Synthesis and O2/N2 Performance Evaluation of Hollow Fiber Carbon Membrane of 2,6-Dimethyl-1,4-Phenylene Oxide
Carbon membrane is one of the most promising separation materials in gas separation technology due to is molecular-sieving properties. This paper reports on the fabrication and maximizing the performance of hollow fiber carbon membrane in O2/N2 separation. Polymer of 2,6-dimethyl-1,4-phenylene oxide was used as the carbon membrane precursor. The carbon membrane was synthesized with different setting of pyrolysis which were pyrolysis temperature, heating rate, and thermal soak time. The corresponding performance of O2 permeability and O2/N2 ideal selectivity were recorded and analysed. To seek the optimum point in which both the permeability and selectivity is highly desirable, Robeson's 2008 upper bound was utilized to select the best performance out of the pyrolysis settings. Classic approach of one-factor-at-time was used to optimize the performance and understand the phenomenon of the pyrolysis temperature effects. The optimum values after optimizing the pyrolysis temperature, heating rate and thermal soak time were 222 Barrer for the O2 permeability and 40 for the O2/N2 ideal selectivity. The optimum carbon membrane was pyrolyzed at 600 °C with heating rate of 4 °C/min and thermal soak time of 0 hr. Characterization through scanning electron microscope (SEM) depicted that the carbon membrane morphology appeared dense, symmetrical with homogenous structure. The thickness was measured to be approximately 14 µm. The gas diffusion through the carbon membrane was dominated by molecular sieving mechanism. This study indicated that the carbon membrane performance was highly influenced, even, by a small change of pyrolysis temperature. There was an optimum point to get the best performance between the low and high pyrolysis temperature. Too high or low pyrolysis temperature caused the performance to be poor both in permeability and selectivity aspects. Excessively high heating rate increased the O2 and N2 permeabilities. Implementing even a short time of thermal soaking period greatly decreased the O2 permeability and O2/N2 ideal selectivity. According to the O2/N2 mixed gas test, the O2 permeability was measured to almost the same with the O2 permeability of single gas test. Unfortunately, the O2/N2 selectivity of the mixed gas test was found to be lesser than the O2/N2 ideal selectivity.