Spherical MgO nanoparticles and Flake-like clay minerals modified with polyaniline (PAni) are applied in Matrimid in order to fabricate mixed matrix membranes (MMMs) having improved gas separation performance. The CO2 permeability, CO2/CH4 selectivity and CO2-induced plasticization pressure of MMMs are assessed at 4-30 bar feed pressure. The chemical structure, morphology and thermal properties of MMMs are analyzed by Fourier transform infrared equipped with attenuated total reflection (FTIR-ATR), X-ray diffraction, field emission scanning electron microscopy (FESEM) and differential scanning calorimetry/thermogravimetric analyses. Crater-like morphology is seen for MMMs by FESEM tests Physical interaction of Matrimid with MgO and PAni, confirmed by FTIR-ATR, helps better distribution of both the fillers in the polymer matrix. Mechanistical analysis of gas sorption and diffusion terms reveals that PAni/clay (PC) enhances gas permeability by controlling diffusivity paths, while MgO nanoparticles improves both gas solubility and diffusivity. The permeability tests reveal that the sample PC5M5 with 5 wt% PC and 5 wt% MgO, has the best gas separation properties. High pressure tests reveal that this sample can enhance plasticization pressure and CO2 permeability as well by 76% and 200%, respectively; resulting in 425% enhancement in MMMs capacity in natural gas sweetening.

Asymmetrically mixed matrix Matrimid-MIL-53 membranes with silicone cover layer were fabricated. For better understanding of membrane fabrication process, three main parameters of fabrication, Matrimid concentration, silicone concentration and weight percentage of metal organic framework (MIL-53) particles, were optimized by an experimental design method. Cross-section SEM images were used to study the membrane structure and polymer-particles interface. Moreover, thermal resistance of the membranes and the existence of various bonds in them were investigated by FTIR and TGA analyses. The results showed that membranes had porous structure with finger-like morphology. At low and moderate percentages of particles, there were no non-selective voids observed at polymer-particles interface.The thermal resistance of membranes increased with the increase of MIL-53 weight percentage and the destruction temperature of polymer increased from 410°C to 450°C.The permeability tests results showed that the Matrimid (20% wt) -MIL-53 (15% wt) / PMHS (10%wt) membrane exhibited the highest level of CO2/CH4 selectivity (23.6).However, in the membrane with 30 wt% particles loading, selectivity decreased due to particles agglomeration and void formation. The experimental design results showed that the concentration of silicone in covering solution had significant effect. CO2 and CH4 permeability decreased and ideal selectivity of CO2/CH4 increased with silicone concentration enhancement. Although the Matrimid concentration had a little effect on CO2/CH4 ideal selectivity, its enhancement increased the selectivity of the gases.The optimization results showed the membrane with 17.8% of Matrimd polymer, 13.2% of silicone polymer and 15.5 wt% of MIL-53 particle displayed the highest selectivity and CO2 permeability.

Gas separation membranes with enhanced performance were developed by the introduction of nanosized palladium particles. In this study, gas separation performance of Matrimid membrane incorporated with palladium-zeolitic imidazolate framework-8 (Pd@ZIF-8) particles, prepared by solution casting method, has been investigated. ZIF-8 nanoparticles were first synthesized using rapid room-temperature method and Pd@ZIF-8 nanoparticles were prepared through an assembly approach. The synthesized nanoparticles were characterized by scanning electron microscopy (SEM), transmission electron microscopy (TEM), energy-dispersive X-ray spectroscopy (EDX), X-ray diffractometry (XRD), and Fourier transform infrared equipped with attenuated total reflection (FTIR-ATR). TEM images exhibited that the incorporated Pd nanoparticles are well confined within the ZIF-8 crystals. The prepared mixed matrix membranes (MMMs) with various Pd@ZIF-8 contents (0–, 30 wt%) were characterized by SEM and XRD. All the modified membranes exhibited enhanced gas separation performance for selected gas pairs compared to a neat Matrimid membrane. The highest performance was observed for nanocomposite membranes with incorporation of 30 wt% of Pd@ZIF-8 into Matrimid which resulted in 72% increase in H2 permeability. The separation factors of H2/CO2, H2/CH4 and H2/N2 were improved from 3. 25, 126. 13, 95. 21 for Matrimid/ZIF-8 to 5. 82, 228. 2 and 169. 1 for Matrimid/(Pd@ZIF-8) membranes, respectively. Finally, the increase in feed pressure significantly improved the separation performance quality in all the membranes.

Reducing greenhouse gas emissions is a critical environmental challenge, with fossil fuel consumption and deforestation as major contributors to rising CO2 levels. Membrane-based gas separation has emerged as a promising technology due to its low energy requirements, compact design, and environmental compatibility. In this study, a novel Sulfonated PolyEther Ether Ketone (SPEEK)/polyvinyl alcohol (PVA)/zinc oxide (ZnO) nanocomposite membrane was fabricated and evaluated for CO2/CH4 separation. The incorporation of ZnO nanoparticles into the SPEEK/PVA matrix was optimized to achieve uniform dispersion, enhancing both permeability and stability. Characterization by SEM, FT-IR, XRD, AFM, and TGA confirmed the successful integration of ZnO nanoparticles and demonstrated favorable structural and thermal properties. The SPEEK (70 %)/PVA (28 %)/ZnO (2 %) membrane exhibited a CO2 permeability of 40–60 Barrer with CO2 /CH₄ selectivity of 20–25, representing an effective balance of transport and selectivity. Compared to state-of-the-art membranes such as Pebax/PEG/SiO2 and Matrimid/ZIF-8, the proposed membrane delivers comparable separation performance at significantly lower production costs (≈7.5–8.5 USD/m²), highlighting its industrial scalability and economic advantage. In addition, Artificial Neural Network (ANN) modeling using a multilayer perceptron (MLP) with the Levenberg–Marquardt algorithm achieved excellent predictive accuracy (R² = 0.9999), confirming strong agreement between experimental and predicted values. This combined experimental and modeling approach demonstrates the potential of SPEEK/PVA/ZnO membranes as cost-effective, high-performance candidates for CO2 capture and natural gas purification.