Segmental dynamics and physical aging of polystyrene/silver nanocomposites

Abstract

We investigate the effects of silver (Ag) nanoparticles on the segmental and chain dynamics, physical aging and rheological behavior of polystyrene (PS) via a combination of broadband dielectric spectroscopy, calorimetry, and dynamic rheological measurement. The segmental dynamics of PS is found to be unchanged with increasing nanoparticle loading. After annealing below the glass transition temperature (T_g) for various time periods and measuring the recovered enthalpy values of PS, it is surprising that an acceleration and a suppression of the physical aging in PS/Ag-3% and 10% nanocomposites can be observed, respectively, corresponding to the decreased and increased calorimetric T_g, which can be interpreted by plasticizing and antiplasticizing effects. Furthermore, the filler reinforcement in rheological behavior is observed with increased weight fraction of Ag nanoparticles. The temperature-dependent horizontal shift factor reveals that the overall chain dynamics speed up in the presence of Ag nanoparticles. We also emphasize recent discrepancies in the prior studies of polymer nanocomposites and polymer thin films by comparing results.

---

1. Introduction

Introduction of nanoparticles into polymers generates polymer nanocomposites (PNCs) with a combination of outstanding comprehensive properties because of the integration of stiffness and stability of nanoparticles with the excellent properties of polymers such as high transparency, mechanical and dielectric properties and processability. More particularly, due to the unique electrical conductivity and antibacterial characteristics of silver (Ag) nanoparticles, polymer/Ag hybrid advanced materials have a vast array of applications in microelectronic devices, sensors and biomedical applications. A fundamental understanding about polymer–nanoparticle interactions and dynamics will be of great significance to optimize structure and properties of such materials. Unfortunately, these factors have been poorly understood to date.^1,2

The glass transition temperature (T_g), which is directly related to the segmental dynamics, is one of the most fundamental parameters for amorphous polymers. In recent decades, the effects of nanoparticles on the T_g of PNCs have been extensively investigated. Due to the complexity of interactions, including polymer–nanoparticle, polymer–polymer and nanoparticle–nanoparticle, a diverse range of effects can be manifested. Green et al.³ found that the T_gs of polymer–C₆₀ mixtures increase and local polymer chain motions in the glassy state are suppressed compared with bulk polymers. Cabral et al.⁴ and Ding et al.⁵ also reported similar deviations of T_g in other polymer–C₆₀ nanocomposite systems. Schönhals et al.^6,7 found that the T_g is 30 K lower than that of pure polymers in polyethylene (PE)- and polypropylene (PP)-based layered double hydroxide (LDH) nanocomposites. Pandis et al.⁸ also revealed that the glass transition becomes systematically faster in poly(methyl methacrylate) (PMMA)/Ag nanocomposites. Moreover, Green et al.⁹ observed that PNCs may exhibit an increased or decreased T_g through controlling the grafting density and chain length of nanoparticles. On the other hand, it has been reported that due to the reduced segmental mobility of bound layers in the vicinity of nanoparticles, PNCs containing different particles may even exhibit two T_gs, corresponding to close-bound and bulk chains.^10,11 However, other investigators claimed no change in the segmental dynamics associated with T_g for PNCs.^12–15 Therefore, it is of great importance to interpret the basis for such discrepancies.

It is well known that amorphous polymers in the glassy state are in a nonequilibrium state and exist with an excess of thermodynamic quantities such as volume or enthalpy. Note that physical aging refers to structure relaxation toward equilibrium, and it is generally accompanied by significant changes in physical properties such as mechanical strength, modulus and viscoelastic properties.^16–18 The investigation of physical aging of PNCs will be of great significance because it is closely related to the long-term storage stability of PNC materials. However, similar to the abovementioned T_g results, the reported physical aging process of PNCs also lack consistency in previous studies. Torkelson et al.^19–21 found that the physical aging is suppressed in the presence of nanoparticles; moreover, many other studies also show the same reduction trends.^22–24 Nevertheless, Boucher et al.^25–27 demonstrated that physical aging speeded up in PMMA/silica and polystyrene (PS)/gold nanocomposites. Such controversies are generally attributed to the possible interactions between polymer and nanoparticles in PNCs.²⁷ However, as of now, the mechanisms responsible for physical aging processes in PNCs are not well understood.

