending on the nature of the interparticle interactions.9−13 Electrostatic, van der Waals, and dispersion forces are often used to describe the dominant interactions that balance the repulsive and attractive forces.9,14−17 Colloids are often suspended in polymer solutions, and thus, the particle stability of colloid−polymer mixtures is of particular interest, while improving our understanding of the polymer-induced interactions has also increased in importance. It is widely known that nonadsorbing polymers induce depletion interactions.18−20 When two particles approach each other and the surface-to-surface distance becomes smaller than the characterization length-scale of the polymer, the polymers are excluded, bringing the particles closer together, which is a purely entropic and attractive interaction. Thus, one can induce attractive interactions between particles to form a gel, glass, or crystal by either decreasing the temperature or adding polymers that cause depletion interactions.21,22 If adsorbing polymer chains are added to a particle suspension, the steric repulsions arising from the adsorbed polymer chains may contribute to the particle stability, while depletion interactions may still occur because of the © 2016 American Chemical Society

Received: December 30, 2015 Revised: February 18, 2016 Published: March 11, 2016 2307

DOI: 10.1021/acs.macromol.5b02798 Macromolecules 2016, 49, 2307−2317

Article

Macromolecules

was added to the concentrated particle suspension and the solvent was quickly removed in a vacuum oven at 70 °C, creating polymer-based nanocomposites. The polymer nanocomposites were then mixed with a known amount of deionized water in a vortex mixer. To avoid the water evaporation which might occur at 70 °C experiments, samples were stored at water bath set at 70 °C with saturated humid air. The total weights of the samples were measured and recorded before and after the experiments, which ensured there was no significant evaporation during the experiments. The sample was stored at 20 °C for 2−3 h before the temperature was raised to 70 °C for the DWS measurements, while the temperature was immediately raised to 70 °C for the time-resolved SAXS experiments. The particle volume-fraction (ϕc) was varied from 0.10 to 0.25. The polymer concentration will be described in terms of the parameter Rp, which is defined as the ratio of the polymer volume to that of polymer plus the solvent; Rp was varied from 0.1 to 0.5 to ensure the particles were suspended in highly concentrated systems. Small-Angle X-ray Scattering (SAXS) Measurements. SAXS experiments were conducted at the 4C beamline of the Pohang Accelerator Laboratory (PAL) to explore the microstructure of the silica nanoparticles in the polymer solutions; a sample-to-detector distance of 5 m and radiation wavelength, λ, of 0.73 Å were employed. The scattered X-rays were recorded with a Mar charge-coupled device (CCD) area-detector. The two-dimensional SAXS patterns were then azimuthally averaged and the relative one-dimensional scattering intensity was plotted as a function of the scattering vector, q (q = (4π· sin(θ/2))/λ), where θ is the scattering angle. All particles used in the SAXS experiment were from the same synthesis batch except for the data of “7 days” in Figure 2a. To obtain the data over 7 days in a given beamtime, samples were prepared in advance, 7 days before the measurement and kept at 70 °C. The average diameter of silica nanoparticles prepared for the data after “7 days” was slightly smaller than the particles, thus the q-vector was adjusted such that q′ = qD2/D where D2 was the diameter of particles used for the data of “7 days” Fourier Transform Infrared (FT-IR) Spectroscopy. The FT-IR spectra were collected with a Shimadzu IR Tracer-100 spectrophotometer using a spectral resolution of 0.25 cm−1. Aliquots (volume = 2 μL) of the solutions were loaded into a temperature-controllable sample cell (Harrick Scientific), which consisted of two 1-mm-thick CaF2 windows separated by a 6-μm-thick Teflon spacer. To compensate for the IR absorption of the two CaF2 windows in the spectral region of interest, an FT-IR spectrum of a 1 mm-thick CaF2 window was also collected. ζ-Potential and pH Measurements. The ζ-potential of the silica nanoparticles was measured with a Zetasizer Nano ZS90 (Malvern Instrument) using disposable folded capillary cells. The light source was a He−Ne laser (λ = 633 nm, maximum power = 4 mW) and the light scattered at an angle of approximately 13° was detected. The pH of each solution was measured with a S220 SevenCompact pH/ion meter (METTLER TOLEDO) using an InlabExpert Pro-ISM electrode that consisted of a solid polymeric electrolyte stored in a 3 M buffer solution of KCl. Small-Angle Neutron Scattering (SANS) Measurements. SANS experiments were performed on the NG7 30-m SANS instrument at the National Institute of Standards and Technology (NIST) Center for Neutron Research, USA. The samples were loaded into 1 mm path-length, demountable,

depending on the solvent quality with light scattering experiments. They showed polymers in Θ solvent prefer to adsorb onto particles, providing better stability than in good solvent and concluded particle stability can be a subtle balance of electrostatic repulsion, depletion attraction, and kinetic slowdown of diffusion-limited aggregation. From these examples, it is clear that depletion interactions in the presence of adsorbing polymers involve more parameters related to the particle−polymer interactions. In this study, we investigated the effects of adsorbing polymers on the temperature-dependent nanoparticle stability, where the degree of polymer adsorption was controlled by the temperature. Although the recent study by Feng et al.25 described in detail the competing entropic and enthalpic interactions that cause the temperature-dependent phase behavior, our study revealed the role of other mechanisms in the temperature-dependent nanoparticle stability. The model systems used in this study were composed of silica nanoparticles and low-molecular-weight poly(ethylene glycol) (PEG) solutions. Silica nanoparticles and PEG solutions are widely studied because of their easy accessibility and wide variety of applications. In addition, silica nanoparticles are easily dispersed in concentrated PEG solutions.27,28 At room temperature, the particles are stable and the polymers are preferentially adsorbed. However, at elevated temperatures, the particles become less stable and form aggregates depending on the polymer concentration because the polymers leave the particle surfaces. PEG dissolved in water is known to exhibit the lower critical-solution temperature (LCST),29−31 and the closed loop, which indicates the solvent quality, gradually decreases with increasing temperature. Therefore, one might hypothesize that the polymer chains would move to the particle substrate in a poor solvent, which is presumably due to the decreased solvent quality introducing extra steric effects, resulting in more stable particle dispersions. However, our experimental results show that polymer chains leave the particle surfaces and return to the bulk with increasing temperature, and thus, particle aggregates are formed because of the decreased steric repulsions. The detailed evolution of the microstructure of the particle dispersions was explored with small-angle X-ray scattering (SAXS), diffusing-wave spectroscopy (DWS), and multiple static light-scattering techniques, while the microstructure of the polymeric layer was observed with small-angle neutron scattering (SANS).

■

METHODS Sample Preparation. The samples consisted of silica nanoparticles suspended in poly(ethylene glycol) (PEG) 400 solutions. The silica nanoparticles were synthesized according to the method reported by Stöber et al.,32 which involves the base-catalyzed hydrolysis and condensation of tetraethylorthosilicate (TEOS). It yielded silica particles with diameters (D) of 39 ± 4 nm, which was determined from both SEM micrographs and the fitting of the particle form-factor obtained from the angle dependence of the neutrons scattered from a dilute suspension. The PEG was hydroxyl-terminated and was purchased from Sigma-Aldrich. PEG 1000 was used for the small angle neutron scattering experiments. After the particles were synthesized, a mixture of particles and ethanol was concentrated approximately 10-times by heating the mixture in a ventilation hood. During this process, the excess ammonium hydroxide was removed. Then, a known amount of the PEG 2308


Article

Macromolecules

Figure 1. Temperature-dependent nanoparticle stability in polymer solutions. The particles remain stable at 20 °C and become a gel at 70 °C: (a) after 48 h with ϕc = 0.1 and Rp = 0.1 and (b) after a week with various ϕc and Rp values.

In the previous study,28 silica stability in PEG−ethanol mixtures was examined: the sum of attractive forces was maximized when polymer concentration was around 10 wt %. From the experimental similarity between these, we hypothesized the net attraction in our system is maximized near Rp = 0.1 in agreement with the observations. The details for the interaction potential will be discussed later in this article. To observe the time-dependent structural evolution in more detail, we performed SAXS experiments. The SAXS results, after subtracting the background scattering of the corresponding particle-free polymer solution, were considered to arise only from the silica nanoparticles, thereby enabling the use of the effective one-component model. The X-ray scattering intensity from a one-component material can be written as

titanium cells. The cell temperature was maintained at 20 and 70 °C ± 0.1 °C by the 10CB sample holder with a NESLAB circulating bath. A large range of scattering wave-vectors, q = 0.001−0.1 Å−1, was used by combining the sector-averaged scattering intensity from two different instrument configurations at detector distances of 4 and 13.5 m. The SANS data reduction and analysis of the scattering intensity, I versus q, was performed with the SANS reduction and analysis program IGOR Pro, which is available from NIST. A detailed discussion of the data reduction techniques is given by Kline.33 The moderate nanoparticle polydispersity was accounted for by using a Gaussian diameter-distribution to calculate an average form-factor. Fitting the experimental form-factor to the experimental data yielded a particle-size standard deviation of 0.13 D̅ , where D̅ is the average diameter. Diffusing-Wave Spectroscopy (DWS). Light scattering experiments were performed with a DWS Rheolab II (LS Instruments) in the transmission mode to obtain the autocorrelation function of the concentrated nanoparticles in the polymer solutions. The coherent source was a diode laser (λ = 658 nm; maximum power = 30 mW) and the temperature was maintained at 20 ± 0.02 °C. The samples (volume = 2 mL) were loaded into 1 mm-thick glass cuvettes and sealed. Turbidity Measurements. The gelation of the nanoparticles in the polymer solutions was monitored with a Turbiscan Online optical analyzer (Formulaction, France). This instrument employs a near-IR (λ = 850 nm) focused LED as the light source and illuminates the scattering medium in the z-direction. Thus, the time-dependent aggregation of silica nanoparticles was probed via the increasing turbidity, which was caused by the aggregation of particles in the concentrated dispersions.

I(q , ϕc) = ϕcVc Δρ2 Pc(q)Scc(q , ϕc) + B

(1)

where ϕc is the nanoparticle volume-fraction, Δρ is the excess electron scattering-length density (SLD) of the particles relative to the PEG solution phase, Vc is the particle volume, Pc(q) is the single-particle form-factor, Scc(q,ϕc) is the collective nanoparticle-structure factor (normalized to unity at large q values), and B is the background-scattering amplitude. Parts a and b of Figure 2 show the time-resolved scattering intensities for the nanoparticles at 20 and 70 °C, respectively, when ϕc = 0.1 and Rp = 0.2. The polymer concentration of Rp = 0.2 was chosen because the gelation of Rp = 0.1 was expected to be too fast, thus one may not observe the structure change in the early stage. The well-defined peak near q* = 0.009 Å−1 indicates that the nanoparticles are well dispersed and have formed a cage where the neighboring particles are separated by 69 nm (= 2π/q*). At 20 °C, all of the scattered intensities are nearly identical, with no noticeable changes in the intensity over the q-range measured. This indicates that the particles are well dispersed, with no aggregation in the Rp = 0.2 polymer solution, and it did not undergo any aggregation within a week. One may observe a slight change of scattered intensity after 7 days in Figure 2a, which originated from that the sample was separately prepared to others with slightly smaller size of particles at the same sample composition as mentioned in the Methods. The peak is located slightly right to the others; however, no upturn at low q was observed and peak width remains same. However, at 70 °C, the scattered intensity noticeably varies with time; the intensity at low q values slightly increases with increasing time. This increase at low q values indicates that the fluctuations in concentration are increasing because of the increased net attractions between the particles. The microstructure of the nanoparticles is quickly disrupted within a few hours. The well-defined peak q* = 0.009 Å−1 was significantly suppressed, indicating cage-like particle ordering is disrupted. After 24 h, the ordering of the nanoparticles has

■

RESULTS AND DISCUSSION Silica nanoparticles in PEG−water solutions at room temperature exhibit exceptional stability because of the electrostatic repulsion from the silanol (SiOH) groups34,35 and steric stabilization of the adsorbed PEG chains.28,36−38For example, nanoparticles with a particle volume-fraction, ϕc, of 0.1 and volumetric ratio of the polymer (Vpol) to the polymer plus the solvent (Vsolv) (Rp = Vpol/(Vpol + Vsolv)) of 0.1 do not undergo any phase separation at room temperature, as shown in Figure 1a. The average diameter of the silica nanoparticles was determined to be 39 ± 4 nm via SEM images and the particle formfactor fitting. Number-average polymer weight (g/mol) of the PEG is 400. However, the same composition (ϕc = 0.1, Rp = 0.1) became a gel within 48 h at 70 °C (or after a week at 60 °C). Depending on Rp and ϕc, the rate of gelation varies greatly, as shown in Figure 1b and the Supporting Information. The nanoparticles form a gel most rapidly when around ϕc = 0.1 and Rp = 0.05 or 0.1 (see Supporting Information). 2309


Article

Macromolecules

Figure 2. Scattered intensity profiles of the silica nanoparticles in PEG 400 solutions. Rp = 0.2 and ϕc = 0.1 at (a) 20 °C and (b) 70 °C. Rp = 0.5 and ϕc = 0.25 at (c) 20 °C and (d) 70 °C. The particle form-factor used to calculate the structure factor in the inset of part d is shown in part c.

