Method for the removal of micropollutants by the application of heterogeneous catalytic ozonation from water

Field of invention

The invention belongs to the field of chemical engineering and specifically refers to the technology of heterogeneous catalytic ozonation for the removal of organic pollutants of low concentration (μg/L - ng/L), the so-called micropollutants, from groundwater and surface water for human consumption.

Existing state of the art

Research on water quality focuses mainly on nutrients, microbial contamination, heavy metals, and emerging organic pollutants. However, recent studies have shown that there are a number of organic pollutants that have a significant impact on water quality. These pollutants come from different sources and are usually found in concentrations ranging from ng/L to μg/L (micropollutants). These compounds, known as emerging organic contaminants (EOCs), are chemical compounds that have been present in the aquatic environment for years but have only recently been recognised as important pollutants. They are natural or synthetic and consist of compounds that are usually not monitored in the environment and are likely to have adverse effects on humans and the ecosystem. This group includes compounds such as pharmaceuticals, personal care products, pesticides, and hormones (O.M. Rodriguez-Narvaez et. al., Chem. Eng. J. 323 (2017) 361-380).

It is not appropriate for these compounds to remain in the environment, as their continued entry into the aquatic ecosystem creates negative impacts. Furthermore, natural attenuation and conventional treatment methods are not able to remove such micropollutants from both urban wastewater and industrial waste and from surface and groundwater used as drinking water. Therefore, solving this serious environmental problem requires the application of specialized treatment processes (L. Prieto- Rodriguez et. al., Water Res. 47 (2013) 1521-1528), as the present method of micropollutants removal using heterogeneous catalytic ozonation, described in detail below.

Micropollutants removal processes can be divided into three categories: phase change technologies, biological treatment, and advanced oxidation techniques. Phase change technologies have the ability to transfer the contaminant from one phase (e.g., aqueous) to another (e.g., solid). This category includes the adsorption process and membrane technology (O.M. Rodriguez-Narvaez et. al, Chem. Eng. J. 323 (2017) 361-380). The most widely used adsorbent material is activated carbon, with the use of which a large fraction of micropollutants are removed at rates greater than 90% in distilled water environments (R. Baccar et. al., Chem. Eng. J. 211-212 (2012) 310-317; D.P. Grover et. al, J. Hazard Mater. 185 (2011) 1005-1011; A. Zhou et. al., J. Phys. Chem. B 110 (2006) 4699-4707). However, the adsorptive capacity of adsorbent materials is related to the equilibrium concentration, which means that it is extremely low for pollutant removal at < 1 μg/L. Moreover, this process is not selective in terms of removing micropollutants from water, so its efficiency becomes even lower in the presence of organic compounds, such as natural organic matter (humic substances) (M. Opanga et. al, IJSRSET 4 (2018) 11, 223-230). Thus, the requirement of extremely low residual concentration - equilibrium - and the presence of generic organic load exponentially reduces the lifetime of activated carbon by increasing treatment costs to unacceptable levels, thus making this technology not techno-economically viable (S.W. Nam et. al, Environ, Eng. Sci. 34 (2017) 10, 752-761; L. Hernandez-Lead et. al., Water Res. 45 (2011) 2887-2896; A. Zhou et. al., J. Phys. Chem. B 110 (2006) 4699-4707).

As far as membrane technology is concerned, ultrafiltration, nanofiltration and reverse osmosis are mainly used for the removal of micropollutants. In this case, the removal depends on the type of membrane and the type of micropollutant. For example, a polysulfone ultrafiltration membrane can remove bisphenol A by 75% from distilled/pure water, while with a polyvinylidene membrane the removal reaches 98% (J. Heo et. al., Sep. Purif. Technol. 90 (2012) 39-52; A. Melo- Guimaraes et. al. Water Sci. Technol. 67 (2013)). At this point it should be clarified that these percentages have been determined by applying high initial concentrations, which significantly exceed the drinking water limit, as well as the concentrations of micropollutants present in the environment. In conclusion, ultrafiltration and nanofiltration are not effective processes for the removal of micropollutants, except in the specific cases of macro-molecules and polymeric organic compounds. The most effective membrane process is osmosis, due to the small size of the membrane pores, but it removes very few organic compounds (M. Bodzek & K. Konieczny Curr, Org. Chem. 22 (2018) 1070-1102; W. Gao et. al., Desalination, 272 (2011) 1-8) at a low rate (e.g., paracetamol at 50%). Membrane technology can be effective in removing some priority pollutants, but in this case, the challenge is the final disposal of the pollutants, as the treatment creates two different effluent streams, the treated flow, and the concentrated discharge-waste phase. Moreover, in the case of drinking water, membranes degrade its quality characteristics by removing nutrients (Ca - Mg - HCO ₃ - - SiO ₂ ) . In conclusion, in phase change technologies, the pollutants removed end up either in the solid phase during the adsorption process or in the discharge effluent during membrane separation. Therefore, by using this type of technologies the pollutants are not destroyed, they just change phase, continuing to be a problem for the environment (O.M. Rodriguez-Narvaez et. al., Chem. Eng. J. 323 (2017) 361-380).

Of the biological processes, the most widespread is that of activated sludge, as it has a significant efficiency (J. Simpa et. al., Desalination 250 (2010) 653-659). Biological processes are used for the treatment of urban wastewater and industrial waste in the form of aerobic and anaerobic biological processes. The performance of each process is related to the type of micropollutant and therefore they are combined with tertiary treatment processes (A.S. Stasinakis, Bioresour. Technol. 121 (2012) 432- 440; N Dafale et. al., Environ. Rev. 18 (2010) 21-36; J. Martin et. al., Sci. Total Environ. 503-504 (2015) 97-104). Biological processes readily remove mainly biodegradable compounds such as caffeine, diclofenac, trimethoprim, while they fail to remove the more difficult to biodegrade compounds such as bezafibrate, metaprolol, sulpiride (Q. Sui et. al., Environ. Sci. Technol. 45 (2011) 3341-3348). The micropollutants degradation efficiency is significantly related to the treatment conditions: aerobic or anaerobic. Removal rates range from zero (e.g., carbamazepine) during aerobic treatment to almost complete removal (97-100%) of naproxen by either aerobic or anaerobic biological treatment (J. Simpa et. al., Desalination 250 (2010) 653-659; J.L. Acero et. al., Water Air Soil Pollut. 226 (2015)). Also, certain substances, as well as environmental conditions in different geographical areas, can significantly affect the performance of the biological process. At present, there is a significant lack of knowledge regarding the microorganisms involved in the degradation of micropollutants, as well as the effect of other compounds co-existing in the waste. Many of the emerging micropollutants possess biological activity, such as antibiotics, so they behave negatively on the performance of the microorganisms involved in the conventional biological process (O.M. Rodriguez-Narvaez et. al., Chem. Eng. J. 323 (2017) 361-380). In conclusion, biological processes do not effectively remove micropollutants, therefore in the treated waste effluent it is necessary to incorporate a specialized micropollutants removal process.

