Scale-up of photoreactor with TiO₂ thin layer for wastewater treatment

Alarming levels of water pollution have been a serious problem in most countries all over the world. Various micropollutants have negative impacts on the environment regarding their permanent presence not only in wastewater but also in surface water, groundwater and drinking water. Among many water purification technologies currently in use, advanced oxidation processes (AOPs) have a great potential to remove a wide range of emerging contaminants. AOPs include the in-situ generation of highly reactive oxygen species, such as hydroxyl radicals (HO•), H₂O₂, O₃, and superoxide anionic radicals (⁠⁠), which provide pathways of mineralization to CO₂, H₂O, and inorganic ions or acids, or at least, destroy contaminants to residual compounds which are no longer ecotoxic. Reactive hydroxyl radicals are produced by one or more primary oxidants (e.g. ozone, hydrogen peroxide, oxygen) and/or energy sources (e.g. ultraviolet light) or catalysts (e.g. titanium dioxide). AOPs are currently of great interest and are being extensively studied for their potential use in removing non-biodegradable and highly stable pollutants from water (Feng et al. 2013; Kanakaraju et al. 2018; Cuerda-Correa et al. 2019; Sbardella 2019; Garrido-Cardenas et al. 2020; Pino et al. 2020; Mahdi et al. 2021 and many others).

Leyva-Díaz et al. (2021) studied 2,938 articles published related to the pollutant disposal from wastewater in the period of 1979–2020. They monitored the development of individual scientific methods to find the best practices. They found that the most commonly used oxidation process was photocatalysis on titanium dioxide (TiO₂) as the photocatalyst. However, in addition to it, many other heterogeneous photocatalysts based on metal oxides were tested, such as V₂O₅, ZnO, ZnS, Fe₂O₃, CdO, CdS, SrTiO₃, WO₃, and Al₂O₃. The mineralization of organic pollutants into CO₂, NH₄⁺, NO₃⁻, and SO₄²⁻ took place by the formation of radicals such as O₂⁻, HOO•, or OH• through a multistep reduction. Moreover, the mechanism of photocatalysis has also been widely studied, and basic principles of the method are sufficiently described (Ameta & Ameta 2018; Konstas et al. 2019; Fanourakis et al. 2020; Varma et al. 2020; Barros et al. 2021). Furthermore, photocatalysts can be combined, for example, with magnetic material coated with optical fibers, adsorbents or membranes, which improve their efficiency in water treatment. Advances in the research of photocatalytic materials have enabled the development of highly effective materials capable of degrading various contaminants, as well as the development of photocatalysts in visible light (Benotti et al. 2009; Ek et al. 2014; Fanourakis et al. 2020; Mansouri et al. 2021).

Titanium dioxide, either in its mineral forms or in a composite system, revealed a high potential as a photocatalyst due to its favourable physicochemical properties. In addition, compared to many other candidates, TiO₂ is almost the only material suitable for industrial applications, due to its effective photoactivity, high stability and low cost (Krakowiak et al. 2021). For this reason, there are many studies using TiO₂ as a photocatalyst aimed at determining optimal operating conditions in order to achieve the highest efficiency of degradation of pharmaceutical compounds (Kanakaraju et al. 2018; Shi et al. 2018; Zhu et al. 2019). In fact, there are many studies (Gusain et al. 2020; Kumar et al. 2020; Yamazaki et al. 2020) focusing on the application of photocatalysis for degradation of chemical pollutants. However, from the analysis of the results, it is evident that the decontamination efficiency differs significantly from author to author. It is obvious that the oxidation rate together with the efficiency of the photocatalytic system are highly dependent on a number of operating parameters which control the photodegradation of organic molecules. The basic parameters include substrate concentration, photocatalyst amount, solution pH, reaction medium temperature, light irradiation time, light intensity, photocatalyst surface, dissolution of oxygen in the reaction medium, nature of photocatalyst, nature of substrate, doping of metallic and nonmetallic ions, and structure of photocatalyst and substrate, their crystallinity, and crystallite size (Kumar & Pandey 2017; Gusain et al. 2020; Niu et al. 2020; Yamazaki et al. 2020). Moreover, the rate of photocatalytic degradation also depends on the type of pollutant, i.e., on its chemical structure, concentration, and the presence of other substances in water (Doll & Frimmel 2003; Karungamye 2020). Regrettably, the vast majority of these studies are carried out only on a laboratory scale.