In this study, the influence of Ag nanoparticles on the dynamics, physical aging and rheological behavior of PS are systematically investigated by means of broadband dielectric spectroscopy (BDS), calorimetry, and dynamic rheological measurements. The invariant segmental dynamics of PS in the presence of Ag nanoparticles was investigated through BDS isothermal frequency sweep. The physical aging and calorimetric T_g of PS/Ag nanocomposites were studied by monitoring the recovered enthalpy using differential scanning calorimetry. Moreover, the filler reinforcement and the acceleration of overall chain dynamics in the nanocomposites were probed by rheological measurements.

2. Experimental section

2.1. Materials and sample preparation

Polystyrene (PS, M_w = 140 kg mol⁻¹, M_w/M_n = 2.1) was purchased from Sigma-Aldrich. Ag nanoparticles with an average particle size of 50 nm were obtained from Beijing DK Nano technology Co. Ltd., China. All materials were used as received. Solution mixing and rapid precipitation as described elsewhere^2,4 were employed to prepare well-dispersed PS/Ag nanocomposites used in this study. Ag nanoparticles were dispersed in tetrahydrofuran (THF) and ultrasonicated for 30 min in a water bath. Then, the PS was dissolved in the THF solution at 3% by weight with continuous stirring for 2 h, and bath-sonicated again for 30 min. Finally, the solution was precipitated in 500 mL methanol, washed with deionized water and filtered. The samples were maintained at room temperature for 24 h, and then transferred to a vacuum oven at 403 K (well above the T_g of PS) for at least 72 h to remove residual solvent. Thermogravimetric analysis (TGA) was performed to verify the nanoparticle weight fraction and the complete evaporation of solvent. PS/Ag nanocomposites at various nanoparticle concentrations (1, 3, 5, 10 wt%) were prepared using this method. The samples were compression-molded at 10 MPa and 160 °C into approximately 100 μm thick films, and disc samples with a diameter of 25 mm and a thickness of 1.5 mm for dielectric and rheological measurements, respectively.

2.2. Characterization of the nanocomposites

A transmission electron microscope (TEM, JEM-1230, JEOL, Japan) at an acceleration voltage of 100 kV was employed to observe the nanoparticle dispersion morphology in the nanocomposites. Nanocomposite ultra-thin sections of about 100 nm in thickness were microtomed at room temperature from the prepared disc samples.

Broadband dielectric spectroscopy (BDS) measurements were conducted on a Novocontrol Alpha high-resolution dielectric analyzer (Novocontrol GmbH Concept 40, Novocontrol Technology, Germany), and the temperature was controlled by a Novocool cryogenic system with a precision of ±0.1 K during the experiments. The 100 μm thick films were placed between two 20 mm diameter parallel gold electrodes. Isothermal frequency sweeps recording the dielectric function every 3 K were carried out over the temperature range from 363 to 433 K in the frequency range of 10⁻¹ to 10⁷ Hz. Temperature sweeps were performed from 273 to 433 K at the frequency of 10 Hz and the heating rate of 3 K min⁻¹.

Differential scanning calorimetry (DSC) measurements were performed on a differential scanning calorimeter (Q100, TA, USA) to detect the T_g and physical aging of PS/Ag nanocomposites. Pure indium was employed for temperature calibration. All measurements were carried out under a nitrogen atmosphere. For the T_g measurement, the samples were first heated to 423 K and kept for 5 min to eliminate thermal history, and then cooled to 313 K to obtain the T_g as determined by the TA Universal Analysis 2000 software. For the measurement of physical aging, all samples were first heated to 423 K and kept for 5 min to eliminate thermal history, and subsequently cooled to 313 K at the rate of 20 K min⁻¹. Afterwards, the samples were aged in the DSC at the aging temperature (T_a) and kept for enthalpy relaxation at various aging time (t_a) from 5 to 320 min, prior to being cooled to 313 K at the cooling rate of 20 K min⁻¹; finally, they were reheated to 423 K at 10 K min⁻¹ for data collection. For longer-time physical aging measurements, the annealing of the samples was carried out in a vacuum oven at T_a for various aging times (from 640 to 2560 min) after erasing the thermal history and quenching the samples in the DSC. The samples were quenched again after aging and DSC thermograms were recorded for data collection.

