Affiliations 

  • 1 The University of Queensland, FIM²Lab-Functional Interfacial Materials and Membranes Laboratory, School of Chemical Engineering, Brisbane 4072, Australia. [email protected]
  • 2 The University of Queensland, FIM²Lab-Functional Interfacial Materials and Membranes Laboratory, School of Chemical Engineering, Brisbane 4072, Australia. [email protected]
  • 3 The University of Queensland, FIM²Lab-Functional Interfacial Materials and Membranes Laboratory, School of Chemical Engineering, Brisbane 4072, Australia. [email protected]
  • 4 The University of Queensland, FIM²Lab-Functional Interfacial Materials and Membranes Laboratory, School of Chemical Engineering, Brisbane 4072, Australia. [email protected]
  • 5 The University of Queensland, FIM²Lab-Functional Interfacial Materials and Membranes Laboratory, School of Chemical Engineering, Brisbane 4072, Australia. [email protected]
  • 6 The University of Queensland, FIM²Lab-Functional Interfacial Materials and Membranes Laboratory, School of Chemical Engineering, Brisbane 4072, Australia. [email protected]
Materials (Basel), 2016 Nov 18;9(11).
PMID: 28774057 DOI: 10.3390/ma9110938

Abstract

This work investigates the structural formation and analyses of titania membranes (TM) prepared using different vacuum exposure times for molecular weight (MW) cut-off performance and oil/water separation. Titania membranes were synthesized via a sol-gel method and coated on macroporous alumina tubes followed by exposure to a vacuum between 30 and 1200 s and then calcined at 400 °C. X-ray diffraction and nitrogen adsorption analyses showed that the crystallite size and particle size of titania increased as a function of vacuum time. All the TM membranes were mesoporous with an average pore diameter of ~3.6 nm with an anatase crystal morphology. Water, glucose, sucrose, and polyvinylpyrrolidone with 40 and 360 kDa (PVP-40 kDa and PVP-360 kDa) were used as feed solutions for MW cut-off and hexadecane solution for oil filtration investigation. The TM membranes were not able to separate glucose and sucrose, thus indicating the membrane pore sizes are larger than the kinetic diameter of sucrose of 0.9 nm, irrespective of vacuum exposure time. They also showed only moderate rejection (20%) of the smaller PVP-40 kDa, however, all the membranes were able to obtain an excellent rejection of near 100% for the larger PVP-360 kDa molecule. Furthermore, the TM membranes were tested for the separation of oil emulsions with a high concentration of oil (3000 ppm), reaching high oil rejections of more than 90% of oil. In general, the water fluxes increased with the vacuum exposure time indicating a pore structural tailoring effect. It is therefore proposed that a mechanism of pore size tailoring was formed by an interconnected network of Ti-O-Ti nanoparticles with inter-particle voids, which increased as TiO₂ nanoparticle size increased as a function of vacuum exposure time, and thus reduced the water transport resistance through the TM membranes.

* Title and MeSH Headings from MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine.