ABSTRACT
Nanorods of PCN-222, a large-pore, zirconium-based porphyrinic metal-organic framework (MOF), have been prepared through coordination modulation-controlled crystal growth through competing monodentate ligands known as modulators-for incorporation into reverse osmosis thin-film nanocomposite (TFN) membranes. Postsynthetic modification of the MOF node through binding of myristic acid (MA) altered channel dimensions and pore size distribution. The extent of MOF modification was characterized through Brunauer-Emmett-Teller gas sorption and 1H NMR following digestion of the particles. TFN membranes containing PCN-222 nanoparticles modified with varying levels of MA were fabricated via dispersion in the aqueous phase during interfacial polymerization, and the resulting flux and rejection performance of each membrane were evaluated. Increased water flux was observed with increasing MA content in the PCN-222 nanorods. Up to 95% increase in water flux was observed for a TFN containing 0.01 wt % loading of PCN-222 nanorods with a 10:1 MA to linker ratio, while maintaining high salt rejection. The flux change was attributed to tunable water transport through the nanorod pore structure and also through rapid water transport pathways at the nanorod-polymer interface.
ABSTRACT
Thin-film nanocomposite membranes (TFNs) are a recent class of materials that use nanoparticles to provide improvements over traditional thin-film composite (TFC) reverse osmosis membranes by addressing various design challenges, e.g., low flux for brackish water sources, biofouling, etc. In this study, TFNs were produced using as-received cellulose nanocrystals (CNCs) and 2,2,6,6-Tetramethylpiperidine-1-oxyl (TEMPO)-oxidized cellulose nanocrystals (TOCNs) as nanoparticle additives. Cellulose nanocrystals are broadly interesting due to their high aspect ratios, low cost, sustainability, and potential for surface modification. Two methods of membrane fabrication were used in order to study the effects of nanoparticle dispersion on membrane flux and salt rejection: a vacuum filtration method and a monomer dispersion method. In both cases, various quantities of CNCs and TOCNs were incorporated into a polyamide TFC membrane via in-situ interfacial polymerization. The flux and rejection performance of the resulting membranes was evaluated, and the membranes were characterized via attenuated total reflectance Fourier transform infrared spectroscopy (ATR-FTIR), transmission electron microscopy (TEM), and atomic force microscopy (AFM). The vacuum filtration method resulted in inconsistent TFN formation with poor nanocrystal dispersion in the polymer. In contrast, the dispersion method resulted in more consistent TFN formation with improvements in both water flux and salt rejection observed. The best improvement was obtained via the monomer dispersion method at 0.5 wt% TOCN loading resulting in a 260% increase in water flux and an increase in salt rejection to 98.98 ± 0.41% compared to 97.53 ± 0.31% for the plain polyamide membrane. The increased flux is attributed to the formation of nanochannels at the interface between the high aspect ratio nanocrystals and the polyamide matrix. These nanochannels serve as rapid transport pathways through the membrane, and can be used to tune selectivity via control of particle/polymer interactions.