Temperature-dependent particle stability at high polymer concentration is also confirmed with SAXS. For example, when Rp = 0.5 and ϕc = 0.25, a well-defined peak at approximately q* = 0.0124 Å−1 is present in the scattering profile because the well-dispersed nanoparticles have formed a cage-like structure (Figure 2, parts c and d). The distance between the particles was determined to be approximately 51 nm (= 2π/q*). This is in a good agreement with the center-to-center distance calculation for random close packing, 53 nm (= D(0.63/φc)1/3). Similar to the case of particle dispersions at Rp = 0.1, the system becomes more compressible and loses the local structures over time at high temperatures. While there were no noticeable changes at 20 °C, the width and height of the peak increases and significantly decreases, respectively, at 70 °C. The existence of peak even after 3 days indicates particles do not form a gel in a given time; however, particles at 70 °C are much less ordered than in 20 °C. The same trend was observed when Rp = 0.5 and ϕc = 0.1 (Supporting Information). The detailed microstructure of the nanoparticle dispersion can be described by the structure factor. The structure factor of each scattering intensity was obtained by dividing the scattering intensity of the concentrated particle suspension by its dilute-limit analogue (ds) at the same polymer concentration, i.e., S(q,ϕc) = [I(q,ϕc)· ϕc,ds]/[Ids(q,ϕc)·ϕc]. In the dilute-particle limit, Scc(q) = 1, and thus, Ids(q,ϕc) ≈ Pc(q). The structure factor is shown in the inset of Figure 2(d). As the time increases, the fluctuations in the concentration increase, which is indicated by the increasing Scc(0), and the particles become less ordered, which is indicated by the decreasing Scc(q*); there are no signs of aggregation yet.

almost disappeared, and thus, the scattering intensity profile resembles the particle form-factor of a single nanoparticle; the single-particle form-factor is shown in Figure 3c. While upturns at low q and the disappearance of peak indicate particle aggregation, a larger microstructure can also exist before the gel formation, which cannot be confirmed due to the limited low-q range. When Rp = 0.2, we observed that the particles were forming aggregates after 5 days at 70 °C. After 7 days, the scattering profile of the sample decays at low q values according to I(q) ≈ q‑2, indicating the mass fractal structure.39 While the increases of intensity at low q values indicate the presence of aggregates, the oscillation of the scattered intensity at high q values indicates the unperturbed size and shape of a single particle. The dispersions are quite stable at room temperature with Rp = 0.1 or 0.2. They retained the liquid-like properties for more than 9 months, indicating the gelation we reported here is not time-dependent and only observed at high temperature. The gelation of particles is retarded with the increase in Rp. When Rp increases, the attractive depletion force effectively decreases because of the reduced correlation length of the polymers.28 Furthermore, van der Waals attraction decreases because the silica and PEG are refractive index-matched; increasing polymer concentration reduces net attractions.40,41 Thus, the silica nanoparticles can remain stable without any aggregation at high values of Rp (> 0.5).27,35 Indeed, particles do not form gels at high values of Rp (> 0.5). even at 70 °C. However, the particles still exhibit the same temperaturedependent behavior. 2310


Article

Macromolecules

Figure 3. (a) SLD values for silica, PEG 1000, and Rp = 0.45 solutions with various ratios of D2O to H2O (D2O/[D2O + H2O] × 100; wt %). (b) Contrast-matched PEG layers and silica. (c) SANS intensity (I(q)) profiles at 20 °C as a function of the wave vector (q); ϕc was fixed at 0.3 and Rp = 0.45 while the D2O:H2O ratio was varied, as indicated by the legend. (d) Scattering intensity profiles at 70 °C under the same conditions as part c. The size of the errors at each point is smaller than the size of the symbols, and thus, they have been omitted for clarity.

by the adsorbed polymers, ΔρPEG.35 Adsorbed polymers near the particle substrate have the most contrast with the dispersing phase, and therefore, scatter more under the matched conditions. Detailed descriptions of the contrast-matching method for the silica/concentrated-PEG solution systems are given in the Supporting Information and the literature.35,42 The SLD, ρ, of each component is given in Figure 3a. The difference between the SLD values will determine the scattering intensity, e.g., silica scatters most when Δρsilica is the largest and the PEG layer scatters most when ΔρPEG is the largest. When Rp = 0.45, Δρsilica is equal to that of 100 wt % of deuterated water (D2O), and thus, scattering only occurs from the excess polymers near the particle substrate (Δρsilica ≈ 0). On the other hand, scattering from the adsorbed polymers is mostly suppressed with 17.5 wt % of D2O (82.5 wt % of H2O), and thus, the scattering intensity only represents the scattering from the silica particles, as illustrated in Figure 3b. The scattering is dominated by the silica nanoparticles when using 17.5 wt % D2O (labeled case “A” in Figure 3, parts a and c), which produces a scattering profile that is similar to the SAXS measurements. However, substantial changes occur in the scattering profiles at 20 °C (Figure 3c) as the SLD values change from case “A” to case “F” (Figure 3b). First, the oscillation of the scattered intensity at high q values, which represents the single-particle size and shape, has disappeared. Second, the dominant peak near q = 0.011 Å−1, which is associated with the ordering of the silica particles, has also disappeared. Finally, a new peak has emerged around q = 0.020 Å−1, which indicates the presence of adsorbed polymers.35 Further analysis

From these results, one can conclude that the stable structure of silica nanoparticles is disrupted at 70 °C regardless of polymer concentration: particles are aggregated and form gels around Rp = 0.2, or they become significantly less ordered at Rp = 0.5. The decreased nanoparticle stability could be thought of as the bridging aggregation between strongly adsorbed polymers near the particle surfaces.25 However, the PEG solution employed in this study is composed of low-molecular-weight polymers, which cannot form bridges between neighboring particles, and thus, the microstructure of the adsorbed polymers was examined in greater detail. While SAXS can be used to characterize the silica-nanoparticle dispersion, the microstructure of the polymers cannot be probed because of the much smaller electron density of PEG compared to that of silica. Thus, the microstructure of the adsorbed polymer layers was examined via SANS experiments. The intensity, I(q), of the scattered neutrons at wave vector q, has three contributions for a given ϕc: I(q) ∼ AΔρsilica 2 Pc(q)Scc(q) + BΔρsilica ΔρPEG Pc(q) Spc(q) + C ΔρPEG 2 Spp(q)

(2)

where Sij(q) are the structure factors associated with the two components (pp, pc, cc), where the subscript p(c) indicates polymer segments (particles); Δρj is the difference between the SLD of component j and the medium; and A, B, and C are constants. We matched the cross-section density of the silica nanoparticles to that of the dispersing phase, i.e., the polymer solution, thereby ensuring that the scattered intensity is caused 2311


