Paraquat (C₁₂H₁₄Cl₂N₂, methyl-violen or 1,1’-dimethyl-4,4’-bipyridinium according to IUPAC, from Sigma-Aldrich), TiO₂ P25 (average size 20 nm, purity 97%, surface area 50 m²∙g⁻¹ and 80% anatase, 20% rutile, Evonik industries) were used as received. Ethanol anhydride 99.8% and titanium tetra isopropoxide (TTIP, C₁₂H₂₈O₄Ti, 97%) were purchased from respectively Fluka and Sigma-Aldrich. Distilled water was used to prepare all the solutions.

TiO₂ suspension was prepared by adding 10 g of Titanium dioxide P25 in 200 mL of ethanol used as solvent. The resulting concentration of TiO₂ was 50 g/L. Then, 4 mL of TTIP was added in the suspension and used as binder to join the TiO₂ nanoparticles together. TiO₂ slurry is stirred for 24 h with a magnetic stirrer.

The β-SiC alveolar foams were synthesized and supplied by SICAT Company (Willstätt, Germany) [18] . The β-SiC foams were presented in form of parallelepipedic substrates with a dimension of 9.5 cm (length) × 6 cm (width) × 1 cm (thickness), each having a weight of 20 g.

Each piece of β-SiC foam was completely immersed in the TiO₂ suspension for three min at 5 rpm in order to immobilize the TiO₂ photocatalysts on β-SiC. This action was repeated five times. Then the TiO₂/β-SiC foams were dried at room temperature for 20 min. The TiO₂/β-SiC foams were placed in an oven at 110˚C overnight to evaporate the residual solvent. Subsequently, TiO₂/β-SiC foams were brought to the furnace at 450˚C during 2 h at a rise rate of 5˚C∙min⁻¹. The material was calcined to remove all traces of organic matter and to increase the TiO₂ grip on the foam. The average wt % of TiO₂ per foam was 7.5%. Figure 1 shows the material photocatalytic β-SiC before and after coating.

Scanning Electron Microscopy (SEM) images of TiO₂ P25 films onto the β-SiC foams were recorded using a JEOL scanning electron microscope (JSM-6390) at an accelerating voltage of 25 kV.

Photocatalytic treatments were carried out in a continuous flow-through photoreactor made of polypropylene material with a dimension of 20 cm (length) × 7.5 cm (width) × 2.5 cm (depth). One or two TiO₂/β-SiC foams were placed inside the photoreactor to perform the photocatalytic tests. The photoreactor was covered with a quartz plate to filter the UV radiation. Two UV-A lamps (Philips 18 W, China) were placed horizontally 2 cm above the reactor to illuminate the TiO₂/β-SiC photocatalysts. Irradiation wavelength was around 368 ± 20 nm with an irradiance of about 60 W∙m⁻² using a UV-A radiometer (spectral range: 315 - 400 nm; HD 9021; Delta OHM; Italy). Figure 2 shows the photocatalytic experimental device. Paraquat’s solution was obtained with distilled water. The treated effluent left the reactor through overflow mode. The effluent containing the pollutant (paraquat herbicide) was stored in a 1 litre tank immersed in an ice bath (about 0˚C) to avoid evaporation of the solution in the reactor. The effluent circulated through the photocatalytic system in a single pass using a peristaltic pump. The inlet and outlet flow rates were quite the same and ranged between 26 and 87 mL∙min⁻¹ with a retention time varying from 4 to 12 min about. The feed solution contained 10 to 30 ppm of paraquat herbicide.

Preliminary tests were carried out to compare the direct UV-A photolysis (direct irradiation of the solution without TiO₂/β-SiC), the paraquat adsorption on

TiO₂/β-SiC (without UV-A) and the heterogeneous photocatalysis (UV-A and TiO₂/β-SiC) processes. The experimental conditions are: flow (26 mL∙min⁻¹); C_o = 10 ppm; V (PQ) = 1 L, pH = 6.7 and two TiO₂/β-SiC. Photocatalytic device was conducted at room temperature of the laboratory at about 20˚C.