Advanced Oxidation Processes (AOPs) are the best alternative - synergistic processes, due to their ability to show higher organic compound removal efficiencies than both "phase change" technologies and conventional biological processes. It should be clarified here that AOPs are generally applied: to the biological treatment effluent of urban wastewater as tertiary treatment, as well as to the primary treatment of groundwater and surface water mainly intended for drinking. The high rates of the process are related to the production of hydroxyl radicals with an oxidation potential of 2.8 V, which is their main characteristic (E. M. Cuerda-Correa et. al., Water 12 (2020) 102). Processes belonging to AOPs have different free radical production pathways and specific conditions determined by different materials (J. L. Wang & L. J. Xu, Crit. Rev. Environ. Sei. Technol. 42 (2011) 3, 251-325). Although the main characteristic is the production of hydroxyl radicals, the reactions by which they are produced, and the experimental conditions significantly vary the process performance (Q. Sui et. al, Environ. Sei. Technol. 45 (2011) 3341-3348; S. Sarkar et. al., J. Hazard. Mater. 278 (2014) 16-24). However, there is a significant gap in terms of full-scale process development, as there are few studies in this direction. Most AOPs research are mainly related to laboratory-scale photocatalysis processes (O.M. Rodriguez- Narvaez et. al., Chem. Eng. J. 323 (2017) 361-380), which however have a significant application problem for aqueous phase treatment at high continuous flow rates.

Ozone gas is produced on site from pure oxygen or atmospheric air by transforming oxygen in an electric field into ozone and introduced into the aqueous phase in the form of bubbles, resulting in a utilisation of less than 40%. Ozone is an effective disinfectant and oxidant in water and liquid waste treatment, as it can be driven to produce hydroxyl radicals ( ^● OH) either by direct or indirect oxidation. Reactions of soluble ozone with organic matter leads to the formation of aldehydes and carboxylic acids, which do not react with ozone. This is a major disadvantage of ozonation, as it cannot achieve complete mineralisation. Furthermore, these oxidation reactions are relatively slow and selective, whereas the reactions of hydroxyl radicals with inorganic and organic matter are rapid and non-selective. In contrast to single ozonation, the contact of soluble ozone with the catalyst surface (catalytic ozonation) allows efficient production of hydroxyl radicals even at low pH. The catalysts used for the controlled decomposition of ozone and the formation of hydroxyl radicals are typically metal oxides and hydroxy-oxides, metals in substrates, minerals, and carbons (J. Nawrocki & B. Kasprzyk-Hordern, Appl. Catal. B-Environ. 99 (2010) 27-42). Catalytic ozonation is a promising technology for the efficient removal of micropollutants from water and wastewater, which are resistant to conventional treatment methods. Its main advantages compared to non-catalytic methods are the optimisation of ozone use, the increase in micropollutants removal efficiency and a higher degree of mineralization (P.M. Alvarez et. al., Carbon 44 (2006) 3102-3112). However, both in the process of single ozonation and in that of catalytic ozonation, the formation of bromides (BrO ₃ -), which are carcinogenic ions for humans, is an important drawback. The production of bromates is related to the high ozone concentrations and contact time used for efficient removal of organic micropollutants from water (R. Joshi et. al., Crit. Rev. Environ. Sci. Technol. (2020)', A. W.C van der Helm et. al., Water Supply 5 (2005) 35-40). Thus, the present invention focuses on the full utilization of ozone by membrane diffusion and high efficiency of micropollutants removal by a catalytic process to minimize BrO ₃ - formation.

The literature related to the process of heterogeneous catalytic ozonation includes many conflicting results even in cases where the same materials have been used as catalysts. These results combined with the multitude of different experimental conditions that have been tested, do not clearly elucidate the mechanism of micropollutants degradation by this process. The following ozone depletion mechanisms have been proposed (J. Nawrocki & B. Kasprzyk-Hordern, Appl. Catal. B-Environ. 99 (2010) 27-42):

1. Ozone decomposes into the oxidised/reduced form of a metal deposited on the surface of a solid catalyst.

2. Ozone decays in the Lewis centres of metal oxides (AI ₂ O ₃ , TiO ₂ , ZrO ₂ etc.). The possibility that such a reaction takes place is based on the experimental observation that ozone decomposition is reduced in the presence of phosphate and sulphate ions, which are known to have a high affinity for Lewis centres.

3. Ozone decomposition takes place in the hydroxy groups of oxides and hydroxy oxides of metals.

4. In carbons, ozone decomposition takes place in the basic centres of their structure, minimising the catalytic process.

The literature is mainly based on finding the right material that will be an efficient catalyst and therefore catalytic ozonation is still at the laboratory level. Only a few attempts have been made for its technological application. An important factor for the application of the laboratory method to full-scale technology is the dosing of the catalyst by applying a continuous flow process. Most research show that increasing the catalyst dose results in an increase in system performance, regardless of the material used (Z Bai et. al., Environ. Sci. Pollut. Res. 25 (2018) 10090-10101', A. Khataee et. al, J. Tawain Inst. Chem. E. 77 (2017) 205-2015, G. Asgariet. al., JRHS 12 (2012) 2, 93-97). However, in these studies the catalyst dose is in the range of 1 - 100 mg/L, while using a micropollutant concentration of the same order of magnitude as the catalyst. These experimental conditions are not consistent with the corresponding actual conditions, since, as mentioned above, among emerging pollutants, micropollutants are called the compounds found in the environment at concentrations of ng/L-μg/L. Therefore, a heterogeneous catalytic ozonation unit for micropollutants removal should focus on micropollutants concentrations in the ng/L-μg/L range. Few studies have used micropollutants at μg/L concentrations and of these, few are those that study the effect of catalyst dose on the continuous flow process, and there are no studies of micropollutants removal at μg/L concentrations using a continuous flow process.

ZQ. Liu et. al, Sep. Purif. Technol. 200 (2018) 51-58, used in discontinuous (batch) experiments ZnO/graphite as a catalyst for DEP (diethylphthalate) removal with an initial concentration of 3 μM. The experiments showed that even at the low catalyst doses tested (0.05 g/L - 0.2 g/L), increasing the catalyst concentration resulted in a decrease in process efficiency. Even in cases where the concentration of the contaminant is high enough, to the point where it cannot be classified as a micropollutant, increasing the catalyst concentration has a positive effect up to a certain threshold, after which further increase in catalyst concentration reduces the removal of the contaminant. J. Peng et. al, Sep. Purif. Technol. 195 (2018) 138- 148, used the Ni/Al ₂ O ₃ catalyst to remove 200 mg/L succinic acid. Increasing the catalyst concentration to 10 mg/L resulted in almost 100% removal of the compound, while doubling the concentration reduced the removal to 86.1% at 60 min of the process. Similarly, A. Sukmilin et. al, Environ. Nat. Res. 17 (2019) 2, 87-95, used sludge from an urban wastewater treatment plant with 1% iron as a catalyst for the removal of 50 mg/L of phenol and observed that the optimum concentration for its removal was 1 g/L, while further increase in the catalyst dose led to a decrease in system performance. Similar results were observed in the study of N.H. S. Javadi et. al, J. Environ. Chem. Eng. 6 (2018) 6421-6430, who used magnetic alumina nanoparticles for the removal of benzotriazole (10 mg/L). With the addition of 0.5 g/L catalyst, the reaction kinetics increased by 55% compared to single ozonation, while no increase in system performance was observed with doubling the catalyst dose. However, there are studies in which although the concentration of the contaminant is low (μM), increasing the catalyst concentration seems to have a positive effect. The difference of these studies (Y. Ahnet. al., J. Environ. Chem. Eng. 5 (2017) 4, 3882-3894, S. Zhu et. al., Chem. Eng. J. 328 (2017) 527-535: M. Sui et. al, J. Hazard Mater. 227-228 (2012) 227-236: L. Jothinathan & J. Hu, Water Res. 134 (2018) 63-73) from the others lies in the fact that the doses of catalyst tested were very low compared to other studies, with a maximum of 0.4 g/L. Therefore, it is very likely that further increasing of catalyst concentration in these cases as well will have a negative effect on the removal of micropollutants.

