(587e) Ultrafiltration polyvinylidene fluoride (PVDF) membranes prepared by combined crystallisation & diffusion (CCD) technique

Polyvinylidene fluoride (PVDF) is one of the most commonly used materials for making ultrafiltration and microfiltration membranes. It is desirable because of its superior mechanical properties and stability in various chemical and thermal environments. It is one of the best suited membrane materials for waste water treatment applications. Unfortunately, existing commercial PVDF membranes in the ultrafiltration range suffer from low permeation fluxes (less than 150 LMH.bar^-1) [1]. These commercial PVDF membranes are prepared via traditional routes such as non-solvent induced phase separation (NIPS) or thermally induced phase separation (TIPS) or a combination of both [2-7]. To compensate for its poor permeation performance, large areas of membranes are required for a separation application, which drive the capital and operating costs upwards making it less desirable as a separation technology option.

To address this challenge, a new method of fabrication called combined crystallisation and diffusion (CCD), inspired by ice templating, was developed by our research group [1, 8]. This method involves unidirectional cooling of casted membrane using a solid media maintained at -30 °C. The objective is to freeze the solvent of the casted polymer solution into crystals and leach them out in an ice-cold water bath as it can then serve as pore templates in the final membrane. The permeation rates of these CCD membranes were found to be an order of magnitude higher than the commercial membranes having similar pore sizes.

The first part of this study involved performance comparison of PVDF flat sheet membranes prepared via the most common membrane fabrication method, NIPS and the innovative unidirectional freezing technique of CCD. The polymer dope concentration was varied from 10-20wt.% to produce PVDF membranes of varying pore sizes. The results indicated similar pore sizes for CCD and NIPS membranes prepared by specific dope solution concentrations, but significantly higher permeation rates for CCD membranes (10 times higher) than its NIPS counterpart. This was attributed to differences in the morphology of the two membrane types as seen from their SEM images presented in Fig 1. The CCD membranes also had superior mechanical and anti-fouling properties. The flux recovery of fouled membranes after cleaning was significantly higher for CCD membranes.

The second part of this study focused on devising an efficient way to influence the pore sizes of CCD membranes. So far, varying either the cooling conditions or the total polymer dope concentration would influence the pore size. At 20wt.% total polymer dope concentration and existing cooling conditions, the smallest pore size achieved for CCD membranes was about 30 nm. The purpose of this study was to decrease the pore size further, without increasing the total polymer concentration or changing the cooling conditions, as both present operational challenges. To achieve this blending modification was applied using poly(methyl methacrylate) (PMMA) as an additive.

The pore size of CCD PVDF membranes was reduced by 50% to 15nm by carefully adjusting the blend proportion of PMMA in the PVDF-DMSO dope solutions. Systemic study shows the pore size reduces when increasing PMMA concentration in the membranes; but after an initial drop compared to the pure PVDF membrane, the pure water flux doesn’t decline with PMMA concentration. The effect of different molecular weights of PMMA (120kDa and 35kDa) was also investigated, where the results showed a smaller pore size for membranes prepared using the higher molecular weight PMMA. Scanning electron microscopy (SEM), cloud-point temperature measurement and Fourier transform infrared spectroscopy (FTIR) were used to characterise the membranes and help to understand how PMMA affects membrane formation. From the evidence gathered, we attribute the effects of PMMA to the change in the number of nucleation centre, diffusional rate, as well as the gelation point based on a previously proposed CCD mechanism.

References

[1] B. Wang, J. Ji, and K. Li, "Crystal nuclei templated nanostructured membranes prepared by solvent crystallization and polymer migration," Nat Commun, vol. 7, p. 12804, Sep 19 2016.

[2] T.-T. Jin, Z.-P. Zhao, and K.-C. Chen, "Preparation of a poly(vinyl chloride) ultrafiltration membrane through the combination of thermally induced phase separation and non-solvent-induced phase separation," Journal of Applied Polymer Science, vol. 133, no. 5, pp. n/a-n/a, 2016.

[3] J. T. Jung, J. F. Kim, H. H. Wang, E. di Nicolo, E. Drioli, and Y. M. Lee, "Understanding the non-solvent induced phase separation (NIPS) effect during the fabrication of microporous PVDF membranes via thermally induced phase separation (TIPS)," Journal of Membrane Science, vol. 514, pp. 250-263, 2016.

[4] J. Lee, B. Park, J. Kim, and S. B. Park, "Effect of PVP, lithium chloride, and glycerol additives on PVDF dual-layer hollow fiber membranes fabricated using simultaneous spinning of TIPS and NIPS," Macromolecular Research, vol. 23, no. 3, pp. 291-299, 2015.

[5] J. Liu, X. Lu, J. Li, and C. Wu, "Preparation and properties of poly (vinylidene fluoride) membranes via the low temperature thermally induced phase separation method," Journal of Polymer Research, vol. 21, no. 10, 2014.

[6] T. Xiao, P. Wang, X. Yang, X. Cai, and J. Lu, "Fabrication and characterization of novel asymmetric polyvinylidene fluoride (PVDF) membranes by the nonsolvent thermally induced phase separation (NTIPS) method for membrane distillation applications," Journal of Membrane Science, vol. 489, pp. 160-174, 2015.

[7] H. Matsuyama, S. Berghmans, and D. R. Lloyd, "Formation of anisotropic membranes via thermally induced phase separation," Polymer, vol. 40, no. 9, pp. 2289-2301, 1999/04/01/ 1999.

[8] B. Wang, J. Ji, and K. Li, "Separation Membranes," United Kingdom, 2017.