Photocatalytic Ozonation Performance In Landfill Leachate Treatment.

This work aims to identify the best treatment condition for removal of chemical oxigen demand (COD) and increase of biodegradability in an advanced oxidative process of photocatalytic ozonation with titanium dioxide (TiO2), applied to the remediation of the leachate produced at the municipal landfill of Campo Mourão, Paraná. The experiment was carried out using a photocatalytic chamber and an ozonation system based on the corona effect. A statistical model of the central composite rotatable design (CCRD) was elaborated and three variables (pH, TiO2 dose and airflow) were analyzed, to define the optimum condition that allows the highest perfomance of the treatment. The statistical model was valid for the data. The best condition identified was: pH = 3.3 TiO2 dose = 0.012 g and flow = 9.0 L.min-1 (8.79 g, O3.min-1). The removal of filtrate COD was approximately 30% and increase in biodegradability (BOD/COD) = 0.59 in the optimum condition. Factors such as the complex composition of the leachate and treatment time were considered as possible limitants for better results. The treatment did not allow to reach the permissible limits for disposal of the effluent at the environment, in Brazil.


INTRODUCTION
With the increase of the population and the improvement of the quality of life, consumption is boosted by technological development, leading to the generation of a large amount of waste and effluents, such as the leachate. The main sources of this effluent are from the water infiltrated in the soil, the humidity that is already present in the residues and the humidity from the sludge, when present (Guidolini et al. 2005, Tavares 2011.
The leachate has a specific physicochemical composition, according to the original decomposed material, containing components such as dissolved and solubilized organic matter; nutrients; intermediate products of anaerobic digestion of the waste, as volatile organic acids; and chemicals, that include heavy metals like cadmium (Cd) and mercury (Hg), for example. This composition origins an effluent considered extremely damaging to the environment, being responsible for the contamination of soil, surface water bodies and groundwater, on site and in the whole surrounding region, due its toxic caracter. Beside that, the leachate carries pathogenic microorganisms that are harmful to humans, such as Enterococcus, P. aeruginosa, heterotrophic bacteria, fungi and even thermotolerant coliforms (E. coli) (Tavares 2011, Castilhos Junior et al. 2006, Peixoto 2008, Silva et al. 2011, Jucá & Sobrinho 2012. For the treatment of this type of effluent, there is a great diversity of systems, conventionally applied for the degradation and stabilization of pollutants, to promote their safe disposal at the environment. However, Renou et al. (2008) points out that most of the treatments are inefficient to reach the parameters established by the current environmental legislation.
For such, advanced oxidative processes (AOP's) may be a viable alternative. The basic principle of this form of treatment are the hydroxyl free radicals (•OH), with high oxidative power. These, degrades the pollutant molecules decomposing them into stabilized and less harmful substances, leading to the destruction of organic components, such as carboxylic acids and transforming phenolic or chlorinated aromatic derivatives, into more harmless compounds, such as carbon dioxide (CO 2 ), mineral salts and water (H 2 O) (Mehrjouei et al. 2015).
Among the AOP's employed, according to Guidolini et al. (2005), the use of ozonation is common, mainly in the case of leachate, with pollutants in high concentrations, that exceed the limits of the physicochemical parameters accepted by the environmental agencies, being capable of promoting the reduction of chemical oxygen demand (COD), biochemical oxygen demand (BOD), colour, turbidity and odour.
The ozonation process is already mentioned in the literature as a promising technology, with numerous real-scale applications for the treatment of industrial wastewater, using a strong oxidizing agent, ozone (O 3 ). Thus, it can be applied in the treatment of the compounds present in the landfill leachate, not forming an intrinsic source of pollution (Mahmoud & Freire 2007, Assalin & Durán 2007. Also, the treatment performed with O 3 can be intensified in complement with other forms of treatment, to stimulate the formation of radicals that cause a greater degradation. For this purpose, the use of UV radiation can increase the rate production of •OH radicals and bring more satisfactory results (Kammradt 2004).