All the studies mentioned above are laboratory and batch studies. Very few studies have been carried out in a pilot plant either in continuous operation or full scale. One of these tests was carried out by K. Wei et. al, Environ. Sci. Technol. 53 (2019) 6917-6926 with their own catalyst, CuCo/NiCAF. Their pilot plant is based on two-column operation. The catalyst (150 kg) is added to one column, which is a fixed- bed reactor, while single ozonation experiments (blind experiments) were carried out in parallel in the second column. The concentration of ozone, which was injected into the system in the form of bubbles, was 25 mg/L and the retention time was 30 min (optimum conditions). With this unit they wanted to treat the liquid waste produced by carbon gasification (secondary treatment, COD 70-80 mg/L). The catalyst showed high stability, as the column operated for three months without needing to be renewed. The aim of this research was to treat waste to reduce the concentration of COD, DOC, Tph, NH3-N, UV ₂₅₄ to the mg/L level; therefore, no information is given as to whether this process is able to remove micropollutants, which are found in very low concentrations in waste and especially in natural waters.

A similar pilot plant was used by J. Ma et. al., Sci. Rep. 8 (2018) 1, 1-11 for the treatment of dyehouse waste by heterogeneous catalytic ozonation. The unit consisted of a column divided into 6 levels, where each level contained the catalyst at a height of 910 mm (iron filings). The secondary effluent was fed directly into the reactor, as was the ozone in the form of bubbles. The effluent was recirculated through a recirculation pump. The total dose of catalyst used to fill the column was 350 g/L. The difference between this study and the previous one is that it investigated to some extent the effect of catalytic ozonation on the organic molecules in the dye effluent. Forty organic substances were detected, 38 of which were strongly polar molecules and two were non-polar molecules. After catalytic ozonation, all 40 substances were detected in the treated effluent, but in different proportions.

From a thorough study of the literature, it is understood that in order to achieve high rates of micropollutants removal, it is necessary to use a low dose of catalyst, so that the ozone in the structure of the material is not destroyed, but a large part of it can be broken down to produce hydroxyl radicals that will affect the micropollutants. This factor is very important for the operation of a heterogeneous catalytic ozonation unit to the extent that its design is based on it.

In parallel, patent applications have been filed for patents related to heterogeneous catalytic ozonation, but mainly involve the creation of new process-efficient materials (i.e., CN101759278, CN104289229, CN101891297A, CN11 1250087A, CN101811049A, CN107552067A). Few are based on the creation of continuous flow processes. The main differences of the processes are two: 1 ) The way of continuous diffusion of the ozone gas into the liquid without fixing the ozone dose and 2) the form and the way of using the catalyst as granules in column and powder under suspension.

Document WO 97/14657 presents a heterogeneous catalytic ozonation process in which the waste is first contacted with ozone to oxidise the easily degradable pollutants and dissolve the ozone in water, and then the pollutant-ozone solution is contacted with a solid catalyst, which has been activated by ozone, to oxidise the hardly degradable compounds. The effluent from the column can be recirculated for further contact with the catalyst. The originality of this patent for its time was the dissolution of the ozone gas in the liquid, so that the phases during catalytic ozonation are 2 (liquid-solid) rather than three (gas-liquid-solid).

Document CN203625105 presents a fixed bed reactor for the treatment of refinery waste. This process, like the previous one, includes a reactor for initial ozonation of the waste to degrade easily degradable compounds and then the pre-treated waste goes to the catalytic column to degrade the difficultly biodegradable organic molecules. The process was designed to enhance the utilisation of ozone and reduce its consumption. Excess ozone gas is recirculated and reused in the pre-treatment of fresh waste.

Another way of operating the catalytic column is presented in document CN 1223523. The catalyst in this case is activated carbon and the waste, as well as the ozone, enter the column from the bottom of the column in an upward flow. The treated waste from the top of the column is then fed into a 2 ^nd activated carbon column for biological stabilization treatment of the waste.

A two fixed-bed reactor system is also presented in document CN 104710002. Each reactor differs both in the catalyst it contains and in the way it feeds the ozone. In the first column the ozone is fed to the waste in the form of microbubbles (5-10 μm) and the catalyst is two elements and contains iron and manganese in its structure, while in the second column the catalyst consists of 3 metals and the ozone is fed in the form of nano-bubbles (100-400 nm).

In addition to fixed bed reactors, where the catalyst is in the form of a grain, there are also patents in which the catalyst is in the form of a powder and is suspended. Document CN 104529001 discloses a fluidised bed reactor where the catalyst is added to the system in the form of a powder. In this way, the efficiency is increased as the specific surface area of the catalyst is increased. Another patent where the catalyst is in suspension is US6, 8666788 B1. The reactor is divided into two chambers, in one of which the catalyst is in suspension due to a gas (air or oxygen) and enters the second chamber where the waste is located, in which the chemical oxidation takes place by adding oxidizing gas (ozone). The two chambers are separated by a vertical wall, which is open at both ends and allows the waste to be recirculated. For the successful operation of the system the amount of catalyst to be used is 100 g/L. In both cases the ozone gas is added to the system in the form of bubbles.

Document CN104529001A also describes a catalytic ozonation system with a fluidized bed reactor, which includes catalyst recirculation. This system is designed for the treatment of waste, mainly industrial waste, and focuses on COD reduction. The dissolution of ozone in the liquid is done through bubble diffusers thus increasing the energy requirements of the system by not fully utilising the ozone. Excess ozone can lead to the formation of bromine, which are products of secondary pollution.

In all these patents the performance of the systems was evaluated based on the reduction of high COD and TOC concentrations. Only one patent evaluates the removal performance of a specific micropollutant. Document ES2265728 deals with the removal of SDBS (sodium dodecylbenzene sulfonate) through the process of heterogeneous catalytic ozonation using activated carbon (PAC) as a catalyst. For this purpose, a 1 L reactor with agitation was set up, in which the pH and temperature conditions were controlled and remained constant, with the catalyst concentration ranging between 2.5 and 100 mg/L. In addition to the controlled conditions, this process did not involve either recycling of the catalyst or control of the ozone transport efficiency in the liquid, and therefore does not achieve the process optimisation and reduction of treatment costs as brought about by the application of the present invention discussed below. Also, due to the very good adsorptive properties of activated carbon, the removal of the micropollutants was due 60% to the adsorption process and 40% to the oxidation process. Besides, the performance of the system was tested with respect to a specific micropollutant rather than over a range of micropollutants with different physicochemical properties.

