Glass Transitions of Poly(methyl methacrylate) Confined in Nanopores

Article pubs.acs.org/JPCB

Glass Transitions of Poly(methyl methacrylate) Confined in Nanopores: Conversion of Three- and Two-Layer Models Linling Li,† Jiao Chen,† Weijia Deng,† Chen Zhang,† Ye Sha,† Zhen Cheng,† Gi Xue,*,† and Dongshan Zhou*,†,‡ †

Key Laboratory of High Performance Polymer Materials and Technology of Ministry of Education, Department of Polymer Science and Engineering, School of Chemistry and Chemical Engineering, State Key Laboratory of Coordination Chemistry, Nanjing National Laboratory of Microstructures, Nanjing University, Nanjing 210093, P. R. China ‡ Xinjiang Laboratory of Phase Transitions and Microstructures in Condensed Matters, College of Physical Science and Technology, Yili Normal University, Yining 835000, P. R. China S Supporting Information *

ABSTRACT: The glass transitions of poly(methyl methacrylate) (PMMA) oligomer confined in alumina nanopores with diameters much larger than the polymer chain dimension were investigated. Compared with the case of 80 nm nanopores, PMMA oligomer confined in 300 nm nanopores shows three glass transition temperatures (from from low to high, denoted as Tg,lo, Tg,inter, and Tg,hi). Such phenomenon can be interpreted by a three-layer model: there exists an interphase between the adsorbed layer and core volume called the interlayer, which has an intermediate Tg. The behavior of multi-Tg parameters is ascribed to the propagation of the interfacial interaction during vitrifaction process. Besides, because of the nonequilibrium effect in the adsorbed layer, the cooling rate plays an important role in the glass transitions: the fast cooling rate generates a single Tg; the intermediate cooling rate induces three Tg values, while the ultraslow cooling rate results in two Tg values. With decreasing the cooling rate, the thickness of interlayer would continually decrease, while those of the adsorbed layer and core volume gradually increase; meanwhile, the Tg,lo gradually increases, Tg,inter almost stays constant, and the Tg,hi value keeps decreasing. In such a process, the dynamic exchanges between the interlayer and adsorbed layer, core volume should be dominant.

■

INTRODUCTION Confinement of liquids in geometries at the nanoscale usually gives rise to some unusual structural and dynamic properties. In past decades, because of the potential importance for both fundamental and practical implications, considerable efforts have been paid to investigate the confinement effects on glass transitions of polymers. These investigations aim both to achieve a deeper understanding of the finite-size effect in polymer systems and to explore how interface (the substrate or free surface) influences the dynamics of confined polymer chains.1−13 It is assumed that glass forming materials would show deviations from their bulk properties when confined to dimensions similar to the intrinsic length scale associated with the glass transition.1,2 However, the confinement effect has been observed for confining sizes in the range of 5−1000 nm, which in most cases are larger than the single chain dimension, let alone the cooperativity length scale (usually several nanometers).14−18 In these cases, the interfacial effect should be considered, which may mask or compete with the finite-size effect. The free surface has been proven to accelerate the segmental dynamics,4,6,10 while the substrate can either enhance or reduce the segmental dynamics which depends on the © 2015 American Chemical Society

strength of interfacial interaction between polymer and substrate.11,12,19−21 The interfacial interaction is longrange,22,23 which may induce multilayers with a gradient of mobility along the confinement dimension.24−27 With the increase of surface-to-volume ratio, such interfacial effect is expected to increase. In addition to the finite-size effect and interfacial interaction, the interfacial adsorption should also play an important role. When polymer chains are placed in contact with a nonrepulsive interface, the irreversibly adsorbed layer would be formed as long as the monomer−surface interaction is on the order of kBT. In such cases, the nonequilibrium effect would be dominant, which leads to a nonequilibrium layer whose structure and dynamics depend on the adsorption kinetics and layer aging.3,28,29 At the state of the art, the glass transitions of polymers confined in different confinement geometries have been investigated, including polymer thin films,1−11 polymer confined in cylindrical nanopores,15−17,30−32 polymer nanoReceived: November 10, 2014 Revised: March 19, 2015 Published: March 25, 2015 5047

DOI: 10.1021/jp511248q J. Phys. Chem. B 2015, 119, 5047−5054

Article

The Journal of Physical Chemistry B particles,33−35 and nanocomposites,21,36−38 etc. Among them, the studies of polymer thin films take up a large proportion. However, it is worth noting that the pioneering works of the confinement effect are the calorimetric studies of Jackson and McKenna which involve the organic glass forming liquids confined to the nanopores of Vycor glasses.39 Recently, several groups have paid attention to the structural and dynamic properties of polymer chains confined in cylindrical alumina nanopores.15−18,30−32 Shin and Russell observed an unexpected enhancement in flow and a reduction of the intermolecular entanglements for large polystyrene molecules inside the nanoscopic cylindrical channels of anodic alumina oxide (AAO).31 Calorimetric,17 dielectric,18,30 NMR,15,16 and neutron scattering32,40 experiments have all revealed altered dynamics for polymer chains confined in AAO nanopores. It is reported that the distribution of relaxation times is broadened even within nanopores with size 50 times the unperturbed chain dimension, which can be ascribed to the interfacial effect.18 Fourkas et al. demonstrated that liquids restricted to volumes on a molecular distance scale would show strongly perturbed orientational and translational dynamics due to the confinement.41−43 On account of the chain connectivity, such perturbations are more complicated for polymer chains. Close to the confining substrate, a very thin irreversibly adsorbed layer would be formed even for the weak polymer−substrate interaction system. In the adsorbed layer, the dynamics of polymer chains are restricted and complex chain conformations including trains, loops, and tails are developed.44,45 Using neutron spin echo technique, Krutyeva et al. first observed an interphase between the bulklike chains in the core volume and anchored polymer chains at the surface, and the full chain relaxation in the interphase is impeded through the interaction with the anchored chains.32 The studies of Pye and Roth on glass transitions of the freestanding polymer films indicated the possibility of two reduced Tg values upon nanoconfinement. They proposed that two separate mechanisms simultaneously acted on the film to propagate enhanced mobility from free surface into the material.46,47 Recently, we also reported a two-Tg behavior of poly(methyl methacrylate) (PMMA) oligomer confined in 80 nm AAO nanopores, and a two-layer model induced by the propagation of interfacial interaction into the center part during the slow vitrifaction process was proposed.17 Actually, such a two-Tg scenario has been commonly observed for organic glass forming liquids confined in nanopores,48−53 although there is a debate on whether the gradient or two-Tg scenario should be adopted in these cases. For the latter scenario, there needs to be a break in cooperativity at the core−shell boundary, which appears counterintuitive. An explanation of such a break is that molecules are adhered to the surface in a preferred orientation.41,42,48,54 Koga et al. studied the chain conformations of polymer molecules accommodated at the polymer− solid interface, and two different nanoarchitectures were detected in the adsorbed layer.44 Another important concept is the nonequilibrium effect: polymer chains near the interface can be trapped in metastable states, in which the chain conformations associate with local and not global minima of the free energy. Recent works of Napolitano et al. have shown a striking correlation between the deviation from bulk behavior and the irreversible chain adsorption. Their results indicate that glass transitions of polymers under confinement could be tuned by controlling the local free volume at the interface through thermal annealing.3,9,29,55,56 At the core−shell boundary,

