Preparation of antibacterial and antifouling PSF/ZnO/eugenol membrane for peat water ultrafiltration

Peat water is a potential clean water source for people living in water-stressed places, especially in peatland areas. However, peat water needs proper treatment since it contains a high concentration of dissolved organic matter. Established purification technologies, such as adsorption and ozonation, were limited by the process complexity related to the regeneration of adsorbent and the generation of ozone in a reactor (Liu et al. 2013; Zulfikar et al. 2014; Sumisha et al. 2015). Ultrafiltration (UF) membranes have gained much attention as the alternative technology due to the low energy consumption, process simplicity, and being environmentally friendly (Liu et al. 2013; Sianipar et al. 2017; Rockey et al. 2018). However, fouling is still inevitable and becomes the major drawback of UF operation (Díez et al. 2017). Among different types of fouling, biofouling is the most challenging as the biofilm cannot be easily removed from the membrane surface (Orooji et al. 2017).

Various strategies for controlling biofouling formation on UF membrane have been studied. The most common strategy is the addition of additives, such as TiO₂, silicon dioxide, and ZnO (Jo et al. 2016; Aryanti et al. 2017; Chew et al. 2018; Ma et al. 2018; Tian et al. 2019). ZnO is a low-cost additive which has good antibacterial activity and can improve membrane performance due to its hydrophilic nature (Jo et al. 2016). Eugenol is one of the natural extracts showing excellent antibacterial activity (Devi et al. 2010; Harun et al. 2016). Harun et al. (2016) investigated the synergistic effect of eugenol and ZnO as antibacterial agents in polysulfone (PSF) UF membrane. Despite its antibacterial activity, the membrane has not been used in peat water treatment.

PSF is widely used for UF membrane preparation due to its good chemical resistance, temperature stability, and mechanical strength (Wenten et al. 2016; Jiang et al. 2019; Rodrigues et al. 2019). The performance of PSF membrane for peat water treatment has been demonstrated in the previous study (Aryanti et al. 2015). However, the membrane should be improved since it had no antibacterial activity. Therefore, in this research, antibacterial PSF membrane was prepared by introducing ZnO and eugenol via a blending method. The additive blending method is simple and scalable, which is expected to decrease the cost of membrane fabrication.

PSF UDEL-3500 purchased from Solvay Advanced Polymer, N,N-dimethylacetamide (DMAc) obtained from Jingsan Jingwei Chemical Co. Ltd, and PEG400 obtained from PT. Saary Indoraya (Indonesia) were used for membrane solution preparation. Zinc oxide (ZnO) and eugenol (99% purity) obtained from a local supplier were used as additives. Nutrient Agar (Himedia) was purchased from Merck. Natural peat water was obtained from Pekanbaru river, Riau province, Indonesia. Demineralized water was used in all experiments.

PSF membranes were synthesized by phase inversion and additive blending methods (Figure S1(a) in the Supplementary Materials). The detailed composition of the membrane solution is summarized in Table S1 (Supplementary Materials). The procedure for membrane preparation is explained in the Supplementary Materials. Briefly, membrane preparation consists of dissolving PSF and PEG400 in DAMc, blending additives in PSF solution, evaporating the solvent from the solution, casting the membrane solution, and immersing the cast solution in a coagulation bath.

The membrane cross-section was probed by scanning electron microscope (SEM, JEOL JSM 6510A). Membrane chemical structures were analyzed by Fourier transform infrared spectroscopy (FTIR, Bruker Tensor 27 FTIR). Water contact angle (WCA) was determined by the sessile drop technique (Aryanti et al. 2015).

Antibacterial activity was examined by the disc diffusion technique. Bacteria culture (Escherichia coli) and peat water (in a separate assay) were placed on the membrane surface (6 mm in diameter) prior to the incubation period of 24 h at 37 °C (and at room temperature for the peat water test). The formation of the inhibition zone around the membrane disc was measured (Harun et al. 2016).

As shown in Figure 1, the membranes exhibited asymmetric structure with finger-like macropores in the porous sublayer. PSF-3 to PSF-6 display longer macropores in the porous sublayer than PSF-1 and PSF-2. This may imply that ZnO and eugenol induce an increasing rate of solvent–non-solvent exchange during the immersion in the coagulation bath. This suggests an increasing diffusion rate of deionized water into the membrane solution with the increasing eugenol and ZnO concentration (Ali et al. 2011; Ma et al. 2011). The increasing diffusion rate may be due to the hydrophilic nature of the additives.

Before immersion, the cast solution was kept at room condition for partial solvent evaporation. This was intended to ensure membrane integrity when immersed in the coagulation bath since direct immersion will result in a wavy membrane and the leaching out of polymer or particles from the membrane solution (Khoiruddin & Wenten 2016). In addition, evaporation leads to the formation of a dense skin layer (Figure 1).

The SEM images showed that ZnO nanoparticles were dispersed well as there were no ZnO agglomerations. This indicates that simple additive blending is able to produce mixed-matrix membrane with a fine distribution of inorganic particles.