In all of the heterogeneous catalytic ozonation patents, ozone is added to the system in the form of bubbles. All attempts to optimise its use have been based on recirculating the excess ozone. Another way to achieve full utilisation of ozone is to use membranes to diffuse the ozone gas into the liquid without creating bubbles (bubble-less). Document EP3202721A1 describes a system to remove micropollutants using ozone and hydrogen peroxide (H ₂ O ₂ ), which describes the removal of micropollutants from water. Ozone is introduced into the reactor at a controlled concentration through a PVDF hollow fibre membrane without the formation of gaseous bubbles. Another patent combining ozone with membranes is CN105060458. In this case as well, the membranes used are hybrid PVDF membranes with nanoparticles in their structure that promote ozone decomposition. Hydroxyl radicals inhibit the growth and reproduction of microorganisms on the membrane surface resulting in reduced fouling to increase the lifetime of the membrane, but PVDF membranes do not have a long lifetime when using ozone. A different way for controlled ozone channelling and reduction of bromine production was proposed in document US6024882. The researchers designed a pressurized plug-flow reactor where ozone and H ₂ O ₂ are injected into the system in small amounts at different points in the reactor. However, this system is quite complex, large, with many valves and requires close monitoring during operation.

Patent CN101050036A focuses on the control of bromine production through the catalytic ozonation process. Materials containing cerium oxide (CeO ₂ ) in their structure are proposed as catalysts. Such catalysts have the ability to reduce the production of hydroxyl radicals, which are responsible for the formation of bromine. The reduced production of hydroxyl radicals makes the process inefficient in terms of micropollutants removal, as this is largely dependent on the strong oxidising action of hydroxyl radicals.

In conclusion, no technology has been developed for the removal of organic micropollutants from water intended for drinking through the heterogeneous catalytic ozonation process, which implements the advantages of a plug-flow reactor and at the same time the full utilisation of ozone.

The present invention solves the technical problem of removing micropollutants from water by briefly including the following processes:

- Step 1 : Removal of suspended particles for non-clogging of membranes and optimisation of hydroxyl radical production.

- Step 2: pH adjustment for CaCO ₃ nucleation development.

- Step 3: Full utilization of ozone by diffusion into water through a hollow fibre membrane.

- Step 4: Use of a plug-flow reactor airtightly closed creating low back-pressure gravity pressure based on the overlying level above the ozone diffusion membrane, to prevent ozone loss by diffusion into the removed gas phase and thus total utilisation (full consumption) of the added ozone, reducing fixed and energy costs. The rapid reaction kinetics, which implies a micropollutant destruction efficiency of more than 90% in up to 2 min, also implies a corresponding reduction in ozone concentration, resulting in the prevention of BrO ₃ - formation, as the catalytic oxidation of micropollutants is completed using a plug-flow reactor in less than 10 min.

- Step 5: Reuse of the catalyst by recirculating in the context of the circular economy, reducing the operational cost of processing.

- Step 6: Biological filtration for optimal removal of the molecules resulting from the breakdown of the micropollutants.

Brief description of the drawings

The present invention is presented in the following description of drawings and examples of embodiments.

Figure 1 shows the flowchart of the organic micropollutant degradation/decomposition process proposed in the present invention by the application of heterogeneous catalytic ozonation.

Figure 2 shows the diagram of the activity coefficients of ions in aqueous solutions, calculated from the Debye-Huckel and Guntelberg equations. Figure 3 shows the flow diagram of the plug-flow catalytic ozonation reactor, which generates low back- pressure, gravity pressure and contributes to the total dissolution of ozone.

Figure 4 shows the correlation diagram of the pH value with the concentration of CO ₂ .

Description of the invention

The present invention relates to the method of catalytic ozonation with high efficiency of degradation of micropollutants from water intended for human consumption. In particular, it achieves the reduction of the concentration of a micropollutant, typically not exceeding 10 μg/L, with the following technical characteristics.

- Reduction of the micropollutant concentration below the drinking water limit (mainly < 0.1 μg/L).

- Application as a continuous flow process.

- Operating time less than 10 min.

- No degradation of the physical and chemical characteristics of the water.

- Acceptable capital and operating costs.

The present catalytic ozonation process includes the following equipment (Figure 1):

1. CO ₂ storage tank; 2. Blower for stirring the pre-treated water to remove CO ₂ and increase the pH; 3. In line mixer for adding CO ₂ ; 4. pH-meter to control and adjust the pH of the water; 5. Pre-treated water storage tank for pH adjustment, 6. Pump for feeding the water to be treated;

7. Hollow fibre membrane for diffusion of ozone into the water; 8. High purity O ₂ production generator; 9. O ₃ production generator, 10. O ₃ -meter for regulating the ozone feed flow rate; 11. In-line mixer for dispersing the catalyst in water, 12. Plug-flow reactor; 13. Catalyst suspension stirrer; 14. Catalyst suspension storage tank; 15. Ultrafiltration membrane tank; 16. Catalyst recirculation pump; 17. Suction pump for suction of treated water from the ultrafiltration membranes; 18. Air supply blower to suspend membrane clogging; 19. PLC (Programmable Logic Controller); 20. Clean water tank; 21. Biological Filtration Cell. Also shown in Figure 1 are: treated water after particle removal (A); air (B) for feeding to the clean oxygen production unit; and treated water (C). The treatment modules are described below.

STEP 1. Removal of suspended particles

All groundwater and surface water contains a significant number of suspended particles, which have a negative effect on the catalytic ozonation process:

- Creating clogging problems in the ozone diffusion hollow fibre membrane and accumulation in the catalyst suspension.

- They neutralise the hydroxyl radicals produced by the ozonation catalysts, significantly reducing the decomposition efficiency of the micropollutants.

For this reason, pre-treatment of the water is required to remove suspended particles either by the classical flocculation-agglomeration-filtration process, or by simple filter filtration, which additionally removes partly natural humic components, improving the efficiency of catalytic ozonation.

Step 2: Adjust the pH of the water before starting the catalytic ozonation process.

Most importantly, the determination of the saturation pH of CaCO ₃ (pH _s ) in water is required. Usually, the pH of surface waters is significantly higher than pH _s due to CO ₂ uptake by photosynthetic algae, while the pH of groundwater is usually lower than pH _s . Subsequently, adjusting the pH in the pH _s < pH < pH _s + 0.2 range (Figure 1 ), which on the one hand entails negligible operational costs, but on the other hand leads to CaCO ₃ nucleation, which in this pH range does not generate any precipitation.

- Reducing pH by adding CO ₂ . Figure 4 specifies the determination of CO ₂ addition dose with in-line mixer in the range of 20±10 g/m ³ to adjust the pH in the range pH _s <pH<pH _s +0.2.

- Increase the pH with natural ventilation to remove CO ₂ .

Table 1. Temperature dependence of partial equilibrium constants of carbonates

The pH _s is calculated thermodynamically as follows: pH _s = pK _α,2 - pK _sp + p[Ca ²⁺ ] + p[HCO ₃ ] - logγ _Ca 2+ - logγ _HCO3-

From Table 1 , the values of the equilibrium constants of carbonate ionic forms with respect to temperature are selected to determine the saturation pH of CaCO ₃ (pH _s ) in order to adjust the pH of water for CaCO ₃ nucleation.

For example, suppose the concentration of calcium in water is [Ca ²⁺ ]= 84 mg/L and the concentration of [HCO ₃ - ]= 351 mg/L and the conductivity is 730 μmhos/cm:

The Molecular Weight of calcium implies Mrca2 ⁺ = 40 g/mol, so the moles of calcium:

The Molecular Weight of HCO ₃ implies Mr _HCO3 - = 61 g/mol, so the moles of HCO ₃ ^_ :

To calculate the activity coefficient of ions in aqueous solutions, knowledge of the ionic strength is necessary. The ionic strength (p) is calculated from the relationship (Figure 2): μ = 1.6X10 ^-5 x SEC where SEC is the Specific Electrical Conductivity in μmhos/cm, which applies to conductivity values less than 10.000 μmhos/cm and ionic strength p < 0.1 , which is consistent with the water quality to be treated by the present method. μ = 1.6x10 ^-5 x (specific electrical conductivity, μmhos/cm).