polymer chains with different relaxation rates meet, and the molecular dynamic exchange may happen.41,57 The dynamic exchange effect indicates that individual processes would show the apparently increased relaxation rates, and the slower process gains intensity from the faster one.58 Kremer et al. have demonstrated that such an effect is suited to explain the changes in dielectric spectra of a microconfined glass forming liquid in porous glasses.58,59 In this work, we mainly investigate glass transitions of PMMA oligomer confined in 300 nm alumina nanopores that are much larger than the single polymer chain dimension. Interestingly, three Tg values are observed for the moderate cooling rates, which can be interpreted by a three-layer model. Compared with the case of PMMA confined in 80 nm nanopores previously reported, the differences may be ascribed to the weaker interfacial effect. As the cooling rate decreases, the glass transitions gradually recover to the two-Tg scenario. The variations in Tg value and thickness of each layer during the nonisothermal annealing process are explained by the dynamic exchange effect.

■

EXPERIMENTAL SECTION Materials. The monodisperse PMMA sample with numberaverage molecular weight of Mn = 6 kg/mol (denoted as PMMA6K, its Rg is 2.4 nm, and the polydispersity index is 1.1) was purchased from Polymer Source Inc. (Dorval, Canada). The stereoregular composition is 6% isotactic, 38% atactic, and 56% syndiotactic, which is confirmed by the 1H NMR characterization. The calorimetric glass transition temperature of PMMA6K is 110 °C. The anodic aluminum oxide (AAO) templates with the average pore diameter of 80 nm and 100 μm of length were purchased from PUYUAN NANO Co. (Hefei, China) prepared via a two-step anodization process, and those with the average pore diameter of 300 nm and 60 μm of length were purchased from Whatman. All the membranes were first rinsed thoroughly with chloroform and methanol to remove the possible impurities on surfaces and then annealed at 150 °C for several hours in vacuum before use. Sample Preparation. First, the polymer films with thickness about 100 μm were prepared by solution-casting the 10 wt % PMMA6K/toluene solution onto clean cover glasses. After the films were dried under ambient conditions for several days, they were dried in vacuum for 24 h at 150 °C to remove the possible residual solvent. The dried polymer film was placed onto the AAO template and sandwiched between two glass slides. Then, the assembly was annealed at 160 °C for 12 h under vacuum. During the annealing process, the PMMA melt was drawn into the nanopores by capillary forces. The result of thermogravimetry (as shown in Figure S1 of the Supporting Information) indicates that the thermal degradation should be ignored during annealing. At last, the assembly was slowly cooled back to room temperature, and the excess PMMA on the surface of AAO template was carefully removed by a sharp razor blade. The PMMA nanorods were obtained by immersing the prepared PMMA-filled AAO sample in a sodium hydroxide aqueous solution (1 mol/L) for 12 h, rinsing thoroughly with deionized water and then drying under vacuum at room temperature. Characterization. Scanning electron microscopic (SEM) micrographs were recorded using the HITACHI S-4800 scanning electron microscope with an acceleration voltage of 20 kV. DSC measurements were performed on a PerkinElmer DSC-Pyris1 system under a nitrogen atmosphere (20 mL/ 5048

DOI: 10.1021/jp511248q J. Phys. Chem. B 2015, 119, 5047−5054

Article

The Journal of Physical Chemistry B

Table 1. Tg Values of PMMA6K Bulk and PMMA6K Confined in 80 and 300 nm AAO Nanoporesa

min). The temperature was calibrated with indium and zinc standards before measurements. The mass of PMMA-filled AAO sample was about 20 mg and the heating rate was chosen to be 10 K/min to reduce the influence of thermal lag. The amount of polymer infiltrated into AAO template was determined by the thermogravimertic analysis (TGA), and the measurement was carried out on a PerkinElmer TGA-Pyris system. The samples were heated from room temperature to 700 °C at 10 K/min under dry nitrogen, and the results are shown in Figure S1.

sample

Tg,bulk (°C)

dAAO (nm)

Tg,lo (°C)

Tg,inter (°C)

Tg,hi (°C)

PMMA6K

110

80 300

100 103

130

146 142

a

The Tg values are determined by the middle point of the glass transition region, and both the heating and cooling rates are 10 K/min.

