60th Annual Report on Research 2015

Scheme 1. The procedure of PDMS/CNTs PNMs synthesis   1. Characterization and analysis of the membranes 1.1 Solvent effects on PSU supported neat PDMS membranes             Figure 1 shows the effect of two solvents (ethyl acetate and n-heptane) on PSU supported neat PDMS membranes. a&d, b&e and c&f images present the cross-section, top surface (PDMS) and bottom surface (PSU) of the membranes, respectively. Comparing the cross-sections of the membranes made from two different solvents (a&d), the most obvious distinction is that it is single layer from ethyl acetate while double layer with higher pore density from n-heptane. The main reason is that PDMS has higher solubility parameter in ethyl acetate (9.0 cal^1/2cm^-3/2) than that in n-heptane (7.4 cal^1/2cm^-3/2),¹ which also results in pores on the top surface of the membrane from ethyl acetate solvent, Figure b. Moreover, the bottom surface of the membrane from ethyl acetate is much rougher than that from n-heptane, Figure 1, c&f.    

Figure 1. SEM microstructures of PDMS/PSU membranes made from ethyl acetate (a-c) and n-heptane (d-f). a&d, b&e and c&f images depict the cross-section, top surface (PDMS) and bottom surface (PSU) of the membranes, respectively   1.2 Thermogravimetric analysis (TGA)   

Figure 2. TGA of PSU substrate, neat PDMS and PNMs, with CNTs loading of 0.5, 1.0, 1.5, and 2.0 wt%.

Table 1. Onset temperature T_1onset and T_2onset of the PSU substrate, neat PDMS and the PNMs Thermal decomposition curves of PSU substrate, neat PDMS and PNMs under air atmosphere are shown in Figure 2. The detailed thermal decomposition temperatures are shown in Table 1. T_1onset and T_2onset are defined as the temperature at 10 and 20 wt% loss of the tested specimen, respectively. For the neat PDMS and PNMs, there is a sharp weight loss stage in the temperature range from 350 to 700 °C, which is caused by the chain breakdown of the PDMS structure. The T_1onset values (Table 1) indicate that introduction of CNTs caused no obvious difference; however, the T_2onset value decreases with increasing CNTs loading due to the high thermal conductivity of CNTs which facilitate the heat transfer to the inner part of the polymer. 1.3 Dielectric property

Figure 3. Real permittivity (ε´) of PSU substrate, neat PDMS and PNMs with CNTs loading of 0.5, 1.0, 1.5, and 2.0 wt%, respectively             For the dielectric property study, effects of the CNTs loading on the real permittivity (ε´) of PNMs are shown in Figure 3. The PSU substrate, neat PDMS and PNMs with 0.5 wt% CNTs loading are observed to have a constant ε´ in all the frequency range. For PNMs with 1.0, 1.5 and 2.0 wt% CNTs loading, the ε´ value decreases with increasing oscillation frequency, which indicated a dielectric relaxation behavior.²

2. Absorbance measurement at 226 nm by UV-vis spectrometry

Table 2. Summary of concentration (ppm) of thiophene in thiophene/hexadecane mixture and corresponding absorbance (A) at 226 nm

Figure 4. Calibration curve with R² = 0.99999 and equation C/A=0.27932 C+1.97519             In this project, UV-vis spectrometry is employed to determine the concentration of thiophene in thiophene/hexadecane mixture. A calibration curve was constructed by measuring the absorbance at 226 nm for different concentrations (62.5, 125, 250, 500, 1000 and 2000 ppm), followed by plotting C/A against C. The high R² value indicates that this method is reliable for quantification across the range of 2000 ppm thiophene.

3. Nanoparticle surface treatment Nanoparticles were functionalized according to the following scheme to facilitate uniform dispersion and strong particle-polymer interactions, thus improve sulfur compound transportation and fuel processing yield. The functionalization was verified by FTIR in Figure 5, in which the strong absorption peak at 1074 cm^-1 is corresponding to Si-O-C bond in the curing agent, while for modified NPs, three representative peaks were observed, namely Si-O-Si, Si-C and N-H bands at 1044, 1349 and 1595 cm^-1, respectively.

Scheme 2. Silanization of the hydroxyl group functionalized nanofiller.

Figure 5. FT-IR spectra of the as-received, modified Fe₃O₄ NPs, and the coupling agent.

4. Desulfurization equipment

Figure 6. Pervaporation system Pervaporation system has been successfully assembled to achieve desulfurization process as shown in Figure 6. The flow rate and solution temperature can be well controlled by flow meter and heater, respectively.

In summary, PNMs preparation has been accomplished and the following work will be pursued to test the desulfurization performance and the results will be used to guide further optimization.

a)      Thermal properties will be investigated by TGA/DSC.

b)      Mechanical properties will be assessed by tensile test machine following the ASTM standard.

c)      Morphology of nanofiller and membranes will be studied by SEM, transmission electron microscopy (TEM) and atomic force microscopy (AFM).

d)      Particle-polymer interactions will be further investigated by X-ray fluorescence elemental mapping, X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS).

e)      Thiophene/hexadecane will be used to study desulfurization capacity of PDMS and the PNMs. In addition, we have done some related work in membrane separation: a) lab synthesized bio-inspired CO₂-philic network membrane for sustainable gas separation; and b) novel polyamide (PA) thin-film-composite (TFC) nanofiltration (NF) membranes (TFC NF) which have achieved dual resistance to fouling and chlorine.