Effect of Amine-Functionalized MIL-53 Metal Organic Frameworks on the Performance of Poly(4-methyl-1-pentyne) Membrane in CO2/CH4 Separation Gas Mixture

Document Type : Research Paper

Authors

1 Faculty of Chemical Engineering, Babol University of Technology, P.O. Box: 484, Babol, Iran

2 Faculty of Chemical Engineering, Tarbiat Modares University, P.O. Box: 14115-143, Tehran, Iran

Abstract

The effect of NH2-MIL 53 metal organic framework (MOF) on gas transport properties of poly(4-methyl-1-pentyne) (PMP) was investigated. Various characterization methods such as FTIR, DSC, SEM and gas adsorption test as well as a series of CO2/CH4 gas separation tests (i.e., pure and mixed gas test) were conducted in order to determine the effect of ligand functionalization (–NH2) on the properties of the prepared mixed matrix membranes and their gas transport characteristics. The results of DSC showed that glass transition temperature (Tg) increased by increasing NH2-MIL 53 loading. The SEM images also demonstrated that the NH2-MIL 53 particles were dispersed well in the PMP matrix with no noticeable agglomeration. The gas adsorption test of NH2-MIL 53 particles revealed there was a selective adsorption behavior with respect to CO2. It was also found that, incorporation of NH2-MIL 53 into the PMP resulted in an increase in gas permeability (especially towards CO2) and a higher CO2/CH4 selectivity. Adding 30 wt% NH2-MIL 53 into the polymer matrix increased CO2 permeability and CO2/CH4 selectivity of the mixed gas from 83.35 to 210.21 barrer and 7.61 to 19.88, respectively. Rising the temperature from 30 to 60°C led to the permeability increment of both CO2 and CH4 in the mixed gas test, while the CO2/CH4 selectivity decreased. Moreover, the results showed that amino groups required no regeneration and their performance did not decline during 120 h of permeation test. A comparison between the permeation data and those calculated from permeation models revealed that the Bruggeman model could fit the CO2 permeability data better than the Maxwell and Lewis models.

Keywords