Based on conductivity = 730 pS/cm or 730 μmhos/cm, → μ = 1.6x10 ^-5 x 730 → μ = 0.012 Based on Figure 2, we observe that:

For μ = 0.012 the activity coefficient of Ca ²⁺ is 0.68 and of HCO ₃ 0.90

Therefore, the activity of Ca ²⁺ is: -log 0.68 = 0.17 → -log _Ca 2+ = 0. 17

The activity of the HCO ₃ is: - log 0 . 90 = 0.05→ -log _HCO3 = 0. 05

Therefore, for a temperature of 10 °C. pH _s = pK _α,2 - pK _sp + p[Ca ²⁺ ] + p[HCO ₃ ] - logγ _Ca 2 ₊ - logy _H co3 = 10.38 - 8.28 + 2.68 +2.24 + 0.17 + 0.05 → pH _s = 7.24

STEP 3. Adding ozone by diffusion to a hollow fibre membrane, which:

1. In combination with the plug-flow reactor, which generates low back-pressure gravity pressure with an overlying level of 2 - 6 m above the ozone diffusion membrane, it results in total ozone dissolution due to the synergistic effect of back-pressure.

2. In the presence of CaCO ₃ cores, the ozone diffusion hollow fibre membrane also acts as a plug-flow catalytic ozonation reactor, increasing the micropollutants decomposition efficiency up to 80%, while in the absence of CaCO ₃ cores the efficiency with single ozonation does not exceed 25%, as the retention time of water in the hollow fibre membrane does not exceed 1 min.

In conclusion, in the present method, it is the first time that the growth and use of CaCO ₃ cores as the first stage of catalytic ozonation within the hollow fibres of the ozone diffusion membrane is presented, exponentially increasing the decomposition efficiency of micropollutants in 1 min. Also, the application of back-pressure implies the complete dissolution/utilization of ozone. Thus, the low residual micropollutants concentration in the second stage of catalytic ozonation at the inlet of the plug-flow reactor implies a lower ozone concentration requirement, resulting in the prevention of BrO ₃ - formation in a contact time of up to 10 min.

STEP 4. Catalytic ozonation in a plug-flow reactor.

The decomposition efficiency of micropollutants is affected by the type and concentration of catalyst, as well as by the ozone concentration. The most efficient catalysts are perlite, zeolite, talc and SiO ₂ , especially after thermal treatment to remove crystalline waters and hydroxyls, as well as iron hydroxy- oxide in the form of goethite. The catalyst application requires:

- Choice of catalyst type depending on the type of micropollutant.

- Completion of the process in up to 10 min.

- A catalyst concentration between 0.1 - 1 g/L with an optimum concentration range of 0.5 ± 0.2 g/L in the form of a particle size of 5 - 50 μm. The optimum catalyst concentration is related to the type of micropollutant and water temperature.

- Addition of catalyst suspension to the ozonated water at the exit of the water from the hollow fibre membrane and homogenisation of the dispersion with an in-line mixer (in-line mixer - Figure 1 - number 11 ). Changing the recirculation flow rate of the catalyst suspension enables the catalyst concentration to be changed within 10 min.

- Calculation of settling of catalyst particles with maximum diameter ds = 50 μm and maximum apparent porosity < 3,500 kg/m ³ and water density 1000 kg/m ³ : (precipitation in the Stokes region) Therefore, U

Calculated the

Thus, the initial assumption is correct, so substituting the Reynolds number into the equation Uk = 3.4x10 ^-3 m/s or 12.25 m/h for a maximum particle size of 50 μm and a maximum solid density of 3,500 kg/m ³ . For safety reasons the flow rate should exceed 60 m/h due to upward/downward flow in the plug-flow reactor.

- The reactor will have 10 consecutive chambers (Figure 3) with a depth of 2 - 6 m and a retention time in each chamber of 0.5 -1 min, with water being introduced into the first chamber at the bottom and overflowing from the last chamber.

In conclusion, the present method uses a plug-flow reactor without continuous stirring and full utilization of ozone. The advantage of applying plug-flow reactor is the high efficiency of micropollutants degradation in a short retention time. The retention time in a complete mixing reactor (CSTR) to reduce the initial concentration (CAO) of the micropollutant to a residual concentration (C _AE ) is:

The corresponding time using a plug-flow reactor (PF) is:

So, for 95% and 99% degradation efficiency the ratio of the required retention time is: for efficiency for efficiency

And for a degradation efficiency o Based on the design of the plug-flow reactor (PF) with a maximum retention time of 10 min, using a complete mixing reactor (CSTR), for 95% efficiency a retention time of 60 min is required, for 99% efficiency a retention time of 220 min is required, and for 99.9% efficiency a retention time of 1.450 min is required. STEP 5. Catalyst separation and recirculation.

- The effluent from the plug-flow reactor enters the ultrafiltration membrane tank, where the treated water is separated from the catalyst (Figure 1 - number 15).

- The catalyst remains in the storage tank in suspension form in communication with the membrane tank, which is returned by recirculation and mixed with the ozonated water at the outlet of the ozone addition hollow fibre membrane by diffusion.

- For optimal membrane performance and lifetime, the catalyst concentration in the storage tank ranges from 3 to 10 g/L with an optimum concentration of 5±2 g/L.

- Based on the catalyst concentration in the water to be treated in the range 0.1 - 1 g/L = 100 - 1000 g/m ³ , the recirculation (Q _r ) will vary with respect to the treatment flow rate (Q) in the range 0.02 Q < Qr < 0.2Q.

- To inhibit clogging of the ultrafiltration membranes, they shall be cleaned every 10 min by reverse flow/rinsing of treated water and air injection for 1 min.

In conclusion, the present method involves the reuse of the catalyst, which includes 2 important advantages: The use of a catalyst that is pre-ozonated resulting in maximum efficiency, as well as the continuous reuse of the catalyst meeting the criteria of circular economy, while reducing the operational cost of treatment.

STEP 6. Biological filtration process of the treated water with catalytic ozonation.

Groundwater and surface water micropollutants are generally not biodegradable. Catalytic ozonation disorganises a proportion of organic micropollutants and a proportion breaks them down into smaller molecules that are largely biodegradable.

The water treatment is completed with the biological filtration process for optimal removal of the molecules resulting from the breakdown of micropollutants during catalytic ozonation. It involves the use of an activated carbon bed with a height H > 1.2 m and a vacuum velocity V = 7 ± 2 m/h, so that the vacuum time exceeds 10 min (EBCT > 10 min).

This method involves the removal of organic micropollutants belonging to the following categories: pharmaceutical compounds, personal care products, steroid hormones, pesticides/insecticides, surfactants, and super-fluorinated compounds at a concentration of less than 10 μg/L.

Examples of application of the invention

Example of application 1

Surface reservoir water with a temperature of 20 °C and a daily flow rate Q = 2,400 m ³ /day has suspended solids (SS) 5 mg/L, pH 7.8, TOC 1.6 mg/L and the organic micropollutant benzotriazole (BZT) at a concentration of 2 μg/L. The pH _s of CaCO ₃ saturation in the water is 7.25.