■

values of PMMA6K confined in 80 nm nanopores, the deviation between Tg,hi and Tg,lo in 300 nm nanopores is smaller. Previously, we’ve demonstrated that such heterogeneous Tg dynamics are induced by the chain transferring from the center part to the wall surface during the slow cooling process.17 During the slow vitrification process, the strong interfacial interaction between polymer chains and pore wall can propagate into the center part, which forms an adsorbed layer with an increased Tg; meanwhile, the chains with reduced packing density in the core volume show a decreased Tg. Clearly, the stronger is the interfacial interaction, the higher is the Tg in the adsorbed layer. The increase of diameter of AAO nanopores would decrease the surface-to-volume ratio and also the interfacial effect. So it is not diffcult to understand the smaller deviation between Tg,hi and Tg,lo for PMMA oligomer confined in 300 nm nanopores. But the apperance of three Tg values is totally unexpected and the formation of the interlayer between the adsorbed layer and core volume is unclear. The finite-size effect usually induces an increase in segmental dynamics as decreasing the confining size, which is contrary to the effect of interfacial interaction. The collective dynamics of PMMA oligomer confined in smaller nanopores should be mediated by the counterbalance between the finite-size effect and interfacial effect.60,61 Hence, the glass transitions of PMMA confined in AAO nanopores smaller than 80 nm are not included in this manuscript. To investigate the nature of such three Tg parameters, the sub-Tg annealing experiments were carried out, and the enthalpy overshoots could be taken as criteria.47 The thermal procedure of aging tests are shown in Figure S2. After removal of the thermal history, the PMMA6K-filled 300 nm AAO samples were first cooled from 180 °C at 10 K/min, then each annealed at Tg − 5 K for 5 h, respectively. After the aging, a subsequent heating at 10 K/min was taken to study the changes in DSC curves. At last, the standard measurements (heating and cooling at 10 K/min, which were identical to that of sample without aging) were carried out for comparison. The results of aging tests are shown in Figure 2. We confirm that the observed behavior of the three Tg parameters is reproducible upon multiple cooling/heating cycles. After annealing at Tg,lo − 5 K for 5 h, a broad endothermic enthalpy relaxation peak appears in the Tg,lo region, which clearly demonstrates that Tg,lo is a real Tg transition. As the aging temperature is far below Tg,inter and Tg,hi, their physical aging rates are rather low, and their shapes are almost unchanged. After annealing at Tg,inter − 5 K for 5 h, due to the close adjacency of Tg,inter and Tg,hi regions, an ambiguous enthalpy relaxation peak appears in the Tg,inter region. However, the observed enthalpy overshoot can also prove that Tg,inter is a real Tg transition. Besides, we can see that the Tg,lo value increases and the possible change in the Tg,hi region is hidden because of the enthalpy overshoot of Tg,inter. After annealing at Tg,hi − 5 K for 5 h, no obvious enthalpy relaxation peak appears, which can be ascribed to the sluggish physical aging rate in the adsorbed layer.62 But the heat capacity

RESULTS AND DISCUSSION In our previous research, two distinct Tg values were detected for PMMA oligomer confined in AAO nanopores with diameter of 80 nm, among which one was higher than the bulk value and the other was lower, and the deviation in Tg was as large as 45 K (as shown in Figure 1).17 A two-layer model was proposed to

Figure 1. Upper: SEM micrographs of PMMA nanorods prepared by infiltrating AAO templates with the PMMA melt: (a) 80 nm and (b) 300 nm. Lower: DSC thermograms of bulk PMMA6K (black), PMMA6K confined in 80 nm (red) and 300 nm (blue) AAO templates. Both the heating and cooling rates equal 10 K/min. On the basis of the TGA results, the curves of PMMA6K confined in 80 and 300 nm nanopores have been normalized. The dashed lines indicate the Tg values.

interpret the phenomenon: a strongly constrained interfacial layer with an increased Tg and a core volume with a decreased Tg. However, the effect of AAO nanopore size on glass transitions of polymer confined in nanopores is absent. Figure 1 presents the SEM micrographs of PMMA6K nanorods with diameters of 80 and 300 nm prepared by melt infiltration and the DSC thermograms of PMMA6K confined in 80 and 300 nm AAO nanopores. Obviously, the glass transitions of PMMA6K confined in 300 nm nanopores are quite different from those of PMMA6K in 80 nm nanopores. First, PMMA6K in 300 nm nanopores exhibits three glass transitions: one is lower than the bulk value, and the other two are higher. Here, we denote the highest Tg as Tg,hi, the lowest Tg as Tg,lo, and the Tg in between as Tg,inter. The Tg values of PMMA6K confined in 80 and 300 nm nanopores are shown in Table 1. Second, compared with Tg 5049

DOI: 10.1021/jp511248q J. Phys. Chem. B 2015, 119, 5047−5054

Article

The Journal of Physical Chemistry B

values with strong enthalpy relaxation peaks. As depicted in Figure 3b, with decreasing Qc, the Tg,lo value gradually increases and the Tg,inter value almost stays constant while the Tg,hi value keeps decreasing. Interestingly, the Tg,lo and Tg,hi values show the linear relationships with the logarithmic Qc. Besides, the heat capacity changes in Tg,lo and Tg,hi regions gradually increase with decreasing Qc. Meanwhile that in the Tg,inter region continually decreases and finally disappears. The clear enthalpy overshoots of PMMA samples suffering ultraslow cooling rate can also demonstrate that Tg,lo and Tg,hi are real Tg transitions. Here we claim that the interphase behavior is totally reversible, which only depends on the thermal history. To investigate the reversibility of the interphase behavior, we designed a thermal procedure as shown in Figure S4. The DSC measurement of PMMA sample cooled at 0.5 K/min was capped by two standard calorimetric measurements (both the heating and cooling rates are 10 K/min). For the cooling rate of 0.5 K/min, the interphase behavior disappears while the heating curves of two standard calorimetric measurements are totally identical, which means that the interphase behavior recovers after removing the thermal history of cooling at 0.5 K/min. Because of the monomer connectivity, the polymer chains have very high conformational entropies, which tremendously increases the possibility of perturbing molecular conformations upon confinement. Here we assume that the variation of Tg in each layer should be attributed to the slight fluctuation of chain conformation or local packing density compared with the bulk state. In principle, such variation in chain conformation should be induced by the counterbalance between the conformational entropy loss of chains and the energy gain of attached segments to the surface in the total free energy.29 According to the simulation results, the density profile of polymer chains confined between two flat hard walls is heterogeneous which depends on the polymer−substrate interaction strength, film thickness, distance from the substrate, and the temperature.63−65 Near the polymer−substrate interface, due to the presence of hard walls, the polymer beads exhibit the strong ordering and sharp density variation. As the center is approached, the ordering decays and the density profile gradually becomes smooth with the bulk characterisics. At high temperatures, the decay length scale is usually smaller than the radius of gyration (Rg) of polymer chains. So it is conceivable that most of the polymer melts confined in nanopores with diameters much larger than the chain