Article

Macromolecules

4 days at 20 °C once the system reaches equilibrium, indicating no substantial change of the structure was observed at 20 °C. (Supporting Information) When the temperature is raised up to 70 °C, the relaxation time for freely moving particles immediately decreases to 7.7 ms, which is 102−103 times shorter than at 20 °C because of the increased particle mobility at the higher temperature. We hypothesize that in addition to the lower viscosity and higher thermal energy at 70 °C, the higher hydrodynamic diameter of the particles with adsorbed polymers at 20 °C contributed to the significant reduction in τ. However, relaxation time of particles in a gel in the later stage is much slower than that of particles at 20 °C. After 5 h, the C(t) curves start to show a two-step decay process with two relaxation times, with the first decay in the order of microseconds and the second decay in the order of milliseconds and two decays have different stretching factors shown in Figure 4(f). To examine the particle dynamics quantitatively, we fitted the C(t) curves with the following equation:

of these results describes the detailed structure of the adsorbed polymer segments from the particle substrate to the bulk, as well as the layer thickness on the scale of the radius of gyration of the polymers (Supporting Information).35,43 Figure 3d shows the scattering profiles after 6 h at 70 °C, with the scattering profiles changing with each D2O/H2O ratio tested. First, the intensity of the scattering profile with 17 wt % of D2O confirms the presence of less ordered nanoparticle structures at 70 °C. While substantial changes occur at 20 °C as the system approached 100 wt % of D2O (Figure 3c), the scattering profiles are less sensitive to the D2O/H2O ratio at 70 °C (Figure 3d). Even for case “F” with 100 wt % of D2O, the scattering profile is not much different from that for case “A”, i.e., resembling the single-particle form-factor for nanoparticles, and it does not indicate the existence of polymeric layers. These results suggest that the adsorbed polymer segments are removed from the particle substrate at 70 °C, and thus, the adsorbed PEG layers disappear after a few hours. These results contradict the conventional notions, i.e., the decreased solvent quality favors the adsorption of polymer chains to the particle surfaces. In our system, the particles were suspended in a PEG−water mixture, which is known to have a LCST.29−31 Raising the temperature decreases the solution quality, and thus, the particles were in a poorer solvent at 70 °C. However, under such conditions, the polymer moved to the bulk from the particle surfaces. We hypothesize that the particles experience significantly higher thermal fluctuations at elevated temperatures, which may cause the light oligo(ethylene glycol) molecules to leave the surface. Our DWS results (discussed later) show that the relaxation time of single particles at 70 °C is decreased by several orders of magnitude compared to that at 20 °C. Feng et al.25 reported that particles can be aggregated from the bridged polymer segments between the particles, i.e., the decreased solvent quality brings polymer segments to the particle surfaces rather than the bulk. However, in our system, the degree of polymerization of the PEG 400 is only 9, which is far below the entanglement molecular weight. The short chain length of the PEG 400 used in this study cannot be bridged, with the SAXS data (Figure 2b) indicating that the surface-to-surface distance between the particles is approximately 22.5 nm [(2π/q*) − D], which is much longer than the radius of gyration of PEG 400 (approximately 1 nm). In addition, we confirmed that there were no adsorbed polymers on the particle surfaces at 70 °C. Before we investigated the changes in the interactions between the particles and polymers at the interfaces, the gelation evolution at 70 °C was probed with DWS measurements. For ϕc = 0.1 and Rp = 0.1, the system becomes a gel after 48 h at 70 °C, as confirmed in Figure 1a. The change in the relaxation time of the particles from the well-dispersed state to the gel state was estimated by analyzing the intensity autocorrelation function, g2(t). Parts a and b of Figure 4 show the normalized intensity (C(t) = g2(t) − 1) decay of the correlated particles (ϕc = 0.1, Rp = 0.1) as a function of the lag time at 70 °C. At 20 °C, the well-dispersed particles only have one relaxation time related to the diffusion of particles within the cages, which is described by C(t) = [exp(−t/τ)α]2,where τ is the relaxation time of the stable particles at 20 °C and α is a stretching factor. From Figure 4a, τ was determined to be 1.9 s. We note that technical limit of the DWS for lag time lies around 10 s and scattering intensity was very low at room temperature, thus requiring longer measurement time. However, the intensity correlation function did not change for

C(t ) = [flong exp(− (t /τlong )βl ) + fshort exp(− (t /τshort )βs )]2 (3)

where f long and fshort are the fraction of particles to travel long and short distances, respectively, with f long + fshort = 1 and f long > fshort. In addition, τlong and τshort are the average relaxation times of each mode. We found that τlong was at least 102-times longer than τshort, distinguishing particles in two different modes. The exponent β represents the stretching factor for the slow dynamics in a glass or gel. The fitted results are shown in Figures 4(c), (d), and (e). While well-dispersed particles can freely diffuse within the interparticle distance, the mobility of particles in aggregates or clusters is restricted by the neighboring particles during the early stages of gelation, i.e., the particles between aggregates travel shorter distances. Therefore, the relaxation time of particles that travel much shorter distances is significantly decreased compared to that of freely diffusing particles. In addition, f long decreases and τshort increases with time as the particle mobility becomes more restricted (Figure 4, parts c and d). As the system becomes a gel, the particles cannot relax because the gel networks have formed, as shown by the nondecaying curves after 23 h. In the latter stages of gelation, the relaxation modes merge into a single mode, and the strongly interacting particles do not relax, resulting in a plateau with a very slow decay. For the latter stages of gelation, the subscripts of the terms in eq 3 are changed to avoid confusion with the early stages of gelation. C(t ) = [ftrap exp(− (t /τtrap)βt ) + frelax exp(− (t /τrelax)βr )]2 (4)

In eq 4, f trap is the fraction of particles trapped in the gel and f relax is the fraction of particles that can relax, with f trap + f relax = 1 ( f relax < f trap) and τrelax ≪ τtrap. The height of the plateau, which is indicated by f trap, increases with time. As the particles become gels, the particles are physically trapped within particle networks, and thus, both τrelax and τtrap increase greatly. After 43 h, gelation is almost complete (see Figure 1(a)), as indicated by the nondecaying correlation curves. The oscillations after such periods arise from the local fluctuations,44 as well as the strong inhomogeneity of the gel. Fitted stretching factors at each mode are shown in Figure 4(f). Before the critical gelation point (t < 23 h), a typical two-step 2312


Article

Macromolecules

Figure 4. Time-resolved intensity autocorrelation functions of the silica nanoparticles (ϕc = 0.1, Rp = 0.1) at 70 °C for selected times: (a) up to 23 h and (b) from 23 to 43 h. The correlation function after 40 h at 20 °C is given in part a. The results of fitting the plots for all times tested are shown in parts (c)−(f). The fitted curves are not shown and the fitting quality is discussed in the Supporting Information. In (f), green squares indicate βlong and βtrap and purple dots indicate βshort and βrelax.

