Recycled wastewater has been considered a reliable source for water supply in areas facing severe water shortages. However, secondary effluent from wastewater treatment plants (WWTPs) contains colloidal particles, pathogenic microorganisms and organic pollutants, as well as natural organic matter (NOM) that is not degraded in the biological treatment process. The presence of NOM can result in colour and odour, and can contain precursors for disinfection by-product formation increasing human health risks [1
In order to increase the use of recycled water by communities, secondary effluent needs to undergo numerous treatment processes until a safe and reusable quality can be achieved. Over the last few decades, a great deal of interest has focused on advanced oxidation processes (AOPs) for the degradation of organic compounds present in wastewater. Among the AOPs, photocatalytic oxidation mediated by a semiconductor catalyst is one of the possible advanced oxidation processes. In particular, photocatalysis with TiO2
has been the focus of numerous investigations in recent years [2
In the TiO2
/UV photocatalytic oxidation process, hydroxyl radicals (·OH) are generated when catalyst (TiO2
) is illuminated by ultraviolet (UV) light. As a result, organic compounds are mineralized to CO2
O and inorganic constituents without the use of chemicals, thus avoiding the need for purchasing of chemicals and the associated chemical waste problems [4
]. In recent years, photocatalysis and membrane filtration have been proposed as a means for effective wastewater reclamation treatment [6
]. This hybrid system combines the advantages of both membrane filtration and photocatalytic degradation of contaminants. Membrane filtration is not only able to separate suspended catalysts, but can also enhance the effluent quality by selective separation at the molecular level if small pore size memranes are used [6
]. Furthermore, semiconductor photocatalysis is able to mineralize compounds causing membrane fouling and thus improve the consistency of the membrane’s operation [4
Several studies have reported on the performance of the photocatalysis and membrane hybrid system for the removal of organic contaminants from water and wastewater. In particular, this hybrid process was used to improve membrane flux and minimise fouling, as well as recovery of the TiO2
]. For example, Lee et al. observed that there was a significant flux decline occurred when a mixture of humic acid (HA) and TiO2
without UV irradiation was applied during ultrafiltration (UF). This is because the formation of a dense filter cake with HA molecules deposited in the void space of the TiO2
particle layers. On the other hand, in the presence of UV light, no flux decline was observed during 6 h of experimentation. The authors explained the observed phenomenon in terms of decomposition of HA into non-aromatic by-products that reduced the extent of organic compound adsorption onto TiO2
particles compared to the initial HA compounds [9
]. Lin et al. found that natural humic substances and other impurities in wastewater retard the photo degradation rates to a greater extent than commercial HA in the TiO2
/UV system [10
]. This is because the HA content of a water body can vary in composition in the natural environment, and thus the effects of natural humic substances cannot be accurately represented by a single commercial HA.
On the other hand, Le-Clech et al. considered the hybrid process for polishing treatment of surface water with a total organic carbon (TOC) concentration of 3 mg/L [11
]. They found that the polyvinylidene fluoride membrane process removed around 18% of the initial TOC concentration, while adsorption on TiO2
of 0.1 g/L and UV light alone achieved 5% and 70% TOC removal, respectively. They also observed synergistic effects when all the three process were used together. Kim et al. used a TiO2
mediated photocatalytic membrane reactor using submerged microfiltration (MF) membranes for the treatment of lake water and seawater [12
]. The authors observed that degradation of organic compounds in the seawater was minimal, whereas 80% removal of TOC was achieved in the case of lake water. The low mineralisation rate in the earlier case was mainly due to the presence of chloride ions in seawater, which were responsible for scavenging of the hydroxyl radicals, thus reducing the photo degradation efficiency. However, the authors did not find any significant membrane fouling during 4 h of filtration with seawater.
In their study, Ho et al. demonstrated that photo degradation with TiO2
/UV could effectively reduce membrane fouling and enhance the permeate flux of a submerged polyethylene membrane reactor when treating biologically treated sewage effluent [6
]. Similar conclusions were reported by Pidou et al. [13
], who found that photocatalytic pre-treatment of synthetic grey water reduced membrane fouling when a minimum UV residence time of 120 min in the continuous stirred-tank reactor (CSTR) was used. All of these studies demonstrated that application of photocatalytic treatment improves permeate flux due to decomposition of organic molecules.
All of the previous studies exclusively focused on the use of polymeric membranes, while little research has been performed with ceramic membranes. The use of ceramic membrane as an alternative to polymeric membrane has gained interest in water and wastewater treatment due to their superior physical integrity, chemical resistance and thermal stability, and, in turn, has lower chemical demand, lower cleaning frequency and longer lifetime compared to their polymeric counterparts [14
]. This suggests that ceramic membranes can operate at higher fluxes and tolerate extreme cleaning procedures without compromising membrane integrity. The decreasing cost of ceramic membranes, coupled with the recent successful development of others’ hybrid membrane processes such as ceramic-ozone process for wastewater applications [15
], allow new perspectives for the ceramic membrane-TiO2
/UV hybrid process. However, few studies have investigated the use of TiO2
/UV photocatalytic treatment as a pre-treatment to ceramic membranes. For example, Benotti et al. used TiO2
/UV and ceramic microfiltration membrane for the removal of pharmaceuticals, endocrine disrupting compounds and estrogenic activity from Colorado River water [16
]. It was reported that twenty-nine of the targeted compounds, in addition to total estrogenic activity, were removed by more than 70%, while only three compounds were removed by less than 50% with the highest level of treatment. No estrogenically active transformation products were formed during the treatment. However, the authors did not present any data on permeate flux behaviour during the process, and given the different surface and adsorption properties of ceramic membranes compared to polymeric membranes, this remains a significant unresolved issue.