Figure 2. Subsequent DSC heating curves of PMMA6K-filled 300 nm AAO samples (a) without aging and annealed at (b) Tg,lo − 5 K, (c) Tg,inter − 5 K, and (d) Tg,hi − 5 K for 5 h. All the heating and cooling rates are 10 K/min. The arrows point out the annealing temperatures. The dash lines are the heating curves of samples without aging.

change in the Tg,hi region becomes larger, and that in the Tg,inter region becomes smaller. What should be noticed is that the shape of the Tg,lo region is almost unaltered during such annealing process. As is well-known, nonequilibrium effect is dominant when the monomer−surface sticking energy is larger than kBT.28 In our system, due to the hydrogen bonding interaction between PMMA chains and AAO wall surface, the effect of interfacial adsorption must be taken into account. As the structure and dynamics of polymer chains in the adsorbed layer depend on adsorption kinetics and layer aging, the cooling rate may play an important role in the glass transitions of PMMA6K confined in 300 nm AAO nanopores. The thermal procedure for the investigation of cooling rate effect is shown in Figure S3. After removal of the thermal history, PMMA samples were cooled at different cooling rates, and the subsequent heating at 10 K/min was to trace the variations of glass transitions. Figure 3a presents the heating traces of PMMA6K-filled 300 nm AAO samples suffering different cooling rates. The hyperqueched sample (cooled by liquid nitrogen, in which the cooling rate is estimated to be about 120 K/s) shows a single Tg around Tg,bulk, which is consistent with our previous studies. As the cooling rate (Qc) gradually decreases from 20 to 0.5 K/min, the glass transition behavior changes from three Tg values to two Tg

Figure 3. (a) Normalized DSC heating traces of PMMA6K-filled 300 nm AAO samples suffering different cooling rates, Qc. (b) Relationships between Tg values and Qc: Tg,lo (square), Tg,inter (star), and Tg,hi (circle). 5050

DOI: 10.1021/jp511248q J. Phys. Chem. B 2015, 119, 5047−5054

Article

The Journal of Physical Chemistry B

shown in Figure S5). They have the same molecular weight (Mn ≈ 6 kg/mol) but different monomer−surface sticking energies. Interestingly, for the high physisorption sticking energy system (such as PMMA6K and P2VP6K), the interphase behavior can be observed, while for the low physisorption sticking energy system (such as PS6K), the interphase behavior is absent. The higher monomer−surface sticking energy means a stronger polymer−substrate interaction. So the results indicate that the polymer−substrate interaction is an essential factor for the interphase behavior as well. The thickness of each layer (ζ) in the three-layer model (Figure 4a-B) can be estimated by assuming that the volume of polymer in a given layer is proportional to the step change of the heat capacity of that layer, and the possible chain packing density variation along the pore radius is ignored. The thickness of core volume (ζlo), adsorbed layer (ζhi), and interlayer (ζinter) can be calculated by eqs 1, 2, and 3 as shown in below, respectively.

dimension should obey the bulk behavior. As the temperature is decreased, the interfacial interaction can propagate into the center part, and the decay length scale of density profile becomes much larger than Rg.66 We can imagine that this process should involve the polymer chains’ transferring from the center part toward the wall surface, which induces the heterogeneous Tg dynamics. However, if the cooling rate is high enough, such transfer would be inhibited. The chains are frozen in a homogeneous state as illustrated in Figure 4a-A, and the

ζlo =

ζhi =

⎞1/2 ΔCp,lo d ⎛⎜ ⎟ 2 ⎜⎝ ΔCp,lo + ΔCp,inter + ΔCp,hi ⎟⎠

(1)

⎡ ⎛ ⎞1/2 ⎤ ΔCp,lo + ΔCp,inter d⎢ ⎜⎜ ⎟⎟ ⎥ 1 − ⎥ Δ + Δ + Δ 2⎢ C C C ⎝ p,lo p,inter p,hi ⎠ ⎣ ⎦

(2)