> Selection of processing parameters:

To reduce the concentration of BZT below the 0.1 μg/L limit of potability, thermally treated at 800 °C expanded zeolite and in a particle size range of 5 - 50 μm will be used as catalyst. Also, due to the relatively good water temperature and high kinetic reaction of BZT with hydroxyl radicals and ozone, a catalyst concentration of 350±50 mg/L, retention time in the plug-flow reactor of 6 min and an ozone concentration requirement of 2 mg/L due to high TOC concentration are selected. To avoid membrane and plug-flow reactor blockages the pH will be reduced to 7.45 = pH _s + 0.2.

1 ^st processing stage:

Removal of suspended particles by flocculation - coagulation - filtration, with application of poly- aluminium chloride (PACI) at a dose of 2 mg Al/L and 0.3 mg/L cationic polyelectrolyte. To optimise the removal efficiency of suspended particles with PACI, as for a residual aluminium concentration significantly below the drinkability limit of 200 μg/L, a reduction of pH < 7.5 is required. Reducing pH 7.5 requires the addition of 20 mg CO ₂ /L = 20 g CO ₂ /m ³ , i.e. 1 kg CO ₂ /50 m ³ (Figure 4) at a cost of ~ 0.25€/50 m ³ .

Pre-treated water parameters: SS < 0.1 mg/L, pH = 7.45, TOC = 0.9 mg/L and BZT = 1.9 μg/L

2 ^nd processing stage:

In pre-treated water with pH = 7.45 = pH _s + 0.2 the nucleation of CaCO ₃ is favoured. Subsequently, feed to the PTFE hollow fibre membranes for the addition of ozone by diffusion.

- Based on the flow rate Q = 2,400 m ³ /day = 100 m ³ /h, membranes with a total surface area of 1.7x10 ³ m ² are selected so that the ozone diffusion parameter is 60 L/m ² h.

- The ozone concentration, using an oxygen content of O ₂ > 98%, in the gas phase is 180 g/m ³ . Thus, based on the ozone requirement of 2 mg/L = 2 g/m ³ , the ozone supply will be:

MO ₃ = 100 m ³ /h x 2 g/m ³ = 200 g/h → Q _O3 = (200 g/h) / (180 g/m ³ ) = 1.11 1 m ³ /h = 18.5 L/min

Parameters at the output of hollow fibre membranes:

C _O3 = 0.8 mg/L, BZT = 0.35 μg/L, pH = 7.45, and TOC = 0.8 mg/L

3 ^rd processing stage:

Catalyst introduction and treatment in a plug-flow reactor.

In the outgoing water from the hollow fibre membranes, an in-line mixer is used to disperse an expanded zeolite suspension with a grain size of 5 - 50 μm, a flow rate of 7 m ³ /h and a concentration of 5 g/L = 5,000 g/m ³ as a catalyst, so that the concentration of the catalyst reactor is: C _zeolite = [(7 m ³ /h) x (5,000 g/m ³ )] / (107 m ³ /h) = 327 g/m ³

Based on the reaction time requirement of 6 min, the volume of the 10-chamber reactor is:

V _PF = (107 m ³ /h) x (6min/60min/h) = 10.7 m ³

Choosing a depth of 4 m for each reactor chamber means that the total length of the reactor is:

L = (10 chambers) x (4 m/chamber) = 40 m So, the cross-section of each chamber is E = V _PF / L = 0.2675 m ² → L = W = 0.52 m

Water flow velocity in the plug-flow reactor ensures the fluid suspension of the catalyst:

V = (107 m ³ /h) / (0.2675 m ² ) = 400 m/h

Parameters at the outlet of the plug-flow reactor:

C _O3 = 0.01 mg/L, BZT = 0.01 μg/L, pH = 7.4, and TOC = 0.4 mg/L

4 ^th processing stage:

Separation of the catalyst from the treated water and recirculation to the inlet of the plug-flow reactor. For filtration of 100 m ³ /h under suction, based on an optimum filtration flux of 25 L/m ² h = 0.025 m ³ /m ² h, a membrane surface area is required:

Q = Flux x A _Membrane A→ 100 m ³ /h = (0.025 m ³ /m ² h) x A → A = 4,000 m ²

To cover the inoperability of each membrane for 1 min every 10 min due to backwashing, 10 surface blocks are selected, each 500 m ² , dimensionally:

LxWxH =1.152 x 1.298 x 2.763 m

The blocks are placed in a row, with a distance between them of 0.5 m and 1 m distance from the wall then:

Tank depth = 4 m

A power pump will be used to suck the treated water from the membrane blocks:

A power pump will be used to backwash each membrane block sequentially for 1 min:

During the backwashing of each membrane block, air with a specific flow rate of 0.25 m ³ /m ² h will be supplied at the same time. So the air supply of the blower will be:

Q _air = 500 m ² x 0.25 m ³ /m ² h = 125 m ³ /h

Calculating the power of the air compressor: Where:

W = air supply weight = 1

R = global gas constant =

Ti = absolute inlet temperature (K) = 293 K (20 °C) n = 0.283 (for air) e = efficiency (typical) for compressor = 0.65

P ₁ = absolute inlet pressure (atm) = 1 atm

P ₂ = absolute outlet pressure (atm) = 1.5 atm 2.3 kW

For the storage (buffer) of a quantity of catalyst suspension to ensure recirculation, a tank LxWxH = 2.5 x 2.5 x 4 m, with a net volume of 20 m ³ , is constructed in communication with the membrane tank. A power stirrer (Drawing 1 - number 13) will be installed to maintain the catalyst suspension in homogeneity:

I _mixer = 20 m ³ x 100 W/m ³ = 2,000 W = 2 kW

The power of the catalyst recirculation pump will be:

P = = 450 W

0.65

Water parameters after catalyst filtration with ultrafiltration membranes:

C _O3 < 0.01 mg/L, BZT < 0.01 μg/L, pH = 7.45, and TOC = 0.4 mg/L

5 ^th processing stage:

A biological filtration process of the treated water, involving the use of an activated carbon bed with a height H = 1 .2 m and a vacuum velocity U = 6 m/h, so that the vacuum time (EBCT) exceeds 10 min.

- Bed surface: E = (100 m ³ /h)/(6 m ³ /h) = 16.7 m ²

- EBCT = H/U = (1 .2 m)/(6 m/h) = 0.2 h → EBCT = 0.2 h x 60 min/h = 12 min

' Volume of activated carbon bed: V _EA = E x H = 16.7 m ² x 1.2 m = 20 m ³

Parameters at the outlet of the biological filtration bed:

C _O3 = 0 mg/L, BZT < 0.01 μg/L, pH = 7.45, TOC = 0.3 mg/L and

BrO ₃ - = 5 μg/L < 10 μg/L = limit of potability

Example of application 2

Drilling water with a temperature of 16.5 °C and a daily flow rate Q = 1 ,200 m ³ /day has suspended solids (SS) of 0.7 mg/L, pH 7.05, TOC 0.4 mg/L and the organic micropollutant atrazine (ATZ) at a concentration of 0.7 μg/L. The pH _s saturation of calcium carbonate in the water is 7.25. > Selection of processing parameters:

To reduce the concentration of ATZ below the 0.1 μg/L potability limit, thermally treated SiO ₂ at 800 °C and in a particle size range of 5 - 50 μm will be used as a catalyst. Also, due to the relatively low water temperature and low kinetic reaction of ATZ with hydroxyl radicals and ozone, a catalyst concentration of 750±50 mg/L, a retention time in the plug-flow reactor of 10 min and an ozone concentration requirement of 1 mg/L due to low TOC concentration are selected. Control of LOD removal without changing the pH of the water.