From the AOP's currently used with UV radiation, heterogeneous photocatalysis with titanium dioxide (UV/TiO 2 ) demonstrated as a highly efficient technique. It acts in the total mineralization and destruction of contaminating pollutants present in water and effluents, being economical and environmentally beneficial, also removing microorganisms without generating dangerous by-products. Thus, with the use of AOP's applied simultaneously, it is carried out the transformation of toxic or recalcitrant compounds into biodegradable substances or easily removed by other simple processes, like biological or physico-chemical (Szczepanik 2017, Ahmad et al. 2016, Mehrjouei et al. 2015, Kammradt 2004. Besides that, the use of O 3 and UV/TiO 2 , individually, has limitations on its effectiveness in the degradation of organic pollutants, because UV/TiO 2 represents a slow process due to the low oxidation rate. Similarly, the oxidation process by molecular O 3 , due to its high selectivity, only leads to the partial degradation of organic compounds, generating carboxylic acids and final products that can not be completely mineralized. These limitations end up making such processes economically unviable for treatment of effluents. For this, its simultaneous application resulting in the socalled photocatalytic ozonation, may compose a strong complementation of the processes because of the increased rate of degradation, leading to a more complete and efficient treatment (Mecha et al. 2016).
This process is considered a topic of recent research, and it is suggested in the literature that, the free electrons generated of TiO 2 , in its surface layer, interact with O 3  With the induction of the photocatalytic process, the pairs (e -/h + ) are generated, from which recombination may occur, causing negative effects of reduction and oxidation on the catalyst surface, reducing its interactions and performance. Because the electronic affinity of O 3 is higher than O 2 , it is possible to promote photocatalytic reactions in the presence of O 3 , more effectively than in the UV/ TiO 2 process alone, reducing these effects. Also, H 2 O 2 formation is favoured through O 3 , which also reacts with the catalyst surface to form •OH radicals (Gomes et al. 2017, Mecha et al. 2016, Mehrjoue et al. 2015, Mena et al. 2012 , O 3 , h + ) in the solution for degradation of a greater amount and diversity of pollutants. For this reason, such reconciliation has recently been studied in the degradation of different emerging contaminants (Gomes et al. 2017, Mecha et al. 2016, Mehrjoue et al. 2015, Mena et al. 2012. For Mecha et al. (2016) the process is effective in the treatment and detoxification of effluents to reach parameters necessary for its disposal, leading to a higher level of mineralization than that obtained by the sum of the processes used individually. Besides, Mena et al. (2012) evaluated the costs of photocatalytic ozonation in comparison to the individual processes, which showed that the operational costs were always higher than the simultaneous processes, making the photocatalytic ozonation not only attractive in terms of reaction rate, but also by the economy that is obtained.
There are not too many studies of the application of photocatalytic ozonation into landfill leachate treatment, especially in Brazil, where none is reported according to a survey carried out in the Capes journal, Science Direct and Scielo. Also, to evaluate the process in the remediation of leachate, it's necessary to determinate the possibilities of the treatment and the best conditions that allows to reach the maximum performance and the limits of disposal present in the legislation.
This work aims to identify the best treatment condition for removal of chemical oxygen demand (COD) and increase of biodegradability in an advanced oxidative process of photocatalytic ozonation with titanium dioxide (TiO 2 ), applied to the remediation of the leachate produced at the municipal landfill of Campo Mourão, Paraná.

Leachate sample collecting
The leachate was collected at the municipal landfill of Campo Mourão. This was done with 50 L polyethene gallons at the entrance of the first treatment lagoon (anaerobic), being hermetically sealed and transported to the laboratory, being allocated in a horizontal refrigerator, with two compartments, at -4 °C, to minimize the biological and chemical reactions.