d − ζlo − ζhi (3) 2 where d is the diameter of AAO nanopores and ΔCp,lo, ΔCp,inter, and ΔCp,hi are the step changes of heat capacity at Tg,lo, Tg,inter, and Tg,hi, respectively. For the two-layer model, the above equations are still available if ΔCp,inter and ζinter are taken as 0. As calculated by eqs 1, 2, and 3, the three layers of PMMA6K glass confined in 300 nm nanopores cooled at 10 K/min are ζhi = 16 nm, ζinter = 47 nm, and ζlo = 87 nm. With decreasing cooling rate, ζinter continually decreases until 0 while ζhi and ζlo gradually increase as shown in Figure 4b. When cooled at the ultraslow rate, only two layers are formed: ζhi = 44 nm and ζlo = 106 nm. At first thought, the ζhi of PMMA glass confined in 300 nm nanopores is much larger than that of 80 nm (14 nm).17 But if normalized by the confining size (such as 2ζhi/d), the one of 80 nm would be larger than that of 300 nm (2ζhi/ d80nm = 0.35; 2ζhi/d300nm = 0.29). On the other hand, it demonstrates that the weaker interfacial effect of PMMA chains would suffer when they are confined in the nanopores with larger size. Our previous annealing experiments on PMMA oligomer confined in 80 nm nanopores reveal that Tg,hi and Tg,lo are inherently correlated; the change of one would cause the variation of the other inevitably,17 while, as illustrated in Figure 2, this kind of direct connection seems to have failed in the case of PMMA oligomer confined in 300 nm nanopores. When aging the PMMA sample at Tg,hi − 5 K (Figure 2d), the Tg,hi region changes but the Tg,lo region is almost unaltered. When aging the PMMA sample at Tg,inter − 5 K (Figure 2c), the Tg,lo value increases but Tg,hi is unchanged. These indicate that the interlayer may act as the role of intermediate buffer during the isothermal annealing process. The results in Figure 3b suggest that the interlayer should be devided into two parts: one part adjacent to the adsorbed layer and the other part close to the core volume. As the cooling rate decreases, the former part gradually blends into the adsorbed layer and the Tg,hi value ζinter =

Figure 4. (a) Schematic graphs of the Tg distribution changes for PMMA melt suffering different cooling rates: A (hyperquenched), B (the intermediate cooling rate, like 10 K/min), and C (the ultraslow cooling rate, like 0.5 K/min). (b) Relationships between the thickness of each layer ζ and cooling rate Qc: ζlo (square), ζinter (star), and ζhi (circle). The dash lines in (a) indicate the possible boundaries between each layers, and the solid lines in (b) are guides for the eye.

hyperquenched PMMA glass in AAO nanopores exhibits a single Tg, while during the slow cooling process, the densification of polymer chains near the pore wall would form an adsorbed layer with frustrated dynamics. Meanwhile, due to the mass conservation, the chains in the core volume would be loose in packing density and show enhanced dynamics. For the same cooling rate (10 K/min), the appearance of the interlayer for PMMA confined in 300 nm, not in 80 nm AAO nanopores (Figure 1) should be attributed to two reasons: the larger nanopore size and the weaker interfacial effect. The former increases the path of chain transfer, and the latter decreases the driven force for chain transfer. Both of them would prolong the characteristic time for the complete chain transfer process that generates only two layers. So for the moderate cooling rate, some polymer chains would be trapped in a nonequilibrium state between the adsorbed layer and core volume (Figure 4a-B), while for the ultraslow cooling rate, it has enough time to complete the chain transfer process, and the two-layers scenario is generated as shown in Figure 4a-C. In this study, to systematically explore the formation of such interphase behavior, the glass transitions of three different polymers (polystyrene, PS; poly(methyl methacrylate), PMMA; and poly(2-vinylpyridine), P2VP) confined in 300 nm AAO nanopores were investigated (as 5051

DOI: 10.1021/jp511248q J. Phys. Chem. B 2015, 119, 5047−5054

The Journal of Physical Chemistry B

■

decreases. Meanwhile, the latter part gradually merges into the core volume and the Tg,lo value increases. Herein we believe that the dynamic exchange effects between the two parts of interlayer and adsorbed layer, core volume should play the important roles.58,59 The dynamic exchange happens only if the relaxation rates of polymer chains reach the order of magnitude of the exchange rate during the characteristc time of measurement. As the temperature gradually decreases, first, the dynamic exchange between interlayer and adsorbed layer takes place in the higher temperature region and then the dynamic exchange between interlayer and core volume happens in the lower temperature region. The linear relationships between Tg,hi, Tg,lo and the logarithmic Qc suggest that such processes are possibly diffusion controlled, which is consistent with the chain transfer concept mentioned above.

AUTHOR INFORMATION

Corresponding Authors

*G.X.: e-mail, [email protected]. *D.Z.: e-mail, [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors appreciate the financial support of National Basic Research Program of China (973 Program, Grant 2012CB821503) and Program for Changjiang Scholars and Innovative Research Team in University (PCSIRT). This work was also supported by the NSF of China (Grants 51133002, 21274059, 21274060, 21274062, 21304003, and 21404055).

■

■

CONCLUSIONS In this manuscript, we investigated the glass transitions of PMMA oligomer confined in AAO nanopores with diameters much larger than polymer chain dimension. In these cases, the finite-size effect should be ignored and the interfacial effect between polymer chains and pore walls would dominate. When cooled at 10 K/min, PMMA oligomer confined in 80 nm nanopores shows two distinct Tg values, which can be interpreted by the two-layer model, while, because of the weaker interfacial effect, PMMA oligomer confined in 300 nm nanopores shows three Tg values. The new phenomenon is illustrated by a three-layer model. Polymer chains near AAO pore walls suffer the strong interfacial interactions and form the adsorbed layer with an increased Tg (Tg,hi), while polymer chains in the core volume with loose packing density show a decreased Tg (Tg,lo). Meanwhile, the polymer chains between these two layers are trapped in a nonequilibrium state with an intermediate Tg (Tg,inter), called the interlayer. Because of the interfacial adsorption and the nonequilibrium effect in adsorbed layer, such behaviors show the strong cooling rate dependence. Fast cooling rate generates a single Tg; the intermediate cooling rate induces three Tg values; while the ultraslow cooling rate results in two Tg values. The Tg value and thickness of each layer reveal different relationships with the cooling rate. As the cooling rate gradually decreases, the Tg,lo gradually increases, Tg,inter almost stays constant, while Tg,hi keeps decreasing. The thickness of interlayer continually decreases. Meanwhile those of the adsorbed layer and core volume gradually increase. In our opinion, the interlayer could be devided into two parts: one part adjacent to the adsorbed layer and the other part close to the core volume. The dynamic exchanges between the two parts of interlayer and the adsorbed layer, core volume should play important roles in these variations.

