Carbon Nanotube Membranes: Synthesis, Properties, and Future Filtration Applications

Abstract

Over the course of the past decade, there has been growing interest in the development of different types of membranes composed of carbon nanotubes (CNTs), including buckypapers and composite materials, for an ever-widening range of filtration applications. This article provides an overview of how different types of CNT membranes are prepared and the results obtained from investigations into their suitability for different applications. The latter involve the removal of small particles from air samples, the filtration of aqueous solutions containing organic compounds and/or bacteria, and the separation of individual liquids present in mixtures. A growing number of reports have demonstrated that the incorporation of CNTs into composite membranes confers an improved resistance to fouling caused by biomacromolecules and bacteria. These results are discussed, along with evidence that demonstrates it is possible to further reduce fouling by taking advantage of the inherent conductivity of composite membranes containing CNTs, as well as by using different types of electrochemical stimuli.

Keywords: buckypapers; carbon nanotubes; desalination; fouling; permeability; pervaporation.

Figure 1

Currently used methods for synthesising carbon nanotubes (CNTs). Reproduced from Reference [17] with permission of the Royal Society of Chemistry.

Figure 2

Schematic illustration of chitosan helically wrapping around the outside of a CNT. The structure of chitosan is also shown. Reproduced with permission from Reference [30]. Copyright Elsevier, 2007.

Figure 3

(A) Scanning electron microscope (SEM) image of a vertically aligned array of CNTs produced using a Fe-catalysed chemical vapor deposition (CVD) process; (B) schematic illustration of the structure of an aligned CNT membrane. From Reference [37]. Reprinted with permission from the American Association for the Advancement of Science.

Figure 4

Production of a macro architecture consisting of aligned multi-walled carbon nanotubes (MWNTs), for use in filtration applications. (A) Schematic of the spray pyrolysis apparatus used for growing aligned MWNTs. The process consisted of a nozzle attached to a ferrocene/benzene solution supply used for releasing the solution into a quartz tube, mounted inside a temperature-controlled cylindrical furnace. A benzene/ferrocene solution was injected into the quartz tube, using argon as a carrier gas, and the temperature of the furnace then increased to 900 °C; (B) Photograph of the bulk nanotube architecture produced by the above method; (C) SEM image of the aligned nanotubes with radial symmetry, resulted in a hollow cylindrical structure (scale 1 mm). Reprinted by permission from Macmillan Publishers Ltd from Reference [45]. Copyright Nature Publishing Group 2004.

Figure 5

(A) Digital photograph of a MWNT buckypaper; (B) an SEM micrograph of a MWNT buckypaper (BP).

Figure 6

Inactivation and metabolic activity of E. coli cells retained on a SWNT/poly(vinylidene fluoride) (PVDF) composite filter and on a bare PVDF membrane filter: (A) Inactivation test results showing the presence of E. coli cells that are not viable; (B) metabolic activity test results indicating the presence of metabolically active E. coli cells. Adapted with permission from Reference [16]. Copyright John Wiley and Sons, 2008.

Figure 7

Effect of time on the amount of bacteria remaining in the filtrate (expressed as colony forming units (CFU)) after passage across Ag/MWNT membranes with different loadings of silver. Reproduced with permission from Reference [84]. Copyright Elsevier, 2015.

Figure 8

(a) SEM micrograph of the surface of a self-supporting MWNT BP prepared from a dispersion of MWNTs in isopropyl alcohol. Scale bar is 100 nm; (b) SEM micrograph of the surface of the BP after filtration of gold NPs. Scale bar is 100 nm. The inset is an high-resolution transmission electron microscopy (HRTEM) image showing the Au NPs; (c) UV–visible absorption spectrum of the colloidal solution of Au NPs before and after filtration through a MWNT BP. Reprinted with permission from Reference [85]. Copyright American Chemical Society, 2012.

Figure 9

Schematic illustration of pervaporation of an azeotropic mixture. (A) Feed solution containing a mixture of ethyl tert-butyl ether (ETBE), tert-butyl alcohol (TBA), and ethanol; (B) intermediate; and (C) final stages of pervaporation using an MWNT/PVA BP. Reproduced with permission from Reference [90]. Copyright Elsevier, 2014.

Figure 10

Performance of a reduced graphene oxide (rGO)/MWNT hybrid membrane for removing fulvic acid (initial feed concentration 50 ppm) from water. Reproduced from Reference [95] with permission from the Royal Society of Chemistry.

Figure 11

Performance of rGO/MWNT hybrid NF membranes during experiments involving feed solutions containing Au nanoparticles, bovine serum albumin (BSA), or phoxim. Reproduced from Reference [95] with permission from the Royal Society of Chemistry.

Figure 12

Effect of SWNT loading on wastewater flux (at 0.5 bar) and the removal of bisphenol A (BPA) and 4-nonylphenol (4-NP) by composite polyacrylonitrile (PAN)/SWNT membranes. Reproduced with permission from Reference [99]. Copyright Taylor and Francis, 2016.

Figure 13

Effect of time on the adsorption of: (A) triclosan; and (B) ibuprofen by SWNT/PVDF and MWNT/PVDF composite membranes, both in the absence and presence of Suwannee River fulvic acid (SRFA). Reprinted with permission from Reference [100]. Copyright Elsevier, 2015.

Figure 14

Effect of application of negative potentials to PVA/MWNT–COOH/PS-35 membranes on the extent of fouling caused by a solution consisting of 5 g·L⁻¹ alginic acid. Reduced levels of fouling led to smaller increases in applied pressure being required to maintain membrane operation. Reproduced with permission from Reference [122]. Copyright Elsevier, 2014.

Figure 15

Effect of an applied electrochemical potential on the performance of a MWNT/PAN/Al₂O₃ membrane exposed to humic acid: (a) effect on total organic carbon (TOC) removal efficiency; (b) normalised permeate flux. Reprinted with permission from Reference [124]. Copyright American Chemical Society, 2015.

Figure 16

Effect of different types of electrochemical stimulation on the performance of an MWNT/ceramic membrane during filtration experiments performed using solutions containing natural organic matter: (a) Effect on normalised permeate flux of water; (b) Effect on TOC removal efficiency. Reproduced with permission from Reference [125]. Copyright Elsevier, 2015.