The rheological measurements were carried out on an advanced rheometric expansion system (ARES-G2, TA, USA) with a parallel plate geometry of 25 mm diameter. Isothermal frequency sweeps recording the viscoelastic properties every 10 K were applied in the frequency range of 10⁻² to 10² rad s⁻¹ from 433 to 473 K. The strain amplitude was 1%, ensuring that all measurements were in the linear viscoelastic region.

3. Results and discussion

3.1. Characterization of PS/Ag nanocomposites

Fig. 1 illustrates the quality of Ag dispersion in PS/Ag nanocomposites at the weight fractions of 3% and 10%. From Fig. 1 one can estimate that the Ag nanoparticles show an average diameter of 50 nm, consistent with the quantitative measurement results using the TEM images and Nano Measurer software, as shown in Fig. S1.† The average diameter of Ag nanoparticles was found to be 52.1 nm with a relatively wide polydispersity, according to a large number of statistical results. Generally, dispersion is reasonably homogeneous across the observed sample area, despite a visible aggregation of Ag nanoparticles being detectable in the PNC samples. This indicates that even after sonication treatment the aggregation of nanoparticles was noticeable, although the solution mixing and rapid precipitation method worked well. On the other hand, the high temperature (403 K) drying process for a long time (72 h) during sample preparation could induce the formation of aggregates, as described in other PNC systems upon annealing.²⁸ Sangoro et al.² and Boucher et al.²⁷ also reported the formation of aggregates in poly(2-vinylpyridine) (P2VP)/oxide nanoparticles and PS/gold nanocomposites. They proposed that the aggregation of the nanoparticles does not play a major role in the molecular dynamics and physical aging of nanocomposites, provided the overall dispersion was good.

Fig. 1 TEM images of nanoparticle distributions in (a) PS/Ag-3% and (b) PS/Ag-10% nanocomposites.

3.2. Segmental dynamics of PS/Ag nanocomposites

Firstly, we investigate the effect of Ag nanoparticles on the segmental dynamics of PS. Fig. 2 shows the temperature dependencies of dielectric loss ε′′ for different PS/Ag nanocomposites. It should be noted that the temperature and main relaxation time corresponding to the maximum of dielectric loss (ε^′′_max) is approximately equal for pure PS and its nanocomposites, indicating that the segmental dynamics of PS is not obviously altered by introduction of Ag nanoparticles into the PS matrix. This is similar to the results of PS/gold nanocomposites reported by Boucher et al.²⁷ Moreover, Fig. 2 shows that the width of the dielectric loss function is not noticeably changed by the addition of Ag nanoparticles, in agreement with results on PMMA/silica nanocomposites,²⁶ which can be attributed to the unchanged heterogeneous dynamics. This differs from the significant broadening of dielectric spectra found in PS/gold nanocomposites.²⁷ It is expected that a broader spectral shape of segmental relaxation can be exhibited due to the increasing heterogeneities in the presence of nanoparticles.^15,29 However, neither the main relaxation time nor the width of the spectra is affected by nanoparticle loading in PS/Ag nanocomposites presented here. The detailed analysis of the shape parameters is also discussed in detail in what follows.

Fig. 2 Temperature dependencies of dielectric loss at a frequency of 10 Hz for pure PS and PS/Ag nanocomposites.

To further quantify the effect of Ag nanoparticles on the segmental dynamics of PS, the detailed analysis of relaxation time should be examined. Fig. 3 shows the frequency dependencies of dielectric loss ε′′ for pure PS. In the temperature range investigated, one relaxation peak identified as α-relaxation, which is associated with the segmental motion of PS chains, can be observed. A similar trend for PS/Ag nanocomposite results can be found in Fig. S2.† In the literature reported previously,^27,29–31 a weak second relaxation process (β*) of PS was displayed in the high frequency region, although the mechanism remains ambiguous. Note that the β*-process was not clearly observed in our study.