■

Article

REFERENCES

(1) Forrest, J. A.; Dalnoki-Veress, K. The Glass Transition in Thin Polymer Films. Adv. Colloid Interface Sci. 2001, 94, 167−196. (2) Alcoutlabi, M.; McKenna, G. B. Effects of Confinement on Material Behaviour at the Nanometre Size Scale. J. Phys.: Condens. Matter 2005, 17, R461−R524. (3) Napolitano, S.; Capponi, S.; Vanroy, B. Glassy Dynamics of Soft Matter under 1d Confinement: How Irreversible Adsorption Affects Molecular Packing, Mobility Gradients and Orientational Polarization in Thin Films. Eur. Phys. J. E 2013, 36, 61. (4) Ediger, M. D.; Forrest, J. A. Dynamics near Free Surfaces and the Glass Transition in Thin Polymer Films: A View to the Future. Macromolecules 2014, 47, 471−478. (5) Keddie, J. L.; Jones, R. A. L.; Cory, R. A. Interface and Surface Effects on the Glass-Transition Temperature in Thin Polymer-Films. Faraday Discuss. 1994, 98, 219−230. (6) Fakhraai, Z.; Forrest, J. A. Measuring the Surface Dynamics of Glassy Polymers. Science 2008, 319, 600−604. (7) Yang, Z. H.; Fujii, Y.; Lee, F. K.; Lam, C. H.; Tsui, O. K. C. Glass Transition Dynamics and Surface Layer Mobility in Unentangled Polystyrene Films. Science 2010, 328, 1676−1679. (8) Forrest, J. A.; DalnokiVeress, K.; Stevens, J. R.; Dutcher, J. R. Effect of Free Surfaces on the Glass Transition Temperature of Thin Polymer Films. Phys. Rev. Lett. 1996, 77, 2002−2005. (9) Napolitano, S.; Wubbenhorst, M. The Lifetime of the Deviations from Bulk Behaviour in Polymers Confined at the Nanoscale. Nat. Commun. 2011, 2, 260 DOI: 10.1038/ncomms1259. (10) Baumchen, O.; McGraw, J. D.; Forrest, J. A.; Dalnoki-Veress, K. Reduced Glass Transition Temperatures in Thin Polymer Films: Surface Effect or Artifact? Phys. Rev. Lett. 2012, 109, 055701. (11) Rittigstein, P.; Priestley, R. D.; Broadbelt, L. J.; Torkelson, J. M. Model Polymer Nanocomposites Provide an Understanding of Confinement Effects in Real Nanocomposites. Nat. Mater. 2007, 6, 278−282. (12) Tsui, O. K. C.; Russell, T. P.; Hawker, C. J. Effect of Interfacial Interactions on the Glass Transition of Polymer Thin Films. Macromolecules 2001, 34, 5535−5539. (13) Chen, J.; Xu, J.; Wang, X. L.; Zhou, D. S.; Sun, P. C.; Xue, G. Thickness Dependence of Glass Transitions Measured by Ac-Chip Calorimetry in Films with Controlled Interface. Macromolecules 2013, 46, 7006−7011. (14) Kausik, R.; Mattea, C.; Fatkullin, N.; Kimmich, R. Confinement Effect of Chain Dynamics in Micrometer Thick Layers of a Polymer Melt Below the Critical Molecular Weight. J. Chem. Phys. 2006, 124, 114903. (15) Ok, S.; Steinhart, M.; Serbescu, A.; Franz, C.; Chavez, F. V.; Saalwachter, K. Confinement Effects on Chain Dynamics and Local Chain Order in Entangled Polymer Melts. Macromolecules 2010, 43, 4429−4434. (16) Hofmann, M.; Herrmann, A.; Ok, S.; Franz, C.; Kruk, D.; Saalwachter, K.; Steinhart, M.; Rossler, E. A. Polymer Dynamics of

ASSOCIATED CONTENT

* Supporting Information S

Thermogravimetric analysis for PMMA bulk and PMMA-filled AAO samples; thermal procedure of sub-Tg annealing experiments and experiments for investigation of the cooling rate effect; investigation of the reversibility of interphase behavior and the glass transitions of three different polymers (PS, PMMA, and P2VP) confined in 300 nm AAO nanopores. This material is available free of charge via the Internet at http:// pubs.acs.org. 5052

