CO2-selective polymer and mixed-matrix membranes for CO2/N2 and CO2/CH4 separations

The separation of CO2 from energy streams such as raw natural gas and bio gas is of high importance in industry because CO2 causes severe infrastructure corrosion and lowers the caloric value of the produced methane. Also in environmental context, CO2 capture has gained increasing interest since CO2 is a key greenhouse gas, strongly contributing to global warming. Over the past three decades, membrane-based gas separation has matured into an industrial standard technology for applications such as N2 production, H2 recovery during ammonia synthesis and CO2 removal from natural gas. The polymeric membranes that are the industrial standard for these applications generally only have mediocre separation performance. Membranes with better performance would therefore increase the membrane market share in existing applications (e.g. natural gas sweetening) and/or unlock new application domains (e.g. membrane-based CO2 capture). Development of new polymer membranes and MOF MMMs with improved separation performance therefore remains relevant and essential for the field. One way to improve membrane performance is to incorporate metal-organic frameworks (MOFs) in polymer matrices to form mixed-matrix membranes (MMMs). MMMs are hybrid materials that combine the advantages of polymers (e.g. good processability) with the excellent separation ability of MOFs. Even though the concept behind MMMs is relatively straightforward, its practical implementation involves a number of challenges. One of them is the lack of distinct structure-property relationships, which currently impedes the rational preparation of tailor-made MMMs for specific gas separation applications. In addition, testing of MMMs and, by extension, membranes in general under industrially relevant testing conditions is of high importance to gain insight in the true potential of the membrane. Yet, membrane developers still remain overly fixated on researching thick free-standing membrane films tested under pure gas conditions or, at best, a binary mixture of gases, whereas this is only the first stage of membrane development. In face of these challenges, the aim of this dissertation was threefold: (1) research new membrane materials with enhanced separation properties for CO2/CH4 and CO2/N2 model mixtures, respectively representing natural gas/bio gas and flue gas streams. In a first part of this dissertation the CO2 affinity of zirconium-based UiO-67 MOFs was enhanced by replacing the standard linker with BPYDC (2,2'-bipyridine-5,5'-dicarboxylic acid) linkers in varying concentrations. At optimal conditions, incorporation of these MOFs in Matrimid membranes simultaneously improved the CO2/CH4 separation factor and the CO2 permeability. However, at too high BPYDC concentrations, the MOF stability was compromised, thus limiting the practical applicability of the UiO-67-BPYDC MMMs. As a response to the limited stability of UiO-67, MOF-808 was investigated as a far more stable alternative for MMM design. The unique structure of MOF-808 allowed direct incorporation of monotopic modulator molecules in the MOF framework. Hence, selection of appropriate modulators bearing CO2-philic chemical moieties led to an in-situ MOF functionalization towards enhanced CO2 affinity (modulator-mediated functionalization, MoFu) of the MOF and strongly improved performance of the MMM. The versatility of the MoFu concept was demonstrated by using a series of fluorinated and non-fluorinated carboxylic acid modulators. Also pure polymer strategies were investigated for membrane performance improvement. 6FDA-DABA polymer membranes, bearing free carboxylic acid groups on the polymer backbone, were exposed to different annealing temperatures (100 °C, 180 °C, 250 °C, 350 °C and 400 °C) to evoke decarboxylation cross-linking. The separation performance of the 6FDA-DABA membranes was found to be governed by two counteracting mechanism: physical tightening of the polymer matrix as a result of a more efficient polymer stacking and dilation of the polymer matrix by decarboxylation crosslinking. The membrane annealed at 350 °C displayed a synergy between the dilation effect of cross-linking and the tightening effect, causing a simultaneous, strong improvement of both CO2/CH4 separation factor (+100%) and CO2 permeability (+40%). Moreover, the cross-linked 6FDA-DABA-350 showed an increased resistance to CO2-induced plasticization thanks to covalent cross-linking; (2) initiate a systematic approach for a more rational design of MMMs. Consistent structure-performance relationships for the rational design of MOF-based MMMs for gas separation are very scarce. A first step in finding such relationships could be the linking of intrinsic MOF parameters, such as the CO2 uptake and the CO2 adsorption enthalpy (Qst), to the performance indicators of the MMM. MOF-808 parameters were correlated with the CO2/N2 separation factor and CO2 permeability of the corresponding MMMs. MOF-808 CO2 uptake correlated poorly with MMM performance, which is in strong contrast to literature where CO2 uptake is one of the dominant factors used to explain MMM separation behavior. Correlation coefficients of CO2 Qst with the separation factor were significantly higher than for CO2 uptake but were still considered only moderate. However, a strong correlation was found between Qst and CO2 permeability, indicating that Qst can be considered as the most effective predictor for MMM CO2 permeability amongst the MOF parameters; (3) the design and assembly of a second generation high-throughput gas separation (HTGS-2) set-up for membrane testing under industrially relevant conditions.

Jaar van publicatie:2020

Toegankelijkheid:Open