Paraquat was chosen as a target pollutant [8] to evaluate the photocatalytic performances of the photocatalytic materials. During the irradiation procedure, 2 mL of the solutions were sampled at regular time intervals. The remaining concentration of paraquat was followed by a LIBRA S12 UV-Vis spectrophotometer at 257 nm (maximum absorption wavelength). The pH of the aqueous solutions of the paraquat was adjusted by HCl (0.1 M) and NaOH (0.1 M) aqueous solutions. pH values were determined by a pH-meter (labChem-CP) equipped with junction pH electrode (Model IJ44C). The evolution of TOC during the degradation kinetics of paraquat was followed by a SHIMADZU TOC-L (Total Organic Carbon Analyzer, Japan). Considering that the degradation follows a pseudo first order kinetics, the apparent kinetic constant k_app was calculated according to the following equations:

r = − d c d t = k a p p C (1)

ln ( c c 0 ) = − k a p p ⋅ t (2)

With

r: reaction rate;

C: residual concentration of the pollutant;

C_o: initial concentration of the pollutant;

k_app: apparent kinetic constant and

t: UV irradiation time.

Five subsequent experiments were carried out using the same sample of TiO₂/ β-SiC photocatalysts. Between each experiment, the TiO₂/β-SiC foams were treated at 450˚C for 2 hours to remove all traces of organic matter. The evaporation temperature of paraquat is about 300˚C. The operating conditions of the continuous aging study were as follows: pump flow rate (26 mL∙min⁻¹), volume of paraquat to be treated (2 L), concentration of paraquat (10 ppm), two TiO₂/SiC materials, duration of the experiment (70 min).

Figure 3 shows the morphologies of the TiO₂ P25 film surfaces with and without TTIP. The analyses showed the presence of micro and macro cracks at the surface of the film prepared without the addition of the TTIP in the suspension (Figure 3(a)). These results were observed by some authors [19] . The size of the cracks varied from 3 to 7 μm. These cracks were due to the drying of the material after the dip-coating and during the heat treatment at 450˚C. However, no cracks are observed on the film surface when we added TTIP in the suspension (Figure 3(b)). This homogeneity of the film was due to the presence of the TTIP that plays the role of binder. This organic polymer made it possible to solder the molecules of the TiO₂ P25 nanoparticles, thus avoiding cracks during temperature drying and calcination in the furnace.

Photocatalytic study was carried out to compare the removal efficiency of paraquat by direct photolysis, adsorption and photocatalysis processes. Figure 4

shows that the direct photolysis process remains ineffective, only 1.50% of paraquat and 0.17% of TOC was eliminated after 70 min of irradiation. On the other hand, adsorption and photocatalytic processes remain effective with paraquat abatement rates of 36.65% and 42.68% respectively (Figure 4(a)) and 22.34% and 27.45% of TOC eliminated (Figure 4(b)). A competition between the two processes occured. However, the photocatalytic process remained as effective as adsorption process [20] . The difference between the two processes can be explained by the fact that during the photocatalytic process, OH˚ radicals were released to destroy some Paraquat molecules adsorbed on the surface of TiO₂ P25.

Beyond 20 min of irradiation, we observed a limit of the diffusion of the paraquat molecules to the surface of photocatalyst. This was due to the plug flow reactor used during these experiments. Kouamé et al. [21] showed that the foam played the role of mixer. Due to the size of the foams, some molecules of the paraquat remain at the bottom of the unlit reactor.

The paraquat kinetics of disappearance after reuse of the TiO₂/β-SiC catalytic material is studied. The results showed that the photocatalytic activity of TiO₂/β-SiC remained constant after five cycles of reuse of photocatalyst [26] . Their photocatalytic properties remained unchanged. This shows that the remaining degradation by-products at the material surface are eliminated after the heat treatment between two tests.

Table 1 shows the percentages of paraquat removal (43.16 ± 1.00)%, TOC

removal (27.13 ± 1.00)% and the apparent rate constant (0.0656 ± 0.0010) min⁻¹ during the five reuse tests. These results are very encouraging because the study of the aging of the photocatalytic material remains a necessity in the step of application of the heterogeneous photocatalysis on a large-scale application.