Analytical methods
To characterize the collected raw leachate, the following parameters were evaluated: pH, temperature, turbidity, crude COD, filtrate COD (COD-F), BOD 5 20 and biodegradability (BOD/ COD). These parameters were obtained through specific equipment and different methodologies, To simplify the methods used, the following Table I is presented, establishing the methodology, the equipment and materials used in the determination of physico-chemical parameters.
From those parameters, COD-F and the relation BOD/COD were used to evaluate the treatment. Thus, for those parameters the measure was made before and after the treatment. In case of COD-F, as established by Amr & Aziz (2012), the following equation was used to determinate the efficiency removal provided by the treatment:

Removal (%) = [(Ci -Cf)/Ci] * 100
(1) In equation (1) Ci and Cf correspond to the initial and final concentration of the parameter monitored, in the sample analyzed. The results of those determinations and the biodegradability measure in each sample were used to create the statistic model proposed. A discussion based in the scientific literature and a comparison with the current environmental legislation was made, to understand better the process and verify if it was possible to meet the levels for the disposal of the effluent "in natura".

Photocatalytic ozonation system
The system used to treat the leachate collected, had two distinct structures: heterogeneous photocatalysis and ozonation. The first structure corresponds to an irradiation chamber made of medium density fiberboard (MDF), with dimensions of approximately 50x33x32 (height, width and depth). On the upper external part of the chamber there is a switch, for activating the lamp, and behind it, a 250 W reactor, powered by an extension cord attached to a 220 V power plug. Inside, a vapour lamp of high pressure Table I. Methods, equipment and/or materials used to measure physicochemical parameters to characterize crude leachate and monitor photocatalytic ozonation process.

Method -equipment and materials
Hydrogen potential (pH) Direct measure -pH meter pH-221 Lutron; mercury with 250 W of power is installed in the upper part, connected to the reactor, without the external bulb, removed manually to avoid the effect of UV filter caused by the glass, as established by Schneider et al. (2014). Under the lamp is placed a 600 mL glass beaker with a magnetic bar, the catalyst (TiO 2 ) and 250 mL of leachate, inserted into a plastic container upon a magnetic stirrer. The plastic vessel was used for insertion of ice cubes, to reduce the effect of increasing temperature of the solution, caused by irradiation. Together with this, on the side surface of the chamber, there is an entrance, in which is inserted a hose, relating to the second structure.
The second structure is composed of an air compressor, connected to a flowmeter with a capacity of 15 L.min -1 , used to regulate the air flow, coupled to a universal support, connected by a hose to the ozone reactor. The reactor consists of a wooden plate with two cylindrical coaxial tubes, of which in one of the tubes the entrance of the air generated by the compressor takes place. This air enters one side of the tube, passing through the inside of it and following to the hose that passes through the side entrance of the photocatalytic reactor, ending in the beaker with leachate.
The generation of O 3 occurs through the electric discharge caused by an electronic circuit located near the tube, which causes the corona effect. This electronic circuit is connected to a source that maintains constant current at 12.5 V and a second power supply that provides 45 W to cause O 3 generation. The system also has fans coolers for cooling effect. Given the excessive heating of the air compressor, a small fan was also directed towards it, to mitigate this effect.

Measure of O 3 production in the system
Because the air flow of the ozonation system compose an independent variable in the CCRD, a determination of the amount of O 3 produced in the treatment was carried out using the iodometric titration method, established by the Standard Methods for the Examination of Water and Wastewater (APHA 2012), also used by authors such as Bassani (2010). Based on this methodology, the production of O 3 obtained with the atmospheric airflow applied by the compressor in the proposed system, can be determined by the equation (2) presented: In the equation: Po 3 = production of O 3 (g O 3 .min -1 ); Ntio = normality of thiosulfate; Vtio = volume of thiosulfate obtained in the titration (mL); VKI = volume of potassium iodide (KI) solution (mL); Vam = sample volume; t = bubbling time. This procedure was performed in duplicate, for some outflows foreseen in the CCRD, presented later. In the case of a very large divergence between the volumes obtained in the method, the process was repeated.
Considering the levels established in the CCRD, the results obtained with the iodometric method are presented below (Table II).
Based on the results obtained, a graph was created to observe the O 3 production behaviour and to obtain the curve equation, allowing the quantification of this production in average and minimum values. Such a question was necessary because the iodometric method is an indirect measure, from which the turning point by titration becomes imperceptible in some values or even the variation is very small between one point and another. Thus, with the elaboration of the curve it was possible to perform reagent saving and the determination of an optimum production point on flow 9.0 L.min -1 (8.79 g.min -1 ).
In the modelling of O 3 production values, the trend line closest to the data behaviour was the polynomial of the second order, with R² of 0.79. The results obtained as presented (Table  II), were taken into account in the proposed experimental design, in order to contribute for the selection of a flow range corresponding to a viable average production of O 3 , in theory, that would lead to a better degradation of the effluent.