Fig. 3 Dielectric loss as a function of frequency for pure PS at various temperatures. The solid lines are fits to the data using the HN function, including the conductivity and interfacial process contributions. The dashed dotted lines represent the α-relaxation process. The dashed lines are the contribution of the conductivity.

In order to extract quantitative information from the isothermal dielectric measurements, the complex dielectric function was analyzed by using the empirical Havriliak–Negami (HN) function.³²

in which ω is angular frequency (ω = 2πf), Δε is the dielectric strength, ε_∞ is the unrelaxed value of the dielectric constant, τ_HN is the HN relaxation time, and the exponents α_HN and β_HN are shape parameters describing the symmetric and asymmetric broadening of the spectra, respectively. An additional conductivity effect was taken into account by adding a contribution −i(σ/(ε₀ω^s)) to eqn (1), where σ is the dc conductivity constant, ε₀ is the dielectric permittivity of vacuum, and the coefficient s characterizes the conduction mechanism. As shown in Fig. 3, the HN function including a conductivity process was used to fit the isothermal dielectric spectra in this study. Furthermore, the HN relaxation time τ_HN is related to the mean molecular relaxation time τ_max, corresponding to the maximum of the dielectric loss given by the equation:

Fig. 4 shows the temperature dependencies of segmental relaxation time for PS and its nanocomposites. The introduction of Ag nanoparticles has no significant effect on the segmental dynamics of PS even at high loadings (10%). The results are in agreement with that of PS/Ag nanocomposites and invariant segmental dynamics in PS thin films with decreasing thickness.^27,33 This is also consistent with the findings in many other PNCs reported previously.^{2,12–15,25,26,34} Nevertheless, a suppression of segmental dynamics of PNCs has been reported in the literature due to the reduced mobility that results from strong attractive interactions between the polymer and nanoparticles.^5,35,36 Even in the absence of specific polymer–filler interactions, natural rubber/silica nanocomposites show restricted segmental mobility compared to bulk polymers.³⁷ However, many other studies reveal that the segmental dynamics become systematically faster in the presence of nanoparticles due to variation in radii and characteristics of the nanoparticle, and interactions between polymer and nanoparticle.^6,7,38 In contrast with our result presented here, Cabral et al.⁴ and Schönhals et al.²⁹ demonstrated that segmental motion speeded up in PS/C₆₀ and PS/phenethyl–POSS nanocomposites. In agreement with these two studies, the accelerated segmental dynamics of PMMA/Ag nanocomposites were obtained due to the small size of Ag nanoparticles (5.6 nm).⁸ Moreover, Green et al.⁹ investigated the dynamics of PS/gold nanocomposites and proved that the gold nanoparticle may exhibit plasticization or antiplasticization effects by controlling the grafting density and chain length of gold nanoparticles. Such differences between our result and the above-mentioned studies could be mainly ascribed to the absence of strong polymer–nanoparticle interactions because the Ag nanoparticle is not treated by grafting polymerization or surface treatment in this study, as indicated by the FTIR spectra results shown in Fig. S3.† From the spectra of pure PS and its nanocomposites, identical peaks can be observed and no shift or new absorption band can be observed. The other important reason can be attributed to the fact that the average particle size (52.1 nm) is much larger in our study, whereas in the previous studies the size of particles are very small (several nanometers) and can be compared with the radius of gyration of the polymer chains.^4,8,9

Fig. 4 Segmental relaxation time as a function of temperature for pure PS and PS/Ag nanocomposites. The solid curve represents VFT fit to the data.

Furthermore, the temperature dependencies of α-relaxation time can be described by the Vogel–Fulcher–Tamman (VFT) equation:

where τ₀ is the infinite relaxation time, A is related to the fragility of the system, and T₀ is the Vogel temperature. As shown in Fig. 4, the fits can be well described by the same fitting parameters for all samples. Moreover, the values of τ₀, A, and T₀ are 10^−11.5 s, 505.4 K and 336.5 K, respectively, which further indicate the invariant segmental dynamics caused by Ag nanoparticle loading. On the other hand, Fig. S4† represents the shape parameters of the α-relaxation at 393 K for PS/Ag nanocomposites with various Ag nanoparticle concentrations. It can be seen that the α_HN and β_HN values are almost unchanged with increasing Ag nanoparticle loading, indicating that the width and symmetry of the α-relaxation spectra are hardly affected, which is consistent with the temperature sweep results shown in Fig. 2. Hence, we can conclude that neither the dynamics nor the distribution width of the segmental motion in PS/Ag nanocomposites is affected by the Ag nanoparticles.