1 ^st processing stage:

To avoid possible clogging of the ozone diffusion membrane, filtration treatment with 1 μm bag filters is preceded, due to the low concentration of suspended solids.

Pre-treated water parameters: SS < 0.1 mg/L, pH = 7.05, TOC = 0.4 mg/L and ATZ = 0.7 μg/L

2 ^nd processing stage:

In pre-treated water with pH = 7.05 < pH _s = 7.25, the production of hydroxyl radicals is not favoured, resulting in low removal efficiency of ATZ during the passage through PTFE films of hollow fibre ozone addition by diffusion.

- Based on the flow rate Q = 1 ,200 m ³ /day = 50 m ³ /h, membranes with a total surface area of 10 ³ m ² are selected so that the ozone diffusion parameter is 50 L/m ² h.

- The ozone concentration, using an oxygen content of O ₂ > 98%, in the gas phase is 180 g/m ³ .

Thus, based on the ozone requirement of 1 mg/L = 1 g/m ³ , the ozone supply will be:

MO ₃ = 50 m ³ /h x 1 g/m ³ = 50 g/h Q _O3 = (50 g/h) / (180 g/m ³ ) = 0.28 m ³ /h = 4.6 L/min

Parameters at the output of hollow fibre membranes:

C _O3 = 0.7 mg/L, pH = 7.1, TOC = 0.3 mg/L and ATZ = 0.55 μg/L

3 ^rd processing stage:

Catalyst introduction and treatment in a plug-flow reactor.

In the outgoing water from the hollow fibre membranes, a SiO ₂ suspension with a particle size of 5 - 50 μm, a flow rate of 6 m ³ /h and a concentration of 7 g/L = 7,000 g/m ³ is dispersed with an "in line mixer" as a catalyst, so that the catalyst concentration of the reactor is:

C _siO2 = [(6 m ³ /h) x (7,000 g/m ³ )] / (56 m ³ /h) = 750 g/m ³

Based on the reaction time requirement of 6 min, the volume of the 10-chamber reactor is:

V _PF = (56 m ³ /h) x (10min/60min/h) = 9.34 m ³

Choosing a depth of 4 m for each reactor chamber means that the total length of the reactor is:

L = (10 chambers) x (4 m/chamber) = 40 m

So the cross-section of each chamber is: E = V _PF /L = 0.234 m ² → L = W = 0.48 m Water flow velocity in the plug-flow reactor ensures the fluid suspension of the catalyst:

V = (56 m ³ /h) / (0.234 m ² ) = 240 m/h

Parameters at the outlet of the plug-flow reactor:

C _O3 - 0.015 mg/L, pH = 7.1, TOC = 0.3 mg/L and ATZ = 0.15 μg/L

4 ^th processing stage:

Separation of the catalyst from the treated water and recirculation to the inlet of the plug-flow reactor. For filtration of 50 m ³ /h under suction, based on an optimum filtration flux of 20 L/m ² h = 0.02 m ³ /m ² h, a membrane surface area is required:

Q = Flux x A _Membrane → 50 m ³ /h = (0.02 m ³ /m ² h) x A → A = 2,500 m ²

To cover the inoperability of each membrane for 1 min every 10 min due to backwashing, 6 surface blocks are selected, each 500 m ² , dimensionally:

LxWxH =1.152 x 1.298 x 2.763 m

The blocks are lined up in a row, with a distance between them of 0.5 m and 1 m distance from the wall then:

L _system = ( 1 . 1 52 X 6) + (0.5 X 4) + (1 X 2) → L _system ⁼ 1 0.9 m

W _system = 1 .298 + ( 1 X 2) → W _system ⁼ 3.3 m

Tank depth = 4m

A power pump will be used to suck the treated water from the membrane blocks:

P = [p * g * Ah * Q ]/0,65

A power pump will be used to backwash each membrane block sequentially for 1 min:

During the backwashing of each membrane block, air with a specific flow rate of 0.25 m ³ /m ² h will be supplied at the same time. So, the air supply of the blower will be:

Q _air = 500 m ² x 0.25 m ³ /m ² h = 125 m ³ /h

Calculating the power of the air compressor:

Where:

- W = air supply weight =

R = global gas constant =

T ₁ = absolute inlet temperature (K) = 293 K (20 °C)

- n = 0.283 (for air) e = efficiency (typical) for compressor = 0.65

Pi = absolute inlet pressure (atm) = 1 atm

P2 = absolute outlet pressure (atm) = 1 .5 atm 2.3 kW

For the storage (buffer) of a quantity of catalyst suspension to ensure recirculation, a tank LxWxH =

2.5 x 2.5 x 4 m, with a net volume of 20 m ³ , is constructed in communication with the membrane tank.

A power stirrer (Drawing 1 - number 13) will be installed to maintain the catalyst suspension in homogeneity:

I _mixer ⁼ 20 m ³ x 100 W/m ³ = 2000W = 2 kW

The power of the catalyst recirculation pump will be:

Water parameters after catalyst filtration with ultrafiltration membranes:

CO ₃ < 0.01 mg/L, pH = 7.15, TOC = 0.3 mg/L and ATZ = 0.13 μg/L

5 ^th processing stage:

A biological filtration process of the treated water, involving the use of an activated carbon bed with a height H = 1.5 m and a vacuum velocity U = 5 m/h, so that the vacuum time (EBCT) significantly exceeds 10 min, due to low bio-degradation of the ATZ.

- Bed surface: E = (50 m ³ /h)/ (5 m ³ /h) = 10 m ²

- EBCT = H/U = (1 .5 m)/(5 m/h) = 0.3 h EBCT = 0.3 h x 60 min/h = 18 min

- Volume of activated carbon bed: V _EA = E x H = 10 m ² x 1.5 m = 15 m ³

Parameters at the outlet of the biological filtration bed:

CO ₃ = 0 mg/L, pH = 7.2, TOC = 0.25 mg/L and ATZ = 0.12 μg/L > 0,1 μg/L = limit of potability and BrO ₃ ^_ = 9 μg/L < 10 μg/L = limit of potability

Example of application 3

Drilling water with a temperature of 16.5° C and a daily flow rate Q = 1 ,200 m ³ /day has suspended solids (SS) of 0.7 mg/L, pH 7.05, TOC 0.4 mg/L and the organic micropollutant atrazine (ATZ) at a concentration of 0.7 μg/L. The pH _s saturation of calcium carbonate in the water is 7.25.

> Selection of processing parameters:

To reduce the concentration of ATZ below the 0.1 μg/L porosity limit, thermally treated SiO ₂ at 800 °C and in a particle size range of 5 - 50 μm will be used as a catalyst. Also, due to the relatively low water temperature and low kinetic reaction of ATZ with hydroxyl radicals and ozone, a catalyst concentration of 750±50 mg/L, a retention time in the plug-flow reactor of 10 min and an ozone concentration requirement of 1 mg/L due to low TOC concentration are selected. Control of ATZ removal by changing the pH of the water to pH _s + 0,2.