Analyzing hundreds of studies devoted to this issue, it is clear that, essentially, the only scale-up solutions have been focused on so-called solar reactors (Abdel-Maksoud et al. 2016; Fendrich et al. 2019) either mostly tubular or falling film type. Glass tubes were placed in parabolic bells concentrating real solar radiation into the treated contaminated water, which is in contact with the photocatalyst. Water usually flowed through the irradiated tubes along with the TiO₂ nanoparticles. Glass tubes 1,250 mm long with the diameter of 32 mm and wall thickness of 1.4 mm were arranged on an aluminum base that reflects sunlight onto the tubes. The reactor volume was 7.7 L, while the liquid was in the recirculation mode at 20 L/min (Diaz-Angulo et al. 2020). Solar collector areas usually achieved units of m² (Prieto-Rodriguez et al. 2012; Cabrera-Reina et al. 2019). Another tested arrangement involved tubes with the irradiated area of 3.08 m² (Saggioro et al. 2014) or as a recirculating glass tube type model, with their radiated area of 0.91 m² (Pereira et al. 2011).

Within the falling film arrangement, water flows in a thin film along the irradiated wall (falling film solar reactors) (Colina-Marquez et al. 2016; Shnain et al. 2021). As a photocatalyst, nanoparticles of anatase, rutile, or their mixture, in concentrations of approximately 10–400 mg/L, were usually used. The areas of falling film reactors used to be over 1 m² (e.g. Abdel-Maksoud et al. 2018; Shnain et al. 2021). The largest detected irradiated area for the solar photocatalytic reactor of 25 m² was precisely for the falling flow type (Zayani et al. 2009). The mentioned solar photocatalytic reactors should be energetically advantageous; however, their applications are limited due to the possibility of these reactors to work only in specific geographical climatic conditions with sufficient solar radiation and only during the day (Chafie et al. 2018). Another solution, which is considered usable for so-called gray water, is a research membrane bioreactor with ultrafiltration and solar photocatalysis with a volume of 3 L, where photocatalysis with anatase particles is simultaneously applied (Ojobe et al. 2021).

Apart from the experimental modified membrane bioreactor mentioned above, we have not been able to find any pilot or operational solutions for the photocatalytic removal of pollutants with a TiO₂ photocatalyst, modified, for example, into a catalytic film in a closed reactor and irradiated with a suitable radiation emitter (UV, LED lamps) enabling the photocatalytic decomposition of pollutants, which would be capable of a scale-up for at least a pilot arrangement.

For these reasons, this study is devoted to the potential of TiO₂/UV photocatalyst for a scale-up. The efficiency of the catalyst was first studied on selected common pollutants in laboratory scale and compared with the classic sorption method further. Moreover, the study shows the potential of TiO₂ as a simple photocatalyst suitable for scale-up. The article is also focused on verifying the sufficient resistance and long life of the photocatalytic TiO₂ layer. The developed reactor will enable the continuous removal of emerging pollutants from real wastewater, especially from point sources using an end-of-the-pipe solution. The novelty of this reactor lies in a robust, however simple, and inexpensive device with simple operation and non-conflicting scale-up.

Two activated commercial charcoal Norit (grain 0.16–0.25 mm, S_BET 733 m² g⁻¹) and Supersorbon (pellets 4–7 mm, S_BET 1,258 m² g⁻¹) were used for the sorption comparable experiments. Specific surface areas (S_BET) were determined by physical adsorption of nitrogen at −195.8 °C on ASAP 2050 and 2020 instruments (Micromeritics) from the adsorption-desorption isotherm using a modified BET equation (Schneider 1995). The comparative sorption experiments were realized in a batch reactor using the orbital shaker GFL 3005, 300 rpm (Lauda DR. R. Wobser GMBH, Germany) with the volume of the reaction solution at 100 mL and amount of sorbent 1 g.

