Search

Polymer membranes are critical to many sustainability applications that require the size-based separation of gas mixtures. Despite their ubiquity, there is a continuing need to selectively affect the transport of different mixture components while enhancing mechanical strength and hindering aging. Polymer-grafted nanoparticles (GNPs) have recently been explored in the context of gas separations. Membranes made from pure GNPs have higher gas permeability and lower selectivity relative to the neat polymer because they have increased mean free volume. Going beyond this ability to manipulate the mean free volume by grafting chains to a nanoparticle, the conceptual advance of the present work is our finding that GNPs are spatially heterogeneous transport media, with this free volume distribution being easily manipulated by the addition of free polymer. In particular, adding a small amount of appropriately chosen free polymer can increase the membrane gas selectivity by up to two orders of magnitude while only moderately reducing small gas permeability. Added short free chains, which are homogeneously distributed in the polymer layer of the GNP, reduce the permeability of all gases but yield no dramatic increases in selectivity. In contrast, free chains with length comparable to the grafts, which populate the interstitial pockets between GNPs, preferentially hinder the transport of the larger gas and thus result in large selectivity increases. This work thus establishes that we can favorably manipulate the selective gas transport properties of GNP membranes through the entropic effects associated with the addition of free chains.

The performance of polymer-based membranes for gas separation is currently limited by the Robeson limit, stating that it is impossible to have high gas permeability and high gas selectivity at the same time. We describe the production of membranes based on the ability of graphene oxide (GO) and poly(ethyleneimine) (PEI) multilayers to overcome such a limit. The PEI chains act as molecular spacers in between the GO sheets, yielding a highly reproducible, periodic multilayered structure with a constant spacing of 3.7 nm, giving a record combination of gas permeability and selectivity. The membranes feature a remarkable gas selectivity (up to 500 for He/CO 2 ), allowing to overcome the Robeson limit. The permeability of these membranes to different gases depends exponentially on the diameter of the gas molecule, with a sieving mechanism never obtained in pure GO membranes, in which a size cutoff and a complex dependence on the chemical nature of the permeant is typically observed. The tunable permeability, the high selectivity, and the possibility to produce coatings on a wide range of polymers represent a new approach to produce gas separation membranes for large-scale applications.

Atactic polystyrene, as reported in a recent contribution by our group, displays a marked change in glass transition when exposed to toluene vapor due to plasticization associated with vapor sorption within the polymer. The dependence of the glass transition temperature of the polymer-penetrant mixture on the pressure of toluene vapor is characterized by the so-called "retrograde vitrification" phenomenon, in that, at a constant pressure, a rubber to glass transition occurs by increasing the temperature. In this contribution, we have used a theoretical approach, based on the nonrandom lattice fluid thermodynamic model for the polymer-toluene mixture, to predict the state of this system, i.e., rubbery or glassy, as a function of fluid pressure and system temperature. The experimentally detectable glass transition is assumed to be a kinetically affected evidence of an underlying II order thermodynamic transition of the polymer mixture. On the basis of this hypothesis, the Gibbs-Di Marzio criterion, stating that equilibrium configurational entropy is zeroed at the glass transition, has been applied to locate the transition. The working set of equations consists of the expression of configurational entropy obtained from the adopted lattice fluid model equated to zero, coupled with the equation expressing the phase equilibrium between the polymer phase and the pure toluene vapor phase in contact and with the equations of state for the two phases. Theoretical predictions are in good qualitative and quantitative agreement with the experimental results previously obtained gravimetrically performing "dynamic" sorption experiments, which represent a neat example of the occurrence of so-called "type IV" glass transition temperature vs pressure behavior. The peculiar retrograde vitrification phenomenon and the glass transition temperature vs pressure envelope determined experimentally are well described by the proposed theoretical approach.

Exposing a glassy polymer to a fluid phase (in gaseous or liquid state) containing a low molecular weight compound results in the sorption of the latter within the polymer, inducing, among other effects, the plasticization of the material which also promotes a change in the glass transition temperature. The amount of sorbed penetrant is often related in a complex fashion to the temperature and pressure of the fluid, thus determining that the locus of glass transition, when represented in pressure-temperature coordinates, may display as well rather complex patterns. This is an issue of particular importance in several applications of glassy polymers. In particular, we investigated the behavior of polystyrene in contact with toluene vapor by performing several modes of dynamic sorption experiments, in which the rate of change of the temperature of the system and/or of the pressure of the vapor phase are controlled with high accuracy, with the aim of creating a map of rubbery and glassy states of the polymer as a function of temperature and pressure of the toluene vapor. Isothermal tests were performed by changing the pressure at a controlled rate, isobaric tests were performed by changing the temperature at a controlled rate, and isoactivity tests were performed by concurrently changing, in a proper way, both temperature and pressure. A relevant feature resulting from these experiments is the presence of a discontinuity in the slope of the mass of toluene sorbed within polystyrene reported as a function of temperature and/or pressure. This discontinuity has been interpreted as the indication of the occurrence of a glass transition. The elaboration of the experimental results allowed identification of the pressure/temperature conditions at which rubbery or glassy states of the polymer mixture are established. Quite interestingly, the system displays the so- called "retrograde vitrification" phenomenon, which consists of the occurrence of a rubbery-to-glassy state transition as the temperature increases at a fixed pressure. The whole set of results has been successfully interpreted on the basis of thermodynamics of II order transitions accounting for the fact that experimental evidence of such transitions is significantly affected by the kinetics of polymer relaxation.