temperature for ϕc = 0.1 and Rp = 0.1 relax with single relaxation time whose stretching factor close to 1. (Supporting Information) If polymers were strongly adsorbed at high temperatures, the silica particles would have simply settled down and would not form bridges. We have confirmed that the aggregated particles form a gel rather than settling down (see Figure 1a and the Supporting Information), indicating that the polymer segments fill up the interstitial spaces. A summary of the mechanisms involved in the temperature-dependent particle stability is shown in Scheme 1. On the basis of these observations and the results of the neutron scattering experiments, the temperature-dependent gelation behavior we have reported in this paper implies that gelation originates from another mechanism that does not involve bridging. First, we determined the change of the

decay was observed. The early decay has a stretched exponential form (βshort < 1) while the later decay has a compressed exponential form (βlong > 1). The two-step decay is commonly observed in a glass or gel.45−48 The stretched exponential relaxation usually results from a broad distribution of particle sizes as they begin to aggregate. The compressed exponential form was observed in a glass or gels46,49−52 where internal stress exists. In the gelation at ϕc = 0.1 and Rp = 0.1, particles traveling long distances are not in an equilibrium, where particles can experience internal stresses, thus showing compressed exponential decays. After the critical gelation point (t ≥ 23 h), most of particles are trapped, thus size distribution of particles becomes much broader. The strong inhomogeneity of the gel system makes particles relax with a broad range of relaxation times, therefore decay curves were stretched both in the trapped and relaxed modes. (βrelax, βtrap < 1) On the other hand, stable particles at room 2313


Article

Macromolecules

relative permittivity and refractive indices are adjusted and calculated for the Hamaker coefficients. All parameters were carefully chosen for different temperatures in the presence of polymers. The nonadsorbed polymers give a depletion attraction between the nanoparticles, which is calculated based on the polymer reference interaction site model (PRISM) integral equation-based theory.28,55 Furthermore, the steric repulsion is estimated based on the previous studies, which reported the degree of PEG 400 adsorption on the silica nanoparticles. The detailed procedures and relevant equations for the total pair interaction potential are given in the Supporting Information. Figure 5 shows the pair interaction potentials as a function of particle distance for electrostatic repulsion, van der Waals attraction, depletion attraction, steric repulsion and summed total interaction. We observed that electrostatic repulsion is not strong enough to provide long-term stability for particles both at 20 and 70 °C. At 20 °C, a contact electrostatic repulsion is only 5 kT while a strong van der Waals is dominant for the system. The depletion is only effective at near particle surfaces and not strong enough to bring phase separation in this system presumably because of the short chain length of PEG 400 and relatively high concentration of PEG. Therefore, the calculation result explicitly implies that the steric repulsion from adsorbed polymer layers is the origin of particle stability at 20 °C, which creates a repulsive barrier of 20−30 kT. We note the magnitude of the steric repulsion is determined from experimental result with some uncertainties, thus barrier height can be varied from 20 to 30 kT. For simplicity, we assumed steric repulsions do not exist at 70 °C based on SANS result. As a result, we conclude the gelation of nanoparticles at 70 °C is driven by strong van der Waals attractions in the absence of steric repulsive layers. The physical adsorption of PEG molecules onto silica can be described in terms of the H-bonding between the silica particles and PEG molecules. The surfaces of the Stöber-synthesized silica particles are covered with SiOH groups and its dissociated form, SiO−. SiOH groups in water disassociate into SiO−, which is responsible for the negatively charged surface of silica nanoparticles. The results of the ζ-potential measurements indicate that the surfaces of the silica nanoparticles are mostly covered with SiO−, and thus, form highly stable nanoparticles. In the presence of PEG, the slightly acidic nature of the PEG solution reduces the pH to 4.0, which promotes the deionization of the SiO− groups (i.e., SiO− + H+ → SiOH).56,57 To verify the changes in the surface chemistry of the silica nanoparticles,

Scheme 1. Schematic Diagram of the Thermally Controlled Particle Dispersion and Aggregation of Silica Nanoparticles in Low-Molecular-Weight PEG Solutions

interparticle potential at different temperatures according to the Derjaguin, Landau, Verwey, and Overbeek (DLVO) theory.28,53 When the temperature is raised, the change of van der Waals attractions is small because the difference between the refractive indices remains nearly identical. To confirm the changes in the electrostatic potential, which is inherently repulsive, we measured the ζ-potential of the particles (Table 1).54 Without the Table 1. ζ-Potentials of the Silica Nanoparticles in Various Solutions 20 °C 70 °C

in water (mV)

Rp = 0.1 (mV)

−39 −37

−13.7 −5.74

PEG polymers, the ζ-potential of the silica nanoparticles is −39 mV in water, while in a polymer solution with Rp = 0.1, the ζ-potential decreases to −13.7 mV at 20 °C, which is marginally stable. Although the ζ-potential of −13.7 mV may not be sufficient to ensure the long-term particle stability of the system, the presence of sterically adsorbed polymers produces additional repulsive interactions and may prevent aggregation, limiting the approach of other particles over a finite distance. To confirm the hypothesis, we have calculated total pair interaction potential (V(r)/kT) for silica nanoparticles in Rp = 0.1 PEG solutions at different temperatures, 20 and 70 °C. Electrostatic repulsion is calculated based on the ζ-potential measurement and conductivity measurement with adjusted relative permittivity values. For van der Waals attraction,

Figure 5. Total pair interaction potential V(r)/kT as a function of surface-to-surface distance (a) at 20 and 70 °C. 2314


Article

Macromolecules

Jouault et al.17 showed particle dispersion in polymeric melts can be dramatically differentiated depending on the solvent where the particles were suspended because solvent quality can affect the degree of polymer adsorption. A critical point was that particle stability is a subtle balance of electrostatic repulsion, depletion attraction, and kinetic slow down effect, emphasizing the role of casting solvent. Our study shows some similarity to theirs in that particle stability can be strongly related to the solvent quality, and steric stabilization is crucial to obtain good dispersion. However, a significant difference is also found. They reported a less good solvent or a Θ solvent drive polymers to move particle surface whereas a good solvent drives polymer to leave surface and move to the bulk. A better dispersion is found in a Θ solvent than in a good solvent similar to the Feng et al.25 while our study reports a better dispersion in good solvent. The as-described particle aggregation is partially reversible. If the temperature returns from 70 to 20 °C before the silica particles create permanent networks, i.e., within the first several hours of the early stages of gelation, the particles can be redispersed, which is similar to the recently reported H-bonding-dependent particle stability.12

Fourier transform infrared (FT-IR) spectroscopy measurements were performed. The characteristic IR band of the SiO-H stretching vibration appears at around 950 cm−1, while the Si−O−Si stretching vibration appears at around 1110 cm−1 (Figure 6). Figure 6 shows that the SiO−H signal in pure water