DOI: 10.1021/jp511248q J. Phys. Chem. B 2015, 119, 5047−5054

Article

The Journal of Physical Chemistry B

Polymer/Nanoparticle Interface in Poly(2-vinylpyridine)/Silica Nanocomposites. Macromolecules 2014, 47, 1837−1843. (38) Zhu, L. L.; Wang, X. L.; Gu, Q.; Chen, W.; Sun, P. C.; Xue, G. Confinement-Induced Deviation of Chain Mobility and Glass Transition Temperature for Polystyrene/Au Nanoparticles. Macromolecules 2013, 46, 2292−2297. (39) Jackson, C. L.; McKenna, G. B. The Glass-Transition of Organic Liquids Confined to Small Pores. J. Non-Cryst. Solids 1991, 131, 221− 224. (40) Martin, J. Direct Observation of Confined Single Chain Dynamics by Neutron Scattering. Phys. Rev. Lett. 2010, 104, 197801. (41) Farrer, R. A.; Fourkas, J. T. Orientational Dynamics of Liquids Confined in Nanoporous Sol-Gel Glasses Studied by Optical Kerr Effect Spectroscopy. Acc. Chem. Res. 2003, 36, 605−612. (42) Coasne, B.; Fourkas, J. T. Structure and Dynamics of Benzene Confined in Silica Nanopores. J. Phys. Chem. C 2011, 115, 15471− 15479. (43) Ding, F.; Rivera, C. A.; Zhong, Q.; Manfred, K.; He, X. X.; Brindza, M. R.; Walker, R. A.; Fourkas, J. T. Structure and Dynamics of Trimethylacetonitrile at the Silica/Vapor, Silica/Liquid, and Liquid/ Vapor Interfaces. J. Phys. Chem. C 2012, 116, 7000−7009. (44) Gin, P.; Jiang, N. S.; Liang, C.; Taniguchi, T.; Akgun, B.; Satija, S. K.; Endoh, M. K.; Koga, T. Revealed Architectures of Adsorbed Polymer Chains at Solid-Polymer Melt Interfaces. Phys. Rev. Lett. 2012, 109, 265501. (45) Granick, S. Perspective: Kinetic and Mechanical Properties of Adsorbed Polymer Layers. Eur. Phys. J. E 2002, 9, 421−424. (46) Pye, J. E.; Roth, C. B. Two Simultaneous Mechanisms Causing Glass Transition Temperature Reductions in High Molecular Weight Freestanding Polymer Films As Measured by Transmission Ellipsometry. Phys. Rev. Lett. 2011, 107, 235701. (47) Pye, J. E.; Roth, C. B. Above, below, and in-between the Two Glass Transitions of Ultrathin Free-Standing Polystyrene Films: Thermal Expansion Coefficient and Physical Aging. J. Polym. Sci., Part B: Polym. Phys. 2015, 53, 64−75. (48) Richert, R. Dynamics of Nanoconfined Supercooled Liquids. In Annual Review of Physical Chemistry; Leone, S. R., Cremer, P. S., Groves, J. T., Johnson, M. A., Eds.; Annual Reviews: Palo Alto, CA, 2011; Vol. 62, pp 65−84. (49) Schuller, J.; Richert, R.; Fischer, E. W. Dielectric Relaxation of Liquids at the Surface of a Porous Glass. Phys. Rev. B 1995, 52, 15232−15238. (50) Park, J. Y.; McKenna, G. B. Size and Confinement Effects on the Glass Transition Behavior of Polystyrene/O-Terphenyl Polymer Solutions. Phys. Rev. B 2000, 61, 6667−6676. (51) Koh, Y. P.; Simon, S. L. Trimerization of Monocyanate Ester in Nanopores. J. Phys. Chem. B 2010, 114, 7727−7734. (52) Merino, E. G.; Neves, P. D.; Fonseca, I. M.; Danede, F.; Idrissi, A.; Dias, C. J.; Dionisio, M.; Correia, N. T. Detection of Two Glass Transitions on Triton X-100 under Confinement. J. Phys. Chem. C 2013, 117, 21516−21528. (53) Bras, A. R.; Fonseca, I. M.; Dionisio, M.; Schonhals, A.; Affouard, F.; Correia, N. T. Influence of Nanoscale Confinement on the Molecular Mobility of Ibuprofen. J. Phys. Chem. C 2014, 118, 13857−13868. (54) Zhu, X.; Farrer, R. A.; Fourkas, J. T. Ultrafast Orientational Dynamics of Nanoconfined Benzene. J. Phys. Chem. B 2005, 109, 12724−12730. (55) Napolitano, S.; Cangialosi, D. Interfacial Free Volume and Vitrification: Reduction in Tg in Proximity of an Adsorbing Interface Explained by the Free Volume Holes Diffusion Model. Macromolecules 2013, 46, 8051−8053. (56) Napolitano, S.; Rotella, C.; Wubbenhorst, M. Can Thickness and Interfacial Interactions Univocally Determine the Behavior of Polymers Confined at the Nanoscale? ACS Macro Lett. 2012, 1, 1189− 1193. (57) Loughnane, B. J.; Farrer, R. A.; Fourkas, J. T. Evidence for the Direct Observation of Molecular Exchange of a Liquid at the Solid/ Liquid Interface. J. Phys. Chem. B 1998, 102, 5409−5412.