3.3. Physical aging of PS/Ag nanocomposites

In this section, we further investigate the physical aging of PS and PS/Ag nanocomposites by monitoring their enthalpy relaxation using DSC. In order to ensure that the segmental mobility is identical for all samples, the physical aging has been studied at the same aging temperature due to the fact that the segmental dynamics is not altered by Ag nanoparticle loading, as discussed above. Fig. 5 shows the typical thermograms obtained from DSC for PS and its nanocomposites after physical aging at 363 K for various aging times. From the DSC traces in Fig. 5, it is obvious that the enthalpy relaxation peaks appear and the intensity increases with increasing aging time. Moreover, longer annealing time should be needed in order to achieve the plateau value of recovered enthalpy.

Fig. 5 DSC heating curves of (a) pure PS, (b) PS/Ag-3% and (c) PS/Ag-10% nanocomposites after physical aging at 363 K for various aging times.

Furthermore, the amount of recovered enthalpy of amorphous polymers annealed at a given aging temperature T_a for various times, t_a, can be calculated by integration of the difference between the measured heat capacity of the aged and nonaged samples, expressed as:^27,39

in which C^a_P(T) and C^u_P(T) are the heat capacity of the aged and nonaged samples, respectively. T₁ and T₂ represent the reference temperatures (T₁ < T_g < T₂).

In order to quantitatively analyze the effect of Ag nanoparticle on the enthalpy relaxation of PS, a comparison of recovered enthalpy ΔH(t_a) as a function of aging time for PS and its nanocomposites is shown in Fig. 6. Surprisingly, totally different physical aging behavior can be observed for PS/Ag-3% and 10% nanocomposites. For PS/Ag-3% nanocomposites, it can be noted that the values of ΔH(t_a) are higher than that of pure PS, suggesting an acceleration of physical aging. These findings agree well with the results of PMMA/silica and PS/gold nanocomposites reported by Boucher et al.,^25–27 and polymer thin or ultrathin films.^40–42 However, the physical aging is shown to slow down in PS/Ag-10% nanocomposites. The suppression of physical aging appears to be consistent with the tendency observed in studies dealing with other types of polymer nanocomposite systems.^19–24 From Fig. 6, we can also observe that the equilibrium values of ΔH(t_a) can be achieved in a shorter time for PS/Ag-3% nanocomposites compared with pure PS, whereas the situation is just the opposite for PS/Ag-10% nanocomposites, further verifying the acceleration and suppression of physical aging mentioned above.

Fig. 6 Aging time dependencies of recovered enthalpy ΔH(t_a) for pure PS and PS/Ag nanocomposites at the aging temperature of 363 K.

It is generally accepted that a model based on diffusion of free volume holes toward the polymer–nanoparticle interfaces can be used to quantitatively describe the physical aging and calorimetric T_g in PNCs samples.^25–27 Hence, the calorimetric T_g, defined as the mid-point of glass transition from DSC curve, is investigated, and is shown in Fig. 7. For PS/Ag-3% samples, the value of T_g is 376.5 K, which is lower than that of pure PS (377.6 K). The results agree well with the decrease tendency in T_g observed in other studies dealing with PNC systems,^{6–8,25–27} and thin or ultrathin polymer films.^40–42 However, the increased calorimetric T_g can be detected in PS/Ag-10% nanocomposites. Such discrepancies in previous works are mainly due to different preparation methods (such as residual solvent, annealing time, or thermal history) or various experimental conditions (such as measurement apparatus, atmosphere, rate, or ambient humidity).^43,44 These factors can be ignored in this study because all experimental conditions are uniform. The difference between PS/Ag-3% and 10% can be interpreted as follows: for PS/Ag nanocomposites with low Ag nanoparticle loading, the plasticization of the PS matrix plays a dominant role due to constraints imposed on packing of the chains by the Ag nanoparticles, and the absence of strong polymer–nanoparticle interactions because Ag nanoparticles are not treated by surface modification in this study, which is consistent with the findings reported by Pandis et al.⁸ However, for PS/Ag nanocomposites with high concentration of Ag nanoparticle, the antiplasticization effects can be observed due to an increase of the fragility of glass formation and the increased chain barrier between nanoparticles. Therefore, the calorimetric T_g increases, and physical aging is shown to slow down with further increasing Ag nanoparticle loading.