Statistic analyses
To perform a statistical analysis of the data, a factorial design of the type central composite rotatable design (CCRD) was developed. The definition of the independent variables (factors) of pH, TiO 2 dose and ozone dose, as well as their respective levels were established through the revised literature. To compose a process of greater viability, considering also the literature and the high recalcitrance of the effluent, it was evaluated as a response of this planning: the efficiency of COD-F (%) removal and the increase of biodegradability through the BOD/COD ratio.
For the creation and analysis of the design, the software Statistica 7.0 was used. The planning adopted in this process, having the three independent variables, compose a planning CCRD 2³, this is, corresponding to a total of 8 experiments, 6 axial points and 3 central points, equal to a total of 17 tests, to determinate the optimal operating condition for photocatalytic ozonation process. To be a complete planning 2³, the axial points (α) define the coded levels -1,68 and +1,68. As follow, it's presented the independent variables, the levels adopted and the codification of those in the matrix of the experiments design (Table III and Table IV). Worth highlighting that in the CCRD, the variable ozone dose is presented as the variation supplied by the compressor, from which was possible to control in a pratic manner, by the flowmeter.
With all the data from the 17 tests, statistical analyzes were performed to obtain the effects of variables, analysis of variance (ANOVA), to verify the lack of adjustment, plotting of responses and especially, the desirability function of the model for visualization of the best condition for removal of COD-F (%) and increase of biodegradability (BOD/COD).
This optimal condition was repeated experimentally to validate the model prediction, and the complementary parameters were analyzed together to obtain the system response, in this condition, in a better way. Descriptive Table III. Variables and defined levels for the CCRD (2³) referring to the factors: pH; TiO 2 dose (g) and air flow (L.min -1 ).

Experimental procedure
The procedure adopted to apply the treatment of photocatalytic ozonation in the leachate consists in the following series of steps: 1) The monitoring parameters are measured in the raw leachate sample (250 mL); 2) The photocatalytic reactor cover is partially closed and the lamp is switched on and left on for 1 min for irradiation stabilization; 3) The air compressor is activated and the flowmeter regulated at the specific flow rate to be tested; 4) Both power supplies are switched on and 15 s are waited for O 3 generation stabilization, the compressor is then switched off briefly; 5) The pH of the raw sample is corrected by titration with 0.01 and 1 N sulfuric acid (H 2 SO 4 ) and sodium hydroxide (NaOH), measured with the pH meter until the adjustment of the value in the condition to be tested; 6) In the beaker with the solution, a magnetic bar is inserted, and it is placed inside the photocatalytic chamber, in the plastic container, on the magnetic stirrer; 7) The outlet of the hose from the ozonation reactor is position the effluent; 8) The TiO 2 catalyst, previously weighed in analytical balance, according to the condition to be tested, in a 25 mL beaker is manually inserted into the solution; 9) Ice cubes are placed inside the plastic container, the magnetic stirrer is started and the cover of the photocatalytic reactor is closed; 10) Soon after (about 5 s), simultaneously the compressor and the lamp are activated,  initiating the treatment process, measured during its course by a stopwatch. The treatment applied had a maximum duration of 60 min, in which after its completion, the treated sample was withdrawn to re-measure the physical-chemical monitoring parameters. The commercial chemical catalyst tested was Degussa® TiO 2 P-25 composed of: 80% anthestase and 20% rutile, as a white powder, odourless, with 21 nm particles.
For the photocatalytic treatment, the UV lamp was selected because of its medium pressure, emitting photons of length 253.7 nm and low operating temperature, besides 90% of the emission being foreseen within the minimum absorption range by the catalyst used. Its use provides a more compact and lower cost system (Ilha 2012, Teixeira & Jardim 2004.

