The seminal work in the field of polymer glass transition temperature (Tg)-confinement effects came in the mid 1990’s when Keddie et al. showed via ellipsometry that a 10-nm-thick polystyrene (PS) film supported on silica exhibited a reduction of up to 29 K in Tg relative to Tg,bulk PS.1 They postulated that a “liquid-like” layer existed at the free surface which increasingly influenced the overall chain dynamics of the film when the ratio of surface area to volume is increased with geometric confinement. Many other (pseudo-)thermodynamic measurements of Tg in supported, thin PS films have shown confinement effects similar to those of Keddie et al. Free-standing PS films show considerably larger Tg-confinement effects than supported PS films.2,3 For PS with Mn< 350 kg/mol, the free-standing films exhibit the same Tg-confinement effect as supported films when Tg is plotted as a function of characteristic length, which is volume divided by free surface area.2Such a plot accounts for free-standing films having twice the number of free surfaces as supported films.
Historically, Tg-confinement studies have focused on the one-dimensional film geometry; far fewer confinement studies have been reported on three-dimensional nanospheres possibly because of the increased complexity in preparation and measurement of samples. Nanoparticles represent an alternative geometric system to investigate the effects of free surfaces and confinement on Tg. Nanosphere confinement studies found in literature vary greatly in methods of sample preparation, strength of the reported Tg-confinement effect, and conclusions reached.4–6 Sasaki et al.4 reported that PS spheres prepared via emulsion polymerization having diameters of 42 - 548 nm exhibited an “unambiguous glass transition very near the Tg of bulk PS” while both Zhang5 and Feng6 reported an apparent Tg-confinement effect as a function of particle diameter. Using surfactant-free emulsion polymerization, Zhang5 reported that particles ~150 nm in diameter exhibited a reduction in Tg of ~45 K from bulk PS Tg while Feng6 showed that ~150 nm diameter “surfactant-free” particles (PS synthesized via conventional emulsion polymerization followed by removal of surfactant via dialysis) exhibited a more modest Tg reduction of ~10 K from Tg,bulk.
We have prepared PS nanoparticles and evaluated the Tg-confinement effects. To ensure that our particles are free of surfactant and unreacted monomer, we employ a facile nanoprecipitation method. Stable nanoparticles possessing diameters ranging from 40 - 950 nm were produced by precipitating in water various concentrations of polystyrene (Mn = 91 kg/mol) dissolved in tetrahydrofuran. The effect of nanoscale confinement on Tg was determined using the temperature dependence of a fluorescence intensity ratio associated with trace levels of covalently labeled pyrene dye. We find that a strong effect of confinement on Tg is obtained for particles with diameters below 170 nm; bulk Tg was measured for particles with larger diameter. A quantitative comparison between the Tg-confinement effect seen in particles and films was made possible by defining a characteristic length equal to the ratio of volume to surface area. The size-dependent Tg values of our PS particles are in good agreement with those of supported films (of all molecular weights) and free-standing films (Mn< 350 kg/mol) when the Tg data are compared as a function of characteristic length (volume/area). This indicates that there is a common physical origin of the Tg-confinement effect in the different geometries which is associated with how the free surface perturbs Tgsome tens of nanometers in the film or particle interior.
(1) Keddie, J. L.; Jones, R. A. L.; Cory, R. A. Europhys. Lett. 1994, 27, 59–64.
(2) Mattsson, J.; Forrest, J. A.; Borjesson, L. Phys. Rev. E 2000, 62, 5187–5200.
(3) Kim, S.; Torkelson, J. M. Macromolecules 2011, 44, 4546–4553.
(4) Sasaki, T.; Shimizu, A.; Mourey, T. H.; Thurau, C. T.; Ediger, M. D. J. Chem. Phys. 2003, 119, 8730.
(5) Zhang, C.; Guo, Y.; Priestley, R. D. Macromolecules 2011, 44, 4001–4006.
(6) Feng, S.; Li, Z.; Liu, R.; Mai, B.; Wu, Q.; Liang, G.; Gao, H.; Zhu, F. Soft Matter 2013, 9, 4614.