To demonstrate the successful introduction of PEG400, ZnO, and eugenol in PSF membrane, PSF-0, PSF-1, PSF-2, and PSF-4 membranes were characterized via FTIR analysis (Figure 2 ). All membranes showed absorption peaks at around 1,300–1,350 cm⁻¹ indicating S = O stretching vibration (Mushtaq et al. 2014). The S = O represents the sulfonic groups of PSF. Absorption peaks at around 1,560–1,650 cm⁻¹ were found which represent the functional group of C = C of the benzene ring (Zhang et al. 2008). The absorption peak at around 500 cm⁻¹ representing the stretch mode of ZnO (Silva & Zaniquelli 2002) was observed in the PSF-2 and PSF-4 membranes. The stretching vibrations of O-H were represented by the strong, broad band at around 3,000–3,700 cm⁻¹ and the weak, broad band at 2,800–3,000 cm⁻¹ (Ali et al. 2011). The increase in the intensity of the strong alcohol band was observed in PSF-2 and PSF-4 indicating more alcohol functional groups provided by eugenol. The results of FTIR analysis show the successful additive introduction in the PSF membranes.

Pure water permeability increases with the addition of ZnO and eugenol (Figure 3(a)). The noticeable improvement of the permeability significantly highlights the important role of hydrophilic additives. In addition, membrane permeability might be affected by the solvent and non-solvent exchange rate during the phase inversion process (Chung et al. 2017). The hydrophilic nature of ZnO and eugenol might attract more water which increases the solvent–non-solvent exchange rate (Rajabi et al. 2015). This results in a membrane with major finger-like structures showing higher permeability.

ZnO and eugenol decrease WCA from 67.7° ± 1.2° to a minimum of 52.8° ± 0.5° for PSF-0 and PSF-4, respectively (Figure 3(b)). The increase in WCA may be due to the reduction of interface energy of PSF membranes as ZnO and eugenol introduce polar properties (Khayyat 2013; Rajabi et al. 2015). This was indicated by an increase in the OH- groups in the membrane matrix (Figure 2). The decreased WCA is also attributed to the hydrophilic nature of the additives.

The addition of ZnO and eugenol successfully increased the membrane permeability (Figure 4(a)). The rejection was also improved with the addition of ZnO and eugenol (Figure 4(a) and 4(b)). The increasing rejection of humic substances might be due to the increase in membrane hydrophilicity. The higher hydrophilicity might result in lower humic substances–membrane surface interaction which helps to repel humic substances from the membrane surface (Aryanti et al. 2017). As the water flux is higher at higher hydrophilicity, the concentration of humic substances in the permeate stream is lower.

The addition of ZnO and eugenol also affected the anti-organic fouling property, shown by the higher FRR (Figure 4(c)). FRR is the ratio of the water flux of the used membrane to the fresh membrane. This implied that membranes with ZnO and eugenol were easier to clean than PSF-0. The improved antifouling was also indicated by the lower RFR of the modified membranes than of the unmodified one (Figure 4(c)). RFR is a ratio of peat water to pure water which means the reduction of flux due to the presence of foulant. This result shows that membrane with additive exhibits lower fouling tendency than unmodified membrane. Moreover, the introduction of additives leads to a decrease in the rate of flux decline (Figure 4(d)). Again, the increasing anti-organic property might be associated with increased hydrophilicity.

Membranes with ZnO and eugenol displayed various distances of the bacteria inhibition zone (Figure 5). In contrast, there was no inhibition zone observed on PSF-0. ZnO and eugenol had a similar effect on antibacterial properties as can be seen on PSF-1 and PSF-2 (1 mm thickness of the inhibition zone). It was reported that combining ZnO and eugenol allows enhancement of electron transfer for producing reactive oxygen species (ROS) which act as the agent of bacterial cell eradication (Chung et al. 2017). Therefore, PSF/ZnO/eugenol membranes display improved antibacterial activity.

Peat water may contain various microorganisms. In this assay, peat water was used without further treatment and purification. PSF-4 showed the best antibacterial performance, with the thickness of the inhibition ring at around 1 cm (Figure 6). Results indicate that the membrane can be potentially used in peat water filtration with an increase in anti-organic and antibacterial properties.

In this work, antibacterial and anti-organic fouling PSF-based membranes for peat water filtration have been designed and prepared by introducing ZnO and eugenol. ZnO and eugenol improved the filtration properties of the membrane, i.e. pure water permeability, peat water flux, and humic substance rejection probably due to higher membrane hydrophilicity. ZnO and eugenol decreased WCA on the membrane from 67.7° ± 1.2° to 52.8° ± 0.5°. The increase in hydrophilicity also resulted in improved anti-organic fouling. The antibacterial test performed on E. coli and the peat water sample resulted in the appearance of an inhibition ring around 3 mm and 1 cm thickness, respectively. This appears to be due to the antibacterial properties of ZnO and eugenol.

The authors gratefully acknowledge the financial support from P3MI (Program Penelitian, Pengabdian kepada Masyarakat dan Inovasi) Institut Teknologi Bandung and KARG (KURITA-AIT research grant).

The Supplementary Data for this paper are available online at http://dx.doi.org/10.2166/ws.2019.103.