Leachate characterization
The characterization of the leachate related to the parameters analyzed is presented below (Table V).
Taking into account the time of operation of the municipal landfill from Campo Mourão, 14 years (according to the operating license), as well as the considerations made by Souto (2009) and Renou et al. (2008), its possible to classify the leachate collected as being in the phase of methanogenic degradation, that is, "mature" leachate. This fact can be highlighted by the values of parameters such as pH, COD and BOD 5 20 previously presented. The BOD/COD ratio, between 0.22 and 0.25, confirms that the effluent is in a more advanced state of degradation. According to Kurniawan et al. (2006a) and Renou et al. (2008), this BOD/COD ratio up to 0.25, between 0.1 and 0.5, is more common in medium/intermediate age leachates, and that a BOD/COD ratio of less than 0.1 represents a more specifically leachate, in the final phase of stabilization. It is possible to consider that the characterized effluent is in a transition phase from the initial methanogen to stabilized. For Renou et al. (2008), the BOD/COD ratio decreases from 0.70 to 0.04 as time passes in the landfills, because large recalcitrant molecules are released by the residues, which consequently renders the landfill leachate "mature" characteristic, presenting low values of BOD/COD, for example.

Process optimization
Based on the experimental design, it was possible to obtain removal efficiencies for COD-F (%) ranging from 7.85 to 28.95% and increase of biodegradability from 0.36 to 0.64 for the leachate treated. The best result, for both response variables, corresponded to test 10, with a maximum reduction of 28.95% COD-F (%) and increase of the BOD/COD ratio of the effluent to 0.64. Obviously, the increase in biodegradability is related to a greater reduction of COD considering the relation used, however, it is worth mentioning that the behaviour of BOD 5 20 was also determinant in this case, justifying tests in which the increase of biodegradability was superior, but the removal of COD-F (%) lower. The results obtained for the variables studied in each of the 17 tests that compose the CCRD (2³), elaborated for the study of the photocatalytic ozonation process are presented below (Table VI).
To understand the influence of the variables on these results, the effects of the pH, dose (g) and flow (L.min -1 ) variables and their interactions, obtained through software Statistica 7.0, are shown below, for COD-F (%) removal and increase of biodegradability (BOD/ COD), at a significance level of 95% (p-value <0.05) in Table VII.
On Table VII, the significant effects (p-value <0.05) are shown in bold. To obtain a more valid fit, most of the insignificant effects on the model were ignored for both COD-F (%) removal and biodegradability increase. Some of the not significant effects, with p-value close or superior from 0.05, and also with the similar linear or quadratic effect significant, were mantained, to keep the high adjustment of the model. For COD-F (%) removal those effects were: pH (L) and air flow (L.min -1 ) (Q); For biodegradability increase the effects were: pH (L) and TiO 2 dose (g) (L). By the obtained values, most of the factors present a synergistic effect, indicating that as the increase of these factors cause the increase in removal of COD-F (%) and biodegradability. The only significant antagonistic effect, that is to say, opposite to better results, was the interaction air flow versus TiO 2 dose (L), indicating that the interaction between these variables is negative, so even if a superior air flow is used, the negative effect of the catalyst dosage may prevail, leading to a decrease in COD-F (%) removal (Barros Neto et al. 2001). For biodegradability increase the antagonistic effect obtained was pH (L), but not significant in the model.
It is worth mentioning in this case, that the interaction air flow x TiO 2 dose (L) is considered synergistic in biodegradability increase, inverse to that obtained in the removal of COD-F (%), probably because it considers the BOD 5 20 data. It can also be inferred that the TiO 2 dosage exerts little influence on the biodegradability results.
To establish the validity of the statistical model of the presented data, based on the significant effects, a variance analysis (ANOVA) was performed considering the pure error, allowing to detect if the lack of fit was significant in the model adopted (Table VIII).