■

CONCLUSIONS In this study, we investigated the temperature-dependent nanoparticle stability and probed the microstructures and dynamics with scattering experiments. At room temperature, silica nanoparticles in PEG solutions are stable and do not aggregate because of the electrostatic repulsions, steric layers of adsorbed polymers, and solvation layers for the case of low molecular weights. Although other previous study reported the good stability of silica nanoparticles in high molecular weight PEG solution,37,62 the role of polymer molecular weights should be stressed more for particle stability. While steric repulsion between the particles increases with polymer molecular weights, effective length for the depletion attraction increases where size ratio between the particle and polymer is still important. When the polymer molecular weight changes; therefore, resulting depletion attractions and steric repulsions should be predicted correctly. The PEG molecules can be strongly adsorbed via H-bonding onto the silica particles covered with SiOH groups. At elevated temperatures, we confirmed that the adsorbed PEG molecules leave the silica surface because H-bonding between PEG and silica disappears. The significantly enhanced particle mobility at higher temperatures causes the PEG molecules to leave the particle surfaces, and the PEG molecules prefer to self-interact in the bulk rather than being adsorbed onto the particle surfaces. The water molecules then interact with the silica surfaces, with the particles forming aggregates via H-bonding. By considering all of the factors, the reduced particle stability at high temperatures is attributed to the loss of the steric and solvation layers, in addition to the decreased electrostatic forces. However, temperature-dependent particle stability will be qualitatively different with higher polymer molecular weights where H-bonding with surface and solvent quality might be less dependent with temperature. Our study has revealed the mechanism that controls the temperature-dependent particle stability when adsorbing polymers are involved. A remarkable result was that the adsorbed polymers leave the particle surfaces and return to the bulk in a poor solvent, and the loss of this steric layer causes aggregation. Therefore, the role of the adsorbed layers that provide steric

Figure 6. FT-IR spectra of the silica nanoparticles in water and PEG 400 solutions at different temperatures.

is very weak (see the Supporting Information), confirming the presence of strongly charged particle surfaces covered with SiO− groups, while the SiO-H signal is very clear in the PEG solution, which supports the ζ-potential measurements. Since the physical adsorption of PEG molecules relies on the H-bonding between the SiOH groups and OH groups (or R−O−R′ groups) of PEG, the increased concentration of SiOH groups helps the PEG molecules be adsorbed onto the particles. At 20 °C, both the steric layer of adsorbed PEG molecules and the solvation layer contribute to the particle stability. Raghaven et al.36 reported that the H-bonding of polyether molecules with OH end-groups is able to form a solvation layer on silica particles.58 Oligomers have additional H-bonding between the SiOH groups and OH groups, which is appreciably stronger than that between the SiOH groups and R−O−R′ groups.59 Thus, this repulsive solvation layer provides an additional stabilization force for the particles. At 70 °C, the surface-charge ζ-potential drops to −5.74 mV with a broader distribution (see the Supporting Information), implying a significantly reduced electrostatic force between the particles. The FT-IR results also show a more pronounced vibrational peak for SiO−H at 950 cm−1, which corresponds well to the ζ-potential drop. In addition, there is a large split in the Si−O−Si stretching band at 70 °C. The PEG segments and water molecules must compete to interact with the silica particles via H-bonding. While PEG segments are preferentially adsorbed at 20 °C, the desorbed PEG segments promote the approach of water molecules to particle surface, increasing the number of SiOH groups at 70 °C. Losing the repulsive steric and solvation layers greatly affects the stability of the particle dispersion, and thus, the particles start to aggregate. The aggregated particles can interact with each other via H-bonding between the SiOH groups, as shown in the weakly H-bonding liquid in Scheme 1.60,61 On the other hand, the particles will experience an increasing depletion force because of the increasing polymer concentration in the bulk. 2315


Article

Macromolecules

(9) DemirciogluD.Surface Active Silica Sols: Effect of PEG-Silica interactions; M.S. Thesis, Chalmers University of Technology: Goteborg, Sweden, 2011 (10) Glomann, T.; Schneider, G. J. A microscopic view on the large scale chain dynamics in nanocomposites with attractive interactions. Soft Matter 2013, 9, 10559. (11) Kim, S. Y.; Zukoski, C. F. Nanoparticle Microstructure in Concentrated Polymer Solutions: Role polymer Segment-Particle Surface Interactions. Langmuir 2011, 27, 10455. (12) Heo, K.; Miesch, C.; Emrick, T.; Hayward, R. C. Thermally Reversible Aggregation of Gold Nanoparticles in Polymer Nanocomposites through Hydrogen Bonding. Nano Lett. 2013, 13 (11), 5297−5302. (13) Jouault, N.; Lee, D.; Zhao, D.; Kumar, S. K. Block-CopolymerMediated Nanoparticle Dispersion and Assembly in Polymer Nanocomposites. Adv. Mater. 2014, 26 (24), 4031−4036. (14) Segets, D.; Marczak, R.; Schafer, S.; Paula, C.; Gnichwitz, J. F.; Hirsch, A.; Peukert, W. Experimental and Theoretical Studies of the Colloidal Stability of Nanoparticles - A General Interpretation Based on Stability Maps. ACS Nano 2011, 5 (6), 4658−4669. (15) Vigil, G.; Xu, Z. H.; Steinberg, S.; Israelachvili, J. Interactions of Silica Surfaces. J. Colloid Interface Sci. 1994, 165 (2), 367−385. (16) Ducker, W. A.; Xu, Z. G.; Israelachvili, J. N. Measurements of Hydrophobic and DLVO Forces in Bubble-Surface Interactions in Aqueous-Solutions. Langmuir 1994, 10 (9), 3279−3289. (17) Jouault, N.; Zhao, D.; Kumar, S. K. Role of Casting Solvent on Nanoparticle Dispersion in Polymer Nanocomposites. Macromolecules 2014, 47 (15), 5246−5255. (18) Asakura, S.; Oosawa, F. On Interaction Between 2 bodies Immersed in a Solution of Macromolecules. J. Chem. Phys. 1954, 22 (7), 1255−1256. (19) Degennes, P. G. POLYMER-SOLUTIONS NEAR AN INTERFACE 0.1. ADSORPTION AND DEPLETION LAYERS. Macromolecules 1981, 14 (6), 1637−1644. (20) Dinsmore, A. D.; Yodh, A. G.; Pine, D. J. Entropic Control of Particle Motion Using Passive Surface Microstructures. Nature 1996, 383 (6597), 239−242. (21) Tuinier, R.; Rieger, J.; de Kruif, C. G. Depletion-induced phase separation in colloid-polymer mixtures. Adv. Colloid Interface Sci. 2003, 103 (1), 1−31. (22) Ye, X.; Narayanan, T.; Tong, P.; Huang, J. S.; Lin, M. Y.; Carvalho, B. L.; Fetters, L. J. Depletion Interactions in ColloidPolymer Mixtures. Phys. Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top. 1996, 54 (6), 6500−6510. (23) Kleshchanok, D.; Lang, P. R. Steric Repulsion by Adsorbed Polymer Layers Studied with Total Internal Reflection Microscopy. Langmuir 2007, 23 (8), 4332−4339. (24) Kim, S.; Hyun, K.; Moon, J. Y.; Clasen, C.; Ahn, K. H. Depletion Stabilization in Nanoparticle-Polymer Suspensions: Multi-LengthScale Analysis of Microstructure. Langmuir 2015, 31 (6), 1892−1900. (25) Feng, L.; Laderman, B.; Sacanna, S.; Chaikin, P. Re-entrant Solidification in Polymer-Colloid Mixtures As a Consequence of Competing Entropic and Enthalpic Attractions. Nat. Mater. 2015, 14 (1), 61−65. (26) Sciortino, F.; Mossa, S.; Zaccarelli, E.; Tartaglia, P. Equilibrium cluster phases and low-density arrested disordered states: The role of short-range attraction and long-range repulsion. Phys. Rev. Lett. 2004, 93 (5), 055701. (27) Kim, S. Y.; Hall, L. M.; Schweizer, K. S.; Zukoski, C. F. Long Wavelength Concentration Fluctuations and Cage Scale Ordering of Nanoparticles in Concentrated Polymer Solutions. Macromolecules 2010, 43 (23), 10123−10131. (28) Kim, S. Y.; Zukoski, C. F. Particle Restabilization in Silica/PEG/ Ethanol Suspensions: How Strongly do Polymers Need To Adsorb To Stabilize Against Aggregation? Langmuir 2011, 27 (9), 5211−5221. (29) Kjellander, R.; Florin, E. Water-Structure and Changes in Thermal-Stability of the System Poly(Ethylene Oxide)-Water. J. Chem. Soc., Faraday Trans. 1 1981, 77, 2053.