Fig. 7 DSC curves of pure PS and its nanocomposites at the cooling rate of 10 K min⁻¹. The solid lines show the position of T_g.

On the other hand, the variation trend of T_g is seemingly anomalous compared with the invariant segmental dynamics discussed above. It should be noted that the T_g of amorphous polymers probed by DSC corresponds to its transition from equilibrium liquid-like to non-equilibrium glassy state,^27,45 whereas BDS tests measure the segmental dynamics at equilibrium. Boucher et al.^25–27,33 reported the T_g depression and invariant segmental dynamics in PS thin films, PMMA/silica and PS/gold nanocomposites. In contrast, we observe the decreased and increased T_g, and also unchanged segmental dynamics in this study. Based on these observations, we can conclude that the variation trend of calorimetric T_g and physical aging in PS/Ag nanocomposites further verify the fact that physical aging is driven by the diffusion toward the polymer–nanoparticle interfaces.^25–27

3.4. Rheological behavior of PS/Ag nanocomposites

As discussed above, the invariant segmental dynamics by Ag nanoparticle loading can be observed. Furthermore, the dynamic rheological measurement has been carried out in order to investigate the dynamics on a larger scale (chain dynamics) of PS with increasing Ag nanoparticle loading. Fig. 8 shows the linear viscoelastic properties of pure PS and PS/Ag nanocomposites during an isothermal frequency sweep at 473 K. It can be observed that the dynamic storage modulus (G′), loss modulus (G′′) and complex viscosity (η*) increase with increasing the Ag nanoparticle loading, indicating that common filler reinforcement takes place. The same tendency appears at other temperatures. Moreover, the influences of Ag nanoparticles on the chain dynamics of PS can be thoroughly described using time–temperature superposition (TTS) curves of rheological measurements. Fig. 9 shows the master curves for G′ and G′′ of pure PS and PS/Ag nanocomposites. One can see that the TTS principle constructed by using the horizontal shift factor α_T holds well for both neat PS and its nanocomposites. However, a suppression of the terminal behavior at low frequencies is not observed, indicating the absence of the formation of a large-scale nanoparticle network, as shown in Fig. 1. These observations appear to be in disagreement with studies finding a second plateau in the terminal region of PNCs with high nanoparticle loading due to the formation of particle network.^2,46,47 Furthermore, the temperature dependencies of the horizontal shift factor α_T is analyzed here in detail, as shown in Fig. 10. It can be found that the value of α_T increases with increasing the weight percentage of Ag nanoparticles, implying that the overall chain dynamics speed up in PS/Ag nanocomposites, although Ag nanoparticles do not significantly influence the segmental dynamics and terminal rheological behaviors at low frequencies. However, Sangoro et al.² reported that neither the segmental dynamics nor the chain dynamics is altered by the oxide nanoparticles in P2VP/TiO₂ nanocomposites. These observed differences mainly result from different interactions, preparation methods or experimental conditions.^43,44 The faster chain dynamics can be attributed to the fact that the packing of the PS chains is constrained by Ag nanoparticles due to the absence of strong polymer–nanoparticle interactions.⁸

Fig. 8 Frequency dependencies of (a) dynamic storage modulus (open symbols) and loss modulus (filled symbols) and (b) complex viscosity for pure PS and PS/Ag nanocomposites at 473 K.

Fig. 9 Master curves of dynamic storage modulus (open symbols) and loss modulus (filled symbols) for (a) pure PS, (b) PS/Ag-3% and (c) PS/Ag-10% nanocomposites at the reference temperature of 433 K. (□, ■) 433 K, (○, ●) 443 K, (△, ▲) 453 K, (▽, ▼) 463 K and (◇, ◆) 473 K.

