Polymer-Based membranes play a critical role in several industrially important gas separation processes, e.g., carbon dioxide removal from natural gas. However, an intrinsic trade-off between membrane flux (characterized by its permeability) and selectivity to one gas over the other has limited their effectiveness in practical environments. While some incremental success has been obtained by empirically developing new polymer chemistries, the best hopes for transformative improvements may require novel architectures employing predictive structure/property relationships.
In this work, we develop a novel hybrid membrane construct comprised of inorganic nanoparticles grafted with polymer chains to form grafted nanoparticles. We find that the grafting architecture almost exclusively results in enhanced gas transport properties, in contrast with those expected from conventional predictions. These enhancements, found to be a result of elevated diffusion constants, are broadly tunable with the grafted chain length and leads to order of magnitude increases in gas permeability. We conjecture that the grafted polymer chains serve to impart added free volume to the composite material, which manifests itself as enhanced gas diffusion relative to the pure polymer. Indeed, multiple experimental and simulation probes verify this picture, and indicate that the free volume increases are a result of the grafted chains adopting anisotropic conformations to fill space.
Building off of this finding, we systematically study the effects of the nanoparticle core size and chain grafting density, and find that both the chain length where the maximum permeability occurs, as well as the extent of the enhancement, varies depending on the relative sizes of the chains and the nanoparticle. A thorough structural analysis of the grafted nanoparticles in dilute solution as well as bulk samples indicate that the relation between the measured polymer brush height and the chain length undergoes a transition at intermediate chain lengths, similar to the observed gas permeability enhancements. Using a simple scaling approach, we show that this transition is related to the crossover from a concentrated polymer brush with higher order scaling to a semi-dilute brush where the chains are more ideal. We hypothesize that this impenetrable concentrated brush phase is the source of the added free volume, and that this effect is diminished when the grafted chains are longer than the transition point and the penetrable, semi-dilute polymer brush begins to dominate gas diffusion. When cast in the framework of free volume theories, this prediction accurately captures the trends in gas diffusion; the result is a unique structure/property relation that can be used to design optimal membrane materials.
We expand on these constructs to probe other grafted nanoparticle-based architectures incorporating free polymer chains and advanced chemistries to further manipulate the gas transport properties of these mixed-matrix materials. The addition of free chains with judiciously chosen molecular weights and loadings gives a nearly independent means to tune membrane selectivity, which when combined with the intrinsic permeability increases in the matrix-free grafted nanoparticles results in superior materials that can exceed the current performance Upper Bound. We relate this result to the spacial distribution of the free chains throughout the grafted polymer corona, and how this affects the distribution of the free volume in the material as it selectively cuts off larger gas molecules. We further leverage this universal grafting platform by grafting polymer chains with novel chemistries to design membranes with record-setting selectivities while also increasing permeability by nearly two orders of magnitude. We conclude that grafted nanoparticle constructs allow for precise and predictive control of gas transport properties through a new structure/property relation, and serve as a novel material design platform with the potential to function as high performance gas separation technologies.
Identifer | oai:union.ndltd.org:columbia.edu/oai:academiccommons.columbia.edu:10.7916/d8-kq78-g545 |
Date | January 2019 |
Creators | Bilchak, Connor R. |
Source Sets | Columbia University |
Language | English |
Detected Language | English |
Type | Theses |
Page generated in 0.0017 seconds