stabilization and interactions between the particles and polymer segments must be considered to understand the temperaturedependent particle stability completely, where increasing the temperature does not always guarantee an increase in the particle stability.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.macromol.5b02798. Temperature-dependent particle stability, turbidimetry, DWS fitting quality, ζ-potential measurements, further analysis of the SANS experiments, and total interaction potentials (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*(S.Y.K). E-mail: [email protected]. Telephone: +82-52217-2558. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the 2014 Research Fund (1.140046.01) of the Ulsan National Institute of Science and Technology (UNIST) and by the Basic Science Research Program through the National Research Foundation of Korea (NRF), which was funded by the Ministry of Education (NRF2014R1A1A2056774). Y.S.K acknowledges financial support from the National Research Foundation of Korea (Grant 20110015061). The SAXS experiments in this work were performed at beamline 4C of the Pohang Accelerator Laboratory (PAL). The work involving the SANS facilities was supported in part by the National Science Foundation under Agreement No. DMR-0944772. Thus, we acknowledge the support of the National Institute of Standards and Technology of the U.S. Department of Commerce, who provided the neutron research facilities used in this work.

■

REFERENCES

(1) Poon, W. C. K. The physics of a model colloid-polymer mixture. J. Phys.: Condens. Matter 2002, 14 (33), R859−R880. (2) Ballauff, M.; Lu, Y. ″Smart″ Nanoparticles: Preparation, Characterization and Applications. Polymer 2007, 48 (7), 1815−1823. (3) Sayari, A.; Hamoudi, S. Periodic mesoporous silica-based organic - Inorganic nanocomposite materials. Chem. Mater. 2001, 13 (10), 3151−3168. (4) Balazs, A. C.; Emrick, T.; Russell, T. P. Nanoparticle Polymer Composites: Where Two Small Worlds Meet. Science 2006, 314 (5802), 1107−1110. (5) Norio Ise, I. S. Structure Formation in Solution: Ionic Polymers and Colloidal Particles; Springer Science & Business Media: 2005. (6) Isojima, T.; Lattuada, M.; Vander Sande, J. B.; Hatton, T. A. Reversible Clustering of pH- and Temperature-Responsive Janus Magnetic Nanoparticles. ACS Nano 2008, 2 (9), 1799−1806. (7) Zhang, Y.; Liu, Q. K.; Mundoor, H.; Yuan, Y.; Smalyukh, I. I. Metal Nanoparticle Dispersion, Alignment, and Assembly in Nematic Liquid Crystals for Applications in Switchable Plasmonic Color Filters and E-Polarizers. ACS Nano 2015, 9 (3), 3097−3108. (8) Thomin, J. D.; Keblinski, P.; Kumar, S. K. Network Effects on the Nonlinear Rheology of Polymer Nanocomposites. Macromolecules 2008, 41 (16), 5988−5991. 2316