Fig. 10 Temperature dependencies of horizontal shift factors for pure PS and PS/Ag nanocomposites at the reference temperature of 433 K.

4. Summary

The influence of Ag nanoparticles on the segmental dynamics, physical aging and rheological behaviors of PS were investigated by means of BDS, DSC and rheological measurements. The BDS measurements reveal that the segmental dynamics of PS is not altered in the presence of Ag nanoparticles. However, totally different physical aging behavior can be observed for PS/Ag-3% and 10% nanocomposites. It can be seen that the calorimetric T_g decreases and physical aging speeds up in PS/Ag-3% nanocomposites, whereas the increased calorimetric T_g and a suppression of physical aging in PS/Ag-10% nanocomposites is observed. These results can be attributed to the plasticizing and antiplasticizing effects for nanocomposites with low and high nanoparticle loading, respectively. Moreover, filler reinforcement takes place with increasing Ag nanoparticle loading, as shown by rheological measurements. We also report an acceleration of overall chain dynamics due to the constrained chain packing by the Ag nanoparticles, despite the invariant segmental dynamics and terminal rheological behavior.

Acknowledgements

This work was supported by the National Nature Science Foundation of China (no. 51173165) and the Fundamental Research Funds for the Central Universities (no. 2013QNA4048).

References

1. C. G. Robertson and C. M. Roland, Rubber Chem. Technol., 2008, 81, 506–522.
2. A. P. Holt, J. R. Sangoro, Y. Y. Wang, A. L. Agapov and A. P. Sokolov, Macromolecules, 2013, 46, 4168–4173.
3. J. M. Kropka, V. G. Sakai and P. F. Green, Nano Lett., 2008, 8, 1061–1065.
4. H. C. Wong, A. Sanz, J. F. Douglas and J. T. Cabral, J. Mol. Liq., 2010, 153, 79–87.
5. Y. F. Ding, S. Pawlus, A. P. Sokolov, J. F. Douglas, A. Karim and C. L. Soles, Macromolecules, 2009, 42, 3201–3206.
6. A. Schönhals, H. Goering, F. R. Costa, U. Wagenknecht and G. Heinrich, Macromolecules, 2009, 42, 4165–4174.
7. P. J. Purohit, J. E. Huacuja-Sanchez, D. Y. Wang, F. Emmerling, A. Thunemann, G. Heinrich and A. Schönhals, Macromolecules, 2011, 44, 4342–4354.
8. C. Pandis, E. Logakis, A. Kyritsis, P. Pissis, V. V. Vodnik, E. Dzunuzovic, J. M. Nedeljkovic, V. Djokovic, J. C. R. Hernandez and J. L. G. Ribelles, Eur. Polym. J., 2011, 47, 1514–1525.
9. H. Oh and P. F. Green, Nat. Mater., 2009, 8, 139–143.
10. G. Tsagaropoulos and A. Eisenburg, Macromolecules, 1995, 28, 396–398.
11. L. Chen, K. Zheng, X. Y. Tian, K. Hu, R. X. Wang, C. Liu, Y. Li and P. Cui, Macromolecules, 2010, 43, 1076–1082.
12. R. B. Bogoslovov, C. M. Roland, A. R. Ellis, A. M. Randall and C. G. Robertson, Macromolecules, 2008, 41, 1289–1296.
13. M. Hernandez, J. Carretero-Gonzalez, R. Verdejo, T. A. Ezquerra and M. A. Lopez-Manchado, Macromolecules, 2010, 43, 643–651.
14. J. Otegui, G. A. Schwartz, S. Cerveny, J. Colmenero, J. Loichen and S. Westermann, Macromolecules, 2013, 46, 2407–2416.
15. J. Mijovic, H. K. Lee, J. Kenny and J. Mays, Macromolecules, 2006, 39, 2172–2182.