For both cases, the p-value less than 0.05 indicates that the regression factors, or at least one of them has a significant impact on the mean response obtained. Also, this value also establishes the probability of Fcalculated (Fcal) being close to the Ftable (Ftab), to be lower than 0.05, considering a 95% confidence interval, indicating that the model is valid. The Fcal superior to Ftab presents, in of regression, that the mathematical model can be considered predictive for the data, in both cases. The p-value for lack of adjustment was greater than 0.05, so the lack of adjustment is not significant, which is also portrayed by the Fcal lower than the Ftab, in this case. With the estimation of effects and analysis of ANOVA, it was possible to construct a predictive statistical model for the COD-F (%) data with a R² of 0.88 and R-adjusted of 0.80, indicating that 80% of the variation of the presented data can be explained by this. Likewise, in the case of increased biodegradability (BOD/COD), the R² was 0.89 and an adjusted R-value of 0.84, indicating that the model can cover 84% of the variation in the data obtained by the treatment.
As shown in Figure 1, the model demonstrates that the use of a high O 3 flow, corresponding to 9.0 L.min -1 (8.79 g O 3 .min -1 ) favours higher removals, together with the association of a higher TiO 2 dosage (0.01 and 0.012 g) (a, b). The pH in this case is presented as a more flexible factor, that is, in both lower value (1.6) and higher (3.3), the higher removals in COD-F (%) can be obtained. For biodegradability increase, in the case of the TiO 2 dosage and the O 3 flow (d,e), values taken as central, in both cases, provided the best results based on their location near the darker region on the surfaces. In these cases, pH is a determinant factor for this, so that its variation in 1.6 and 3 would also lead to a higher performance. This can be seen, in both cases, considering that the pH having only significant quadratic effects. The model also reflects for the factors such as catalyst dose and airflow, that the use of a larger quantity would lead to better results. It is noteworthy that in this cases, pH prevailed as a determining factor to obtain a superior performance according to the presented surfaces. Thus, considering the versatility of the photocatalytic ozonation system as regards the pH factor, as the model itself showed, changes could be made that would bring better performance, considering the PCZ of the catalyst.
Based on these considerations, obtain the optimal condition of the model, with a focus mainly on the increase of biodegradability,it is presented the result of the desirability function of the statistical model, according to Figure 2.
In this function, it was chosen to give a greater weight to the increase of the BOD/ COD ratio. As indicated, based on the statistical model previously presented, the best condition would correspond to the maximum levels defined in the CCRD for all factors. Thus, the Considering the obtained employing the function, a last test was carried out examining which were the answers of the system regarding to the pre-established optimal condition. The results predicted by the model and those obtained experimentally are presented below (Table IX).
By the response of the desirability function, it is possible to consider that the model was shown to be predictive. The pre-established efficiency of 29% according to the presented conditions, resulted in the practice, in a response of 30.13% of removal of COD-F (%), value even more positive than that established by the model. The biodegradability of the treated leachate resulted in a predicted, but lower value of BOD/COD = 0.59, post-treatment.
Sense high R² values were obtained: 0.88 and 0.89, it is expected that the model corresponds to the reality verified. Despite of this, it is important to emphasize that the high R-adjusted is more significant for analysis of the fit of the model, because it considers the degrees of freedom, that must be higher than the number of coefficients of the model, composing a more statistically faithful coefficient. The R-adjusted values were also high, 0.80 and 0.84, confirming that the model is predictive for the data and range studied. The variation obtained occurs because the model predicts only 80% and 84% of COD-F (%) removal and increase of biodegradability (BOD/COD) data.