1 ^st processing stage:

To avoid possible clogging of the ozone diffusion membrane, a filtration treatment with 1 μm bag filters shall be performed prior to filtration, due to the low concentration of suspended solids.

Pre-treated water parameters: SS < 0.1 mg/L, pH = 7.05, TOC = 0.4 mg/L and ATZ = 0.7 μg/L

2 ^nd processing stage:

For the synergistic catalytic action of the CaCO ₃ nuclei, the filtered water is collected in a tank LxWxH = 5x5x3 m, with a surface area of 25 m ² and a net volume of 60 m ³ .

To increase the pH = pH _s + 0.2, the tank is stirred with air at a flow rate of 2 ± 1 m ³ air/m ² h to adjust the pH = 7.4 ± 0,5. Maintaining the air flow rate in the range:

Q = 25 m ² x (2 ± 1 m ³ air/m ² h) = 50+ 25 m ³ air/h

Adjusted by controlling the operation of the blower with Inverter.

Calculating the power of the air compressor: ~ 1

Where:

W _max = weight of Q _max air supply =

R = global gas constant = 8.314

T ₁ = absolute inlet temperature (K) = 293 K (20 °C) n = 0.283 (for air) e = efficiency (typical) for compressor - 0.65

P ₁ = absolute inlet pressure (atm) = 1 atm

P ₂ = absolute outlet pressure (atm) = 1.4 atm

So:

Pre-treated water parameters:

SS < 0.1 mg/L, pH = 7.4±0.5, TOC = 0.4 mg/L and ATZ = 0.7 μg/L

3 ^rd processing stage:

The pH = 7.4±0.5 pre-treated water significantly favours the production of hydroxyl radicals resulting in high removal efficiency of ATZ during the passage through the ozone addition by diffusion PTFE hollow fibre membranes.

- Based on the flow rate Q = 1 ,200 m ³ /day = 50 m ³ /h, membranes with a total surface area of 10 ³ m ² are selected so that the ozone diffusion parameter is 50 L/m ² h.

- The ozone concentration, using an oxygen content of O ₂ > 98%, in the gas phase is 180 g/m ³ .

Thus, based on the ozone requirement of 1 mg/L = 1 g/m ³ , the ozone supply will be:

MO ₃ = 50 m ³ /h x 1 g/m ³ = 50 g/h → QO ₃ = (50 g/h) / (180 g/m ³ ) = 0.28 m ³ /h = 4.6 L/min Parameters at the output of hollow fibre membranes:

CO ₃ = 0.5 mg/L, pH = 7.4+0.5, TOC = 0.2 mg/L and ATZ = 0.18 μg/L

4 ^th processing stage:

Catalyst introduction and treatment in a plug-flow reactor.

In the outgoing water from the hollow fibre membranes, an in-line mixer is used to disperse an expanded zeolite suspension with a grain size of 5 - 50 μm, a flow rate of 6 m ³ /h and a concentration of 7 g/L = 7,000 g/m ³ as a catalyst, so that the catalyst concentration of the reactor is:

C _perlite = [(6 m ³ /h) x (7,000 g/m ³ )] / (56 m ³ /h) = 750 g/m ³

Based on the reaction time requirement of 6 min, the volume of the 10-chamber reactor is:

V _PF = (56 m ³ /h) x (10min/60min/h) = 9.34 m ³

Choosing a depth of 4 m for each reactor chamber means that the total length of the reactor is:

L = (10 chambers) x (4 m/chamber) = 40 m

So the cross-section of each chamber is: E = V _PF /L = 0.234 m ² → L. = W = 0.48 m

Water flow velocity in the plug-flow reactor ensures the fluid suspension of the catalyst:

V = (56 m ³ /h) / (0.234 m ² ) = 240 m/h

> Parameters at the outlet of the plug-flow reactor:

CO ₃ < 0.01 mg/L, pH = 7.4+0.5, TOC = 0.15 mg/L and ATZ = 0.02 μg/L

5 ^th processing stage:

Separation of the catalyst from the treated water and recirculation to the inlet of the plug-flow reactor. For filtration of 50 m ³ /h under suction, based on an optimum filtration flux of 20 L/m ² h = 0.02 m ³ /m ² h, a membrane surface area is required:

Q = Flux x A _Membrane → 50 m ³ /h = (0.02 m ³ /m ² h) x A → A = 2,500 m ²

To cover the inoperability of each membrane for 1 min every 10 min due to backwashing, 6 surface blocks are selected, each of 500 m ² , dimensionally:

LxWxH =1.152x1.298 x 2.763 m

The blocks are lined up in a row, with a distance between them of 0.5 m and 1 m distance from the wall then:

L _system — ( 1 . 1 52 X 6) + (0.5 X 4) + ( 1 X 2)"4 L _system — 1 0.9 m

W _system — 1 .298 + (1 X 2)"4 W _system — 3.3 m

Tank depth = 4 m

A power pump will be used to suck the treated water from the membrane blocks:

P — [p * g * Δh * Q]/0.65

A power pump will be used to backwash each membrane block sequentially for 1 min:

During the backwashing of each membrane block, air with a specific flow rate of 0.25 m ³ /m ² h will be supplied at the same time. So the air supply of the blower will be:

Q _air = 500 m ² x 0.25 m ³ /m ² h = 125 m /h ³

Calculating the power of the air compressor:

Where:

- W = air supply weight = 1

R = global gas constant =

T ₁ = absolute inlet temperature (K) = 293 K (20 °C) n = 0.283 (for air) e = efficiency (typical) for compressor = 0.65

P ₁ = absolute inlet pressure (atm) = 1 atm

P ₂ = absolute outlet pressure (atm) = 1 .5 atm

So:

For the storage (buffer) of a quantity of catalyst suspension to ensure recirculation, a tank LxWxH = 2.5 x 2.5 x 4 m, with a net volume of 20 m ³ , is constructed in communication with the membrane tank. A power stirrer (Figure 1 - number 13) will be installed to maintain the catalyst suspension in homogeneity: I _mixer = 20 m ³ x 100 W/m ³ = 2,000 W = 2 kW

The power of the catalyst recirculation pump will be:

Water parameters after catalyst filtration with ultrafiltration membranes:

CO ₃ < 0.01 mg/L, pH = 7.4±0.5, TOC = 0.15 mg/L and ATZ = 0.01 μg/L

6 ^th processing stage:

A biological filtration process of the treated water, involving the use of an activated carbon bed with a height H = 1 .5 m and a vacuum velocity U = 5 m/h, so that the vacuum time (EBCT) significantly exceeds 10 min, due to low bio-degradation of the ATZ.

- Bed surface: E = (50 m ³ /h)/ (5 m ³ /h) = 10 m ²

- EBCT = H/U = (1 .5 m)/(5 m/h) = 0.3 h → EBCT = 0.2 h x 60 min/h = 12 min

- Volume of activated carbon bed: V _EA = E x H = 16.7 m ² x 1.2 m = 20 m ³

Parameters at the outlet of the biological filtration bed: C0 ₃ = 0 mg/L, pH = 7.4, TOC - 0.12 mg/L and ATZ - 0.01 μg/L < 0,1 μg/L = limit of potability and

BrO ₃ - = 2 μg/L < 10 μg/L = limit of potability