The performance of the photocatalytic process is influenced by the various compounds present in the secondary effluent. In particular, the presence of cations and anions in the secondary effluent causes part of the catalyst surface to become unavailable for photon absorption and organic matter adsorption, thus resulting in lower catalytic reaction [17
]. Generally, the presence of various ions may affect the degradation rate via adsorption of the pollutants, reaction with hydroxyl radicals and absorption of UV light. There are several studies in the literature regarding the effects of various anions and cations [18
]. These studies demonstrated that CO32−
act as radical scavengers and also affect the adsorption process, and that Cl−
ions affect the adsorption step strongly, absorb UV light and have an undesirable effect on the photo degradation process, whereas other anions such as sulphate, phosphate and nitrate affect the degradation efficiency slightly. Yawalkar et al. have studied the effect of SO42−
ions on the overall degradation rates of phenol solutions and reported that detrimental effects on photocatalytic oxidation were observed in the order SO42−
]. Furthermore, anions such as Cl−
and cations such as Na+
present in the waters can be bound to TiO2
particles or very close to its surface, so that they can have substantial effects on the interfacial behaviour of the TiO2
Generally the presence of cations such as Na+
in the secondary effluent retards the photo degradation rates. These species are likely to reduce the rates of oxidation of organic compounds by competing for the oxidizing radicals or by blocking the active sites of the TiO2
]. According to Wang et al., cations such as Na+
are the common cations in natural water. They are all in the highest and most stable oxidation state and cannot capture electrons in the solution. It is hypothesized that these metal ions would not have a significant impact on the photodegradation [23
]. While it has been reported that these simple inorganic cations showed a slight suppression of the degradation reaction, this could be the effect of Cl−
co-present in solution [22
]. This is because Cl−
ions might inhibit the photodegradation due to adsorption on the surface of photocatalyst. It is also suggested that there is competitive adsorption by anions such as bicarbonate, sulphate, chloride and organic matter on the surface of catalyst, especially at lower concentrations of organic matter. Therefore, wastewater composition remains an important parameter in the overall performance of the photocatalysis.
Our previous work described the influence of experimental variables such as solution pH, salinity and TiO2
dose on the removal of a model HA solution using TiO2
/UV photocatalytic oxidation process and ceramic membrane filtration [8
]. These studies showed that relatively high removal of TOC and UV absorbance, removal of HA fouling potential and the complete recovery of TiO2
slurry using this hybrid system were achieved. Furthermore, it was shown that the presence of salt reduces the surface contact between the pollutant and the photocatalyst, and thus reduced photo degradation efficiency. However, for practical application, the feasibility of the combined photocatalysis and ceramic membrane filtration for the treatment of wastewater rather than model HA solution is of more interest. This is because membrane fouling varies with the characteristics of organic compounds present in wastewater, as does the oxidation efficiency of organic compound degradation by TiO2
Therefore, laboratory studies considered the efficiency of TiO2/UV with ceramic membranes for the treatment of a disinfected secondary effluent. The secondary effluent contained dissolved organic carbon (DOC) of approximately 9–10 mg/L, and a liquid chromatography-organic carbon detector (LC-OCD) technique was used to characterise the organic compounds in the effluent. To our knowledge, the present study is the first to report the application of this hybrid system to treat secondary effluent, and to characterise the organic compounds present at various stages of this hybrid process with LC-OCD analysis.
The performance of the hybrid photocatalysis-ceramic membrane system for treatment of secondary effluent was investigated and discussed. The obtained results led to the following conclusions.
The DOC and UV254 removal from the secondary effluent was significantly lower when lower TiO2 concentrations (0.5 g/L) were used, and even when a higher concentration of TiO2 (4 g/L) was used, the maximum removal of DOC and UV254 was only 23% and 52%, respectively, after 200 min of photocatalytic treatment. The reason proposed for the low removals compared to model HA systems is that the chemical structure of organic molecules present in the secondary effluent is less amenable to photocatalysis, and various salts contained in the wastewater can reduce the efficiency of oxidation via free radical scavenging.
The application of photocatalyst improved the permeate flux when higher TiO2 concentration was used. The flux improvement in a photocatalysis-ceramic membrane hybrid system was attributed to the formation of a more porous cake layer on the membrane covered with higher TiO2 particles when higher TiO2 concentration was used compared to that in lower TiO2 concentration. Furthermore, a positive effect of a photocatalyst on the permeate flux was observed after TiO2 adsorption and TiO2/UV treatment, as the filter cake had reduced organic compounds adsorbed between TiO2 particles in the filter cake.
LC-OCD analysis with online organic carbon detection revealed substances eluting in the biopolymer peak to be the main foulants in secondary effluent ceramic membrane filtration. While organic compound removal across the ceramic membrane after TiO2/UV photocatalytic treatment was low overall, the biopolymer fraction, as determined via LC-OCD, was removed to the greatest extent in accordance with its greater size. The hybrid system operated through non-selective hydroxyl radical oxidation and molecular separation, which progressively led to an effluent with lower biopolymer content that resulted in an increased proportion of the organic compounds being of LMW in the permeate. Substantial amounts of humic substances still remained after hybrid treatment. Monitoring the organic matter fractions with LC-OCD demonstrated that the reduction of effluent aromaticity (decreasing SUVA) was not strictly correlated with the complete depletion of humic substances in the effluents after hybrid treatment.