Polybutadiene in Nanoscopic Confinement as Revealed by Field Cycling H-1 NMR. Macromolecules 2011, 44, 4017−4021. (17) Li, L. L.; Zhou, D. S.; Huang, D. H.; Xue, G. Double Glass Transition Temperatures of Poly(methyl methacrylate) Confined in Alumina Nanotube Templates. Macromolecules 2014, 47, 297−303. (18) Alexandris, S.; Sakellariou, G.; Steinhart, M.; Floudas, G. Dynamics of Unentangled cis-1,4-Polyisoprene Confined to Nanoporous Alumina. Macromolecules 2014, 47, 3895−3900. (19) Fryer, D. S.; Peters, R. D.; Kim, E. J.; Tomaszewski, J. E.; de Pablo, J. J.; Nealey, P. F.; White, C. C.; Wu, W. L. Dependence of the Glass Transition Temperature of Polymer Films on Interfacial Energy and Thickness. Macromolecules 2001, 34, 5627−5634. (20) Xu, J.; Ding, L.; Chen, J.; Gao, S. Y.; Li, L. L.; Zhou, D. S.; Li, X.; Xue, G. Sensitive Characterization of the Influence of Substrate Interfaces on Supported Thin Films. Macromolecules 2014, 47, 6365− 6372. (21) Bansal, A.; Yang, H. C.; Li, C. Z.; Cho, K. W.; Benicewicz, B. C.; Kumar, S. K.; Schadler, L. S. Quantitative Equivalence between Polymer Nanocomposites and Thin Polymer Films. Nat. Mater. 2005, 4, 693−698. (22) Qi, D.; Fakhraai, Z.; Forrest, J. A. Substrate and Chain Size Dependence of Near Surface Dynamics of Glassy Polymers. Phys. Rev. Lett. 2008, 101, 096101. (23) Zhang, C.; Fujii, Y.; Tanaka, K. Effect of Long Range Interactions on the Glass Transition Temperature of Thin Polystyrene Films. ACS Macro Lett. 2012, 1, 1317−1320. (24) Ellison, C. J.; Torkelson, J. M. The Distribution of GlassTransition Temperatures in Nanoscopically Confined Glass Formers. Nat. Mater. 2003, 2, 695−700. (25) Li, R. X. N.; Chen, F.; Lam, C. H.; Tsui, O. K. C. Viscosity of Pmma on Silica: Epitome of Systems with Strong Polymer-Substrate Interactions. Macromolecules 2013, 46, 7889−7893. (26) Papon, A.; Montes, H.; Hanafi, M.; Lequeux, F.; Guy, L.; Saalwachter, K. Glass-Transition Temperature Gradient in Nanocomposites: Evidence from Nuclear Magnetic Resonance and Differential Scanning Calorimetry. Phys. Rev. Lett. 2012, 108, 065702. (27) Koga, T.; Jiang, N.; Gin, P.; Endoh, M. K.; Narayanan, S.; Lurio, L. B.; Sinha, S. K. Impact of an Irreversibly Adsorbed Layer on Local Viscosity of Nanoconfined Polymer Melts. Phys. Rev. Lett. 2011, 107, 225901. (28) O’Shaughnessy, B.; Vavylonis, D. Non-Equilibrium in Adsorbed Polymer Layers. J. Phys.: Condes. Matter 2005, 17, R63−R99. (29) Housmans, C.; Sferrazza, M.; Napolitano, S. Kinetics of Irreversible Chain Adsorption. Macromolecules 2014, 47, 3390−3393. (30) Blaszczyk-Lezak, I.; Hernandez, M.; Mijangos, C. One Dimensional Pmma Nanofibers from Aao Templates. Evidence of Confinement Effects by Dielectric and Raman Analysis. Macromolecules 2013, 46, 4995−5002. (31) Shin, K.; Obukhov, S.; Chen, J. T.; Huh, J.; Hwang, Y.; Mok, S.; Dobriyal, P.; Thiyagarajan, P.; Russell, T. P. Enhanced Mobility of Confined Polymers. Nat. Mater. 2007, 6, 961−965. (32) Krutyeva, M. Effect of Nanoconfinement on Polymer Dynamics: Surface Layers and Interphases. Phys. Rev. Lett. 2013, 110, 108303. (33) Martínez-Tong, D. E.; Soccio, M.; Sanz, A.; García, C.; Ezquerra, T. A.; Nogales, A. Chain Arrangement and Glass Transition Temperature Variations in Polymer Nanoparticles under 3D-Confinement. Macromolecules 2013, 46, 4698−4705. (34) Zhang, C.; Guo, Y. L.; Priestley, R. D. Glass Transition Temperature of Polymer Nanoparticles under Soft and Hard Confinement. Macromolecules 2011, 44, 4001−4006. (35) Feng, S.; Li, Z. Y.; Liu, R.; Mai, B. Y.; Wu, Q.; Liang, G. D.; Gao, H. Y.; Zhu, F. M. Glass Transition of Polystyrene Nanospheres under Different Confined Environments in Aqueous Dispersions. Soft Matter 2013, 9, 4614−4620. (36) Tsagaropoulos, G.; Eisenberg, A. Direct Observation of 2 Glass Transitions in Silica-Filled PolymersImplications for the Morphology of Random Ionomers. Macromolecules 1995, 28, 396−398. (37) Holt, A. P.; Griffin, P. J.; Bocharova, V.; Agapov, A. L.; Imel, A. E.; Dadmun, M. D.; Sangoro, J. R.; Sokolov, A. P. Dynamics at the 5053

DOI: 10.1021/jp511248q J. Phys. Chem. B 2015, 119, 5047−5054

Article

The Journal of Physical Chemistry B (58) Stannarius, R.; Kremer, F.; Arndt, M. Dynamic Exchange Effects in Broad-Band Dielectric-Spectroscopy. Phys. Rev. Lett. 1995, 75, 4698−4701. (59) Arndt, M.; Stannarius, R.; Gorbatschow, W.; Kremer, F. Dielectric Investigations of the Dynamic Glass Transition in Nanopores. Phys. Rev. E 1996, 54, 5377−5390. (60) Schonhals, A.; Goering, H.; Schick, C.; Frick, B.; Zorn, R. Glassy Dynamics of Polymers Confined to Nanoporous Glasses Revealed by Relaxational and Scattering Experiments. Eur. Phys. J. E 2003, 12, 173−178. (61) Sha, Y.; Li, L.; Wang, X.; Wan, Y.; Yu, J.; Xue, G.; Zhou, D. Growth of Polymer Nanorods with Different Core−Shell Dynamics via Capillary Force in Nanopores. Macromolecules 2014, 47, 8722− 8728. (62) Priestley, R. D.; Ellison, C. J.; Broadbelt, L. J.; Torkelson, J. M. Structural Relaxation of Polymer Glasses at Surfaces, Interfaces and in between. Science 2005, 309, 456−459. (63) Batistakis, C.; Lyulin, A. V.; Michels, M. A. J. Slowing Down versus Acceleration in the Dynamics of Confined Polymer Films. Macromolecules 2012, 45, 7282−7292. (64) Freed, K. F.; Dudowicz, J.; Stukalin, E. B.; Douglas, J. F. General Approach to Polymer Chains Confined by Interacting Boundaries. J. Chem. Phys. 2010, 133, 094901. (65) Watanabe, K.; Kawasaki, T.; Tanaka, H. Structural Origin of Enhanced Slow Dynamics near a Wall in Glass-Forming Systems. Nat. Mater. 2011, 10, 512−520. (66) Rotella, C.; Wubbenhorst, M.; Napolitano, S. Probing Interfacial Mobility Profiles via the Impact of Nanoscopic Confinement on the Strength of the Dynamic Glass Transition. Soft Matter 2011, 7, 5260− 5266.

5054

DOI: 10.1021/jp511248q J. Phys. Chem. B 2015, 119, 5047−5054