The initial concentrations of pollutants were 1 mg L⁻¹ for Diclofenac, Chloramphenicol and Triclosan in laboratory experiments. The reactions were carried out in a stirred beaker (volume of reaction solution 200 mL) with the prepared photocatalyst (5 g glass beads contained 2.5 mg active amount of TiO₂) or without a photocatalyst at laboratory temperature in the presence of UV light (UV lamp Philips HOK 4/120 SE, 400 W medium pressure mercury lamp (Phillips, the Netherlands) with a wavelength in the range of 250–420 nm). Pollutants concentrations were analysed on Waters Alliance HPLC module equipped with PDA detector, a column thermostat and Empower software. Toxicity tests were evaluated from bioluminescence inhibition of Vibrio fischeri bacteria. Analytical parts (pollutant concentrations and toxicity) were performed in external commercial laboratories. All laboratory experiments were duplicated and the standard deviation never exceeded 5%. The figures show the calculated average values.

Two widely used drugs, diclofenac (DFC) and chloramphenicol (CAM), in addition to one bacterial agent used in personal care products, triclosan (TCS), were selected for laboratory tests. Prior to the start of the photocatalytic reaction, reaction solution with photocatalyst was stirred and kept for 15 min in the dark. Only then the reaction solution was located under UV light. The sorption of the pollutant on the surface of the photocatalyst was low, up to a maximum of 2.5% of the initial concentration; 0.2% for DFC, 0.3% for CAM and 2.3% for TCS.

It is evident that decomposition of diclofenac was nearly the same involving photolysis, as well as photocatalysis (Figure 6(a)); however, toxicity of the resulting solutions revealed significant differences (Figure 6(b)). In the case of a photocatalytic reaction, the toxicity did not increase nor varied around zero, whereas in the case of the purely photolytic reaction, there was a significant increase in toxicity during the course of the photocatalytic reaction.

Thus, it can be concluded that the photocatalytic process is an effective way of removing drugs from water, and its practical application depends on the scale-up of the process and the durability, as well as the lifespan of the catalyst. Based on these results, a specially designed pilot unit was employed, for which a new methodology was developed for covering large areas by titanium oxide layers without the loss of surface strength and quality.

Based on these positive results, the prepared photocatalyst was subjected to an abrasion test. A disc newly covered by TiO₂ was used for the abrasion test. It consisted of two parts, 1,160 cycles with the weight of 0.25 kg and 2,900 cycles with the weight of 0.5 kg. After 4,060 cycles, only 40–50% of the TiO₂ layer was removed, which is an excellent result.

This study verified the potential of TiO₂/UV photocatalysis to remove various pollutants, i.e. diclofenac, chloramphenicol and triclosan, in laboratory and pilot plant tests using a new type of pilot plant reactor. Long-time tests with the pilot photocatalytic unit, which were successfully performed on two real water samples, wastewater from the pharmaceutical industry containing danazol and norethisterone with the concentration varying between 4 and 7 mg L⁻¹ and municipal wastewater containing emerging pollutants bisphenol A, 4-nonylphenol, estron, ethinylestradiol and triclosan in the concentrations between 50 and 600 ng L⁻¹, confirmed not only the efficiency of the photocatalytic system but also the durability and lifespan of the prepared TiO₂ photocatalytic layers. In all cases, the residual concentration was below the officially allowed limits even after six months of operation of the pilot plant photocatalytic unit. Durability of TiO₂ layers was corroborated on a durability tester, with an excellent result. The obtained results document that photocatalysis using TiO₂-based catalyst (mostly anatase) synthesized by a sol-gel technique is an effective method to remove emerging pollutants including pharmaceuticals, and can be conveniently scaled-up.

This study confirmed that the developed reactor enables the continuous removal of emerging pollutants from real wastewater, especially from point sources using an end-of-the-pipe solution from smaller ones, while it is a robust but simple and inexpensive device with simple operation. Further scale-up is possible by vertical setting or parallel addition of additional units. The incorporation of photocatalyst nanoparticles into the film enables long-term operation of the device with constant high remediation efficiency, as well as eliminating demanding separation of titania nanoparticles.

Financial support from the Technology Agency of the Czech Republic under the National Competence Centre Biocirtech [project No. TN01000048] is acknowledged.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.