Article

Macromolecules (30) Saeki, S.; Kuwahara, N.; Nakata, M.; Kaneko, M. Upper and Lower Critical Solution Temperature in Poly(Ethylene Glycol) Solutions. Polymer 1976, 17 (8), 685−689. (31) Polik, W. F.; Burchard, W. Static Light-Scattering From Aqueuos Poly(Ethylene Oxide) Solutions in the Temperature-Range 20−90-Degrees-C. Macromolecules 1983, 16 (6), 978−982. (32) Stober, W.; Fink, A.; Bohn, E. Controlled Growth of Monodisperse Silica Spheres in the Micron Size Range. J. Colloid Interface Sci. 1968, 26 (1), 62−69. (33) Kline, S. R. Reduction and analysis of SANS and USANS data using IGOR Pro. J. Appl. Crystallogr. 2006, 39, 895−900. (34) Philipse, A. P.; Vrij, A. Preperation and Properties of Nonaqueous Model Didspersions of Chemically Modified, Charged Silica Spheres. J. Colloid Interface Sci. 1989, 128 (1), 121−136. (35) Kim, S. Y.; Schweizer, K. S.; Zukoski, C. F. Multiscale Structure, Interfacial Cohesion, Adsorbed Layers, and Thermodynamics in Dense Polymer-Nanoparticle Mixtures. Phys. Rev. Lett. 2011, 107 (22), 225504. (36) Raghavan, S. R.; Walls, H. J.; Khan, S. A. Rheology of Silica Dispersions in Organic Liquids: New Evidence for Solvation Forces Dictated by Hydrogen Bonding. Langmuir 2000, 16 (21), 7920−7930. (37) Qiu, D.; Flood, C.; Cosgrove, T. A small-angle neutron scattering study of adsorbed polymer structure in concentrated colloidal dispersions. Langmuir 2008, 24 (7), 2983−2986. (38) Zhang, Z.; Berns, A. E.; Willbold, S.; Buitenhuis, J. Synthesis of Poly(Ethylene Glycol) (PEG)-Grafted Colloidal Silica Particles with Improved Stability in Aqueous Solvents. J. Colloid Interface Sci. 2007, 310 (2), 446−455. (39) Schmidt, P. W. Small-Angle Scattering Studies of Disordered, Porous and Fractal Systems. J. Appl. Crystallogr. 1991, 24, 414−435. (40) Zivkovic, N.; Serbanovic, S.; Kijevcanin, M.; Zivkovic, E. Volumetric Properties, Viscosities, and Refractive Indices of the Binary Systems 1-Butanol + PEG 200, + PEG 400, and + TEGDME. Int. J. Thermophys. 2013, 34 (6), 1002−1020. (41) Leviton, D. B.; Frey, B. J. Temperature-dependent Absolute Refractive Index Measurements of Synthetic Fused Silica - art. no. 62732K. In Optomechanical Technologies for Astronomy, Pts 1 and 2, AtadEttedgul, E., Antebi, J., Lemke, D., Eds. Spie-Int Soc Optical Engineering: Bellingham, 2006; Vol. 6273, pp K2732−K2732. (42) Kim, S. Y.; Meyer, H. W.; Saalwachter, K.; Zukoski, C. F. Polymer Dynamics in PEG-Silica Nanocomposites: Effects of Polymer Molecular Weight, Temperature and Solvent Dilution. Macromolecules 2012, 45 (10), 4225−4237. (43) Growney, D. J.; Mykhaylyk, O. O.; Derouineau, T.; Fielding, L. A.; Smith, A. J.; Aragrag, N.; Lamb, G. D.; Armes, S. P. Star Diblock Copolymer Concentration Dictates the Degree of Dispersion of Carbon Black Particles in Nonpolar Media: Bridging Flocculation versus Steric Stabilization. Macromolecules 2015, 48 (11), 3691−3704. (44) Zhang, F.; Allen, A. J.; Levine, L. E.; Ilavsky, J.; Long, G. G. Structure and Dynamics Studies of Concentrated Micrometer-Sized Colloidal Suspensions. Langmuir 2013, 29 (5), 1379−1387. (45) Schupper, N.; Rabin, Y.; Rosenbluh, M. Multiple Stages in the Aging of a Physical Polymer Gel. Macromolecules 2008, 41 (11), 3983−3994. (46) Schosseler, F.; Kaloun, S.; Skouri, M.; Munch, J. P. Diagram of the Aging Dynamics in Laponite Suspensions at Low Ionic Strength. Phys. Rev. E 2006, 73 (2), 021401. (47) Ruta, B.; Chushkin, Y.; Monaco, G.; Cipelletti, L.; Pineda, E.; Bruna, P.; Giordano, V. M.; Gonzalez-Silveira, M. Atomic-Scale Relaxation Dynamics and Aging in a Metallic Glass Probed by X-Ray Photon Correlation Spectroscopy. Phys. Rev. Lett. 2012, 109 (16), 165701. (48) Cipelletti, L.; Ramos, L.; Manley, S.; Pitard, E.; Weitz, D. A.; Pashkovski, E. E.; Johansson, M. Universal Non-Diffusive Slow Dynamics in Aging Soft Matter. Faraday Discuss. 2003, 123, 237−251. (49) Cipelletti, L.; Manley, S.; Ball, R. C.; Weitz, D. A. Universal Aging Features in the Restructuring of Fractal Colloidal Gels. Phys. Rev. Lett. 2000, 84 (10), 2275−2278.

(50) Guo, H.; Wilking, J. N.; Liang, D.; Mason, T. G.; Harden, J. L.; Leheny, R. L. Slow, Nondiffusive Dynamics in Concentrated Nanoemulsions. Phys. Rev. E 2007, 75 (4), 041401. (51) Bellour, M.; Knaebel, A.; Harden, J. L.; Lequeux, F.; Munch, J. P. Aging Processes and Scale Dependence in Soft Glassy Colloidal Suspensions. Phys. Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top. 2003, 67 (3), 031405. (52) Bandyopadhyay, R.; Liang, D.; Yardimci, H.; Sessoms, D. A.; Borthwick, M. A.; Mochrie, S. G. J.; Harden, J. L.; Leheny, R. L. Evolution of Particle-Scale Dynamics in an Aging Clay Suspension. Phys. Rev. Lett. 2004, 93 (22), 228302. (53) Sainis, S. K.; Merrill, J. W.; Dufresne, E. R. Electrostatic Interactions of Colloidal Particles at Vanishing Ionic Strength. Langmuir 2008, 24 (23), 13334−13347. (54) Antonio Alves Junior, J.; Baptista Baldo, J. The Behavior of Zeta Potential of Silica Suspensions. New J. Glass Ceram. 2014, 4, 29. (55) Chatterjee, A. P.; Schweizer, K. S. Influence of Solvent Quality and Thermal Fluctuations on Polymer-mediated Depletion Interactions. Macromolecules 1999, 32 (3), 923−934. (56) Hampton Research, PEG Stability: A Look at pH and Conductivity Changes over Time in Polyethylene Glycols. Hampton Research Corporation; 2012. http://hamptonresearch.com/ documents/growth_101/27.pdf (accessed March, 2016). (57) Campen, R. K.; Pymer, A. K.; Nihonyanagi, S.; Borguet, E. Linking Surface Potential and Deprotonation in Nanoporous Silica: Second Harmonic Generation and Acid/Base Titration. J. Phys. Chem. C 2010, 114, 18465. (58) Neyertz, S.; Brown, D. Interface between End-Functionalized PEO Oligomers and a Silica Nanoparticles Studied by Molecular Dynamics Simulations. Macromolecules 2004, 37, 10109. (59) Korn, M.; Killmann, E.; Eisenlauer, J. Infrafed and MicroOrimetric Studies of the Adsorption of Mono-Functional Ketones, Esters, Ethers, and Alcohols at the Silica-Carbon Tetrachloride Interface. J. Colloid Interface Sci. 1980, 76 (1), 7−18. (60) Kobayashi, T.; DiVerdi, J. A.; Maciel, G. E. Silica Gel Surface: Molecular Dynamics of Surface Silanols. J. Phys. Chem. C 2008, 112 (11), 4315−4326. (61) Christy, A. A. Effect of Heat on the Adsorption Properties of Silica Gel. IACSIT Int. J. Eng. Technol. 2012, 4, 484. (62) Zhang, Q.; Archer, L. A. Optical Polarimetry and Mechanical Rheometry of Poly(Ethylene Oxide)-Silica Dispersions. Macromolecules 2004, 37 (5), 1928−1936.

2317