16. I. M. Hodge, Science, 1995, 267, 1945–1947.
17. J. M. Hutchinson, Prog. Polym. Sci., 1995, 20, 703–760.
18. C. H. Ahn, W. H. Jo and M. S. Lee, Polym. J., 2000, 32, 471–475.
19. R. D. Priestley, C. J. Ellison, L. J. Broadbelt and J. M. Torkelson, Science, 2005, 309, 456–459.
20. P. Rittigstein, R. D. Priestley, L. J. Broadbelt and J. M. Torkelson, Nat. Mater., 2007, 6, 278–282.
21. R. D. Priestley, P. Rittigstein, L. J. Broadbelt, K. Fukao and J. M. Torkelson, J. Phys.: Condens. Matter, 2007, 19, 205120–205132.
22. H. B. Lu and S. Nutt, Macromol. Chem. Phys., 2003, 204, 1832–1841.
23. S. Amanuel, A. N. Gaudette and S. S. Sternstein, J. Polym. Sci., Part B: Polym. Phys., 2008, 46, 2733–2740.
24. A. L. Flory, T. Ramanathan and L. C. Brinson, Macromolecules, 2010, 43, 4247–4252.
25. V. M. Boucher, D. Cangialosi, A. Alegria and J. Colmenero, Macromolecules, 2010, 43, 7594–7603.
26. V. M. Boucher, D. Cangialosi, A. Alegria, J. Colmenero, J. Gonzalez-Irun and L. M. Liz-Marzan, Soft Matter, 2010, 6, 3306–3317.
27. V. M. Boucher, D. Cangialosi, A. Alegria, J. Colmenero, I. Pastoriza-Santos and L. M. Liz-Marzan, Soft Matter, 2011, 7, 3607–3620.
28. Y. Q. Tan, X. Z. Yin, M. T. Chen, Y. H. Song and Q. Zheng, J. Rheol., 2011, 55, 965–979.
29. N. Hao, M. Boehning and A. Schoenhals, Macromolecules, 2007, 40, 9672–9679.
30. C. Svanberg, Macromolecules, 2007, 40, 312–315.
31. V. Lupascu, S. J. Picken and M. Wubbenhorst, Macromolecules, 2006, 39, 5152–5158.
32. S. Havriliak and S. Negami, Polymer, 1967, 8, 161–210.
33. V. M. Boucher, D. Cangialosi, H. J. Yin, A. Schonhals, A. Alegria and J. Colmenero, Soft Matter, 2012, 8, 5119–5122.
34. M. Hernandez, M. D. Bernal, R. Verdejo, T. A. Ezquerra and M. A. Lopez-Manchado, Compos. Sci. Technol., 2012, 73, 40–46.
35. M. Hernandez, T. A. Ezquerra, R. Verdejo and M. A. Lopez-Manchado, Macromolecules, 2012, 45, 1070–1075.
36. L. T. Vo, S. H. Anastasiadis and E. P. Giannelis, Macromolecules, 2011, 44, 6162–6171.
37. D. Fragiadakis, L. Bokobza and P. Pissis, Polymer, 2011, 52, 3175–3182.
38. N. Hao, M. Bohning, H. Goering and A. Schonhals, Macromolecules, 2007, 40, 2955–2964.
39. I. Hodge, J. Non-Cryst. Solids, 1994, 169, 211–266.
40. K. D. Dorkenoo and P. H. Pfromm, Macromolecules, 2000, 33, 3747–3751.
41. Y. P. Koh, G. B. McKenna and S. L. Simon, J. Polym. Sci., Part B: Polym. Phys., 2006, 44, 3518–3527.
42. Y. P. Koh and S. L. Simon, J. Polym. Sci., Part B: Polym. Phys., 2008, 46, 2741–2753.
43. M. Erber, M. Tress, E. U. Mapesa, A. Serghei, K. J. Eichhorn, B. Voit and F. Kremer, Macromolecules, 2010, 43, 7729–7733.
44. H. K. Nguyen, M. Labardi, S. Capaccioli, M. Lucchesi, P. Rolla and D. Prevosto, Macromolecules, 2012, 45, 2138–2144.
45. J. A. Forrest and K. Dalnoki-Veress, Adv. Colloid Interface Sci., 2001, 94, 167–196.
46. Q. Zheng, B. B. Yang, G. Wu and L. W. Li, Chem. Res. Chin. Univ., 1999, 20, 1483–1490.
47. G. Filippone and M. S. de Luna, Macromolecules, 2012, 45, 8853–8860.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra00517a

---