COD-F removal and increase of biodegradability discussion
It can be seen from Table VI that the treatment did not allow for high COD-F reductions, achieving maximum efficiency of 28.95% in test 10 and a minimum of 7.85% in the test 11. Even test of validation, with maximum levels of all the variables studied, corresponding to the best condition, resulted in a maximum of 30.13%, as previsouly presented.
Considering the scientific literature, high values were expected in both heterogeneous photocatalysis and ozonation, applied individually, demonstrate high performances in the COD removal. Despite this, in these studies generally higher treatment times are noticed, which was not applied in this work, due to the limitations of the ozonization system used, because of the heating of the compressor.
As for BOD 5 20 , all the tests carried out made it possible to increase this parameter in the raw leachate, being more evident in tests 9 and 10. This associated with the reduction of COD-F made it possible to increase the biodegradability of the effluent from a minimum of 0.16 (tests 3 and 4), indicating a need for more advanced treatment, up to a minimum ratio of 0.36 (in test 3) and a maximum of 0,61 and 0,64 (tests 9 and 10), high, considering the recalcitrance of the leachate.
According to Kurniawan et al. (2006b), for a BOD/COD ratio of 0.3 or higher, the biological treatment is more indicated because the organic compounds in the leachate have already been transformed into more biodegradable compounds. Chemlal et al. (2014) states that in a BOD/COD ratio of 0.6, a complete degradation of the leachate occurs naturally through the action of microorganisms.Taking into account the data in Table VI, the following descriptive statistics are presented (Table X).
Based on the previous  (2004), which associated this effect with the transformation of recalcitrant organic compounds into biodegradable intermediates. The authors noted an increase of up to 265% in the values of BOD 5 20 in the leachate by promoting ozonization, including in other studies in which the same effect was observed. It is not considered that the low removal of COD-F is associated with the variation of the pH range, even if small, taking into account also that only its quadratic interaction was significant, according to the CCRD. As discussed by Cho et al. (2002) the photocatalysis process should be conducted at a pH range lower than 4.5, considering that the extraction of alkalinity and bicarbonate ions present in the leachate, since together with inorganic carbons, can compete with organic materials and cause scavenging effects on photogenerated gaps. The authors obtained removals in COD of 60 to 80% in the conditions of pH 2.0 and 4.0, after 12 h of treatment with 3.0 g.L -1 of TiO 2 .
Besides, under acidic conditions below the PCZ of the catalyst, the attractive forces between the surface of the TiO 2 , from which it is positively charged, producing •OH radicals and negatively charged materials such as Cl -, PO 4 3-, SO 4 2-, NO 2 -, among others, favours its adsorption, decreasing the photocatalytic activity, hindering the production of the radicals and preventing other compounds from approaching the active sites in the catalyst. Especially the Clion, in addition to the PO 4 3and SO 4 2ions, can react with the radicals or the generated h + gaps, forming radicals with lower oxidative power, resulting in a lower rate of degradation (Cho et al. 2002, Ramdhani 2018. This may have been one of the factors due to the low efficiencies obtained. Cho et al. (2004) obtained the reduction of an initial concentration of 706 mg.L -1 to the maximum of 315 mg O 2 .L -1 , about 52% removal efficiency, after 5 h of treatment with 3,0 g.L -1 of TiO 2 of which, for the authors, did not show an increase after this period. The authors considered that complete photocatalytic mineralization in the case of leachate is difficult and that the catalyst deactivation process is evident because of the adsorption of non-photodegradable materials, remaining in the solution after treatment, such as siloxane compounds. These compose inactive species that are deposited on the surface of the catalyst and block the adsorption of O 2 , for example, directly injected with ozonation, in the case of photocalytic ozonation treatment.
As for the dosage of the applied catalyst, it is worth noting that it was selected according to the preliminary tests, in a range that made possible the removal of COD-F (%). Also, Wiszniowski et al. (2003) presented that the studies establish longer treatment periods to obtain better results. Based on the current environmental legislation (CONAMA 2011 andCEMA 2009), none of the analyzed parameters, would hypothetically allow the disposal of treated effluent in the environment, requiring complementary treatment.