Silver-polysulfone composite membranes of three types diff ering in skin porosity and macrovoid structure were prepared to explore the coupled eff ects of host matrix porosity and nanoparticle growth conditions on the properties of resulting nanocomposites [ 79 ]. Silver nanoparticles were either synthesized ex situ and then added to the casting solution as organosol or produced in the casting solution via in situ reduction of ionic silver by the polymer solvent.
Polysulfone was chosen as the matrix material due to its wide use in the preparation of UF membranes. Nanoscale silver was chosen for two reasons.
First, there are established protocols for the preparation and characterization of silver sols in various dispersion media, including organic solvents that can reduce silver and are used in membrane preparation. Second, the presence and availability of silver can be detected by the extent of its biocidal eff ect, which provides a convenient framework for the evaluation of the accessibility of silver in polymer matrices and serves as additional motivation in view of potential use of silver-fi lled nanocomposites as biofouling-resistant surfaces.
All membranes were prepared by the wet phase inversion process wherein cast fi lms of polysulfone (PSf) dissolved in a mixture of N,N-dimethylacetamide (DMAC) and dimethylformamide (DMF) were immersed in a non-solvent with respect to polysulfone. DMAC served as the primary polymer solvent whereas DMF acted as the reducing agent for silver. Membranes of three distinct types (Types I, II, and III) were obtained by varying the composition of the casting mixture and the composition and temperature of the non-solvent in the immersion bath ( Table 5.1 ). Either polyethylene glycol (PEG) or 2-propanol was added to the casting mixture as a pore forming agent (porogen). Either water or propanol was used as the non-solvent medium.
Two silver incorporation approaches were adapted. In the fi rst approach, Ag nanoparticles were synthesized ex situ and were added to the casting solution as Ag-DMF organosol. The organosol was prepared by adding AgNO 3 to DMF (reducing agent) and heating the solution under intense stirring conditions [ 80 ].
Table 5.1 Components of Membrane Casting Mixture (% Mass)
Component (%)
Membrane
PSf I PSf II PSf III PSf/Ag I PSf/Ag II PSf/Ag III
PSf 19.97 10.08 10.12 19.20 9.88 9.92
DMAC 45.04 70.78 60.87 43.32 69.38 59.66
DMF 18.97 19.14 19.22 18.24 18.76 18.84
PEG 16.02 — — 15.40 — —
2-propanol — — 9.79 — — 9.59
AgNO 3 — — — 3.84 1.98 1.98
DMAC: N,N-Dimethylacetamide; DMF: Dimethylformamide; PSf: Polysulfone.
Savage_Ch05.indd 64
Savage_Ch05.indd 64 11/11/2008 2:15:59 PM11/11/2008 2:15:59 PM
The second approach involved an in situ reduction of ionic Ag + by DMF in the membrane casting mixture. In this case, AgNO 3 was fi rst dissolved in DMF at room temperature to minimize Ag reduction.
The AgNO 3 /DMF solution was added to the casting mixture, which was then heated under intense stirring to initiate the reduction of silver ions to Ag 0 , with the concomitant formation of silver nanoparticles. Reduction was allowed to proceed for 1 minute prior to casting the membrane.
The impact of silver incorporation on membrane properties was pronounced only for nanocomposites with lower porosity of the host matrix ( Fig. 5.1 ).
For such membranes, the incorporation of nanoscale silver caused macrovoid broadening, an increase in surface pore size and density, and a signifi cant decrease in the hydraulic resistance accompanied by only a minor decrease in rejection. The distribution of silver within membranes depended on the extent of silver reduction prior to membrane casting. In all cases, nanoparticles seemed to be preferentially concentrated along the internal pore surface ( Fig. 5.1 , insets a, b, and c), with ex situ types showing apparent higher coverage densities. The fi ndings indicate that the mere presence of silver nanoparticles in the casting mixture, regardless of their size, is enough to induce morphological changes in the morphology of the polymeric matrix for suffi ciently dense membranes. The observed morphologies can be interpreted as resulting from three processes that simultaneously take place in the casting mixture: (i) nanoparticle formation via the reduction of ionic silver by DMF, and (ii) nanoparticle re-dissolution into the DMAC, a component of the casting mixture that cannot reduce silver but can dissolve it, and (iii) demixing and polymer precipitation.
Biofouling tests were conducted to determine the biocidal eff ectiveness of the prepared nanocomposites. A thick biofi lm could be observed on surfaces of all PSf membranes whereas on the nanocomposite surfaces the bacterial growth was inhibited ( Fig. 5.2 ). There was an evident decrease in bacterial growth on nanocomposites of both PSf/Ag in and PSf/Ag ex of all porosities with respect to their silver-free counterparts. It was also observed that for the very porous membranes, simulated by Type II membranes operated with the more porous side facing the feed, the biocidal effi cacy was insensitive to the method of silver incorporation. Because the skin, now facing the permeate side, provided an absolute barrier to the bacteria, the number of inoculated bacteria retained by the membrane was the same as that in all other biofouling assays performed in this work, enabling a straightforward comparison with the Type II membranes inoculated in the conventional way. In summary, it was demonstrated that the reduction of ionic silver by polymer solvents during the phase inversion process can be employed to synthesize bioactive silver–polymer nanocomposites of a range of porosities. The morphology of the resulting nanocomposite is a result of the dynamic interplay between the polymer coagulation and nanoparticle formation processes. The built-in antibacterial capacity due to the gradual release of ionic silver by the prepared nanocomposites can be eff ective in reducing intrapore biofouling in porous membranes of a wide porosity range.
Savage_Ch05.indd 65
Savage_Ch05.indd 65 11/11/2008 2:16:00 PM11/11/2008 2:16:00 PM
(a) PSf Type I (b) PSf/Agin Type I (c) PSf/Agex Type I
(d) PSf Type II Macrovoid Matrix
(A)
(B)
(C)
(D)
(A) (C)
(B) (D)
1 μm
1 μm 2 μm
400 nm (e) PSf/Agin Type II (f) PSf/Agex Type II
(g) PSf Type III (h) PSf/Agin Type III (i) PSf/Agex Type II
Figure 5.1 Distribution of silver nanoparticles in the silver–polysulfone nanocomposites prepared using diff erent methods of silver incorporation. Red dots correspond to silver nanoparticles. Insets (a), (b), and (c) show cross-section transmission electron microscope images. Arrows point to nanoparticles embedded within the polysulfone matrix. Inset (d) shows a scanning electron microscope image of the top cross-section layer, with arrows pointing to observed larger silver crystals.
Savage_Ch05.indd 66
Savage_Ch05.indd 66 11/11/2008 2:16:00 PM11/11/2008 2:16:00 PM
Figure 5.2 Scanning electron microscope micrographs illustrating the inhibition biofi lm growth on the surface of PSf/Agin nanocomposite (right) compared to the biofi lm developed on the surface of the silver-free polysulfone membrane (left) under same conditions.
2 μm
20 μm 20 μm
Such nanocomposites could in principle be used as materials for macroporous membrane spacers to inhibit the biofi lm growth on downstream membrane surfaces.