Diohine is a village in Senegal located in the west-central part of Senegal, in the region and department of Fatick, in the commune of Tattaguine. Figure 1 shows its location in the Fatick region. The Diohine borehole, which is the subject of this study, is located in the commune and constitutes the source of water supply for the population.

The assessment of the water characteristics of the Diohine borehole was carried out in two stages: an in situ analysis followed by a laboratory analysis.

The in situ analysis focused on the measurement of physical parameters: temperature, hydrogen potential (pH), electrical conductivity and total dissolved solids (TDS).

A water sample was taken and sent to the laboratory for the analysis of physicochemical parameters and the treatment test process.

The zircon used in this research was supplied by Grande Côte Opérations (GCO), a subsidiary of the Eramet group, specializing in mineral sand mining. GCO operates a deposit located in the dune formations of the Thiès and Louga regions of Senegal, and has two industrial processing units located in Diogo. The extracted ores consist mainly of zircon, ilmenite, rutile, and leucoxene.

As for activated carbon, it was synthesized from the litter of Casuarina equisetifolia (filao) collected on the coastal strip of Mboro. The manufacturing process includes controlled pyrolysis followed by chemical activation using calcium chloride (CaCl₂), before grinding and sieving in the laboratory to obtain a particle size suitable for adsorption application. Figure 2 illustrates in A the image of zircon and in B that of carbon.

Carbonization is a key step in the production of precursor charcoal for activation, aiming to enrich the biomass with fixed carbon and to initiate the formation of porosity. In this study, the previously collected Casuarina equisetifolia (filao) litter ( Figure 3 ) was manually fragmented using metal scissors into small, homogeneous pieces to facilitate carbonization.

The process was carried out using a Nabertherm muffle furnace. The ground material was packaged in a metal pan with a perforated lid to allow controlled evacuation of volatile gases from pyrolysis. The whole was then introduced into the furnace to undergo dry pyrolysis in a controlled atmosphere at 600˚C for 45 minutes.

The muffle furnace, having a limited capacity, required several carbonization cycles in series to accumulate a sufficient quantity of carbon for the subsequent activation and characterization steps. Figure 4 illustrates the visual appearance of the carbon obtained after carbonization of the filao litter called carbonizate at 600˚C.

The carbonized material ( Figure 5(A) ) is washed with distilled water to remove impurities, then dried for 24 hours in an oven at 105˚C. It was then ground in a mortar to obtain two (2) types of particle size: one with a diameter of less than 0.5 mm ( Figure 5(B) ) and the other with a diameter of between 0.5 and 1 mm ( Figure 5(C) ). Figure 5 illustrates the stages of grinding and separation of the two aggregates of the carbonized material of the filao litter obtained in a muffle furnace at 600˚C.

Activation was carried out with calcium chloride (CaCl₂) prepared beforehand in the laboratory.

The charcoal powder, passing through a 0.5 sieve, is poured into a saucepan, then the calcium chloride solution is poured into it while stirring so that all the powder is in contact with the solution. This was done until a pasty mixture was obtained. The saucepan was closed and the mixture was left to stand for 24 hours. Then, the carbonized material was removed from the solution while pressing it to drain the water. The previously activated charcoal was dried in an oven for 24 hours. The activated charcoal was washed with distilled water to remove the activating agent and then dried in an oven for 24 hours at 105˚C. After activation of the carbonized material, the activated charcoal was obtained ( Figure 6 ).

The iodine value is an indicator of the microporosity of an adsorbent; it measures its ability to retain small molecules in aqueous solution. It is expressed in milligrams of iodine adsorbed per gram of material. In this study, a 0.1 N iodine solution was prepared from potassium iodide (KI) and diiodine (I₂), then standardized with a 0.1 N sodium thiosulfate solution. The zircon and activated carbon samples, dried at 105˚C for 24 h, were brought into contact with the iodine solution. After stirring for 5 to 6 min and filtration, a portion of the filtrate was titrated with thiosulfate until discoloration. The amount of iodine adsorbed, calculated by the difference in concentration before and after contact, gives the iodine value of the material. This index allows to compare the adsorption efficiency of zircon and activated carbon and to evaluate their ability to trap small molecules in aqueous phase. The iodine index expressed in (mg/g) was calculated according to Morvan et al. by Equation 1 [19] .

Iodine index $=\frac{\left({\text{C}}_0-\frac{{\text{C}}_{\text{n}}{\text{V}}_{\text{n}}}{2{\text{V}}_{{\text{I}}_2}}\right)\times {\text{M}}_{{\text{I}}_2}\times {\text{V}}_{\text{abs}}}{{\text{m}}_{\text{CA}}}$ (1)

with:

C₀: the concentration of the initial iodine solution (mol∙L⁻¹), C_n: the concentration of the sodium thiosulfate solution (mol∙L⁻¹), V_n: the volume of the sodium thiosulfate solution at equivalence (mL), V_I2: the volume of iodine solution dosed, M _I2: the molecular molar mass of iodine, V_abs: the adsorption volume and m_CA: the mass of activated carbon.

The methylene blue index (MBI) is a parameter used to evaluate the macroporosity of an adsorbent material such as activated carbon or zircon. It is defined as the volume of standard methylene blue solution decolorized by 0.1 g of adsorbent. In this work, a 1200 mg/L stock solution was prepared by dissolving 1200 mg of dye in one liter of distilled water. Adsorption is monitored by UV-Visible spectrophotometry, a quantitative technique based on the Beer-Lambert law that relates absorbance to solution concentration, extinction coefficient, and optical path length. Absorbance measurements are performed for different wavelengths (330 - 690 nm) to determine the maximum absorbance wavelength (λ_max). The latter is used to establish a calibration line, an essential prerequisite for evaluating the adsorption capacity of the materials studied.

To measure the residual concentrations of methylene blue, 100 mg of adsorbent (charcoal or zircon) are brought into contact with 100 mL of dye solution for 20 minutes with stirring. The absorbance values obtained make it possible to determine the residual concentrations (C_mr), then to calculate the apparent adsorption capacity (Q_BM in mg/g) according to Equation 2 [20] :

$Q_{BM}=\frac{(C_{mi}-C_{mr})\times V}{m_{CA}}$ (2)

C_mi: the initial mass concentration (mg/L) of the methylene blue (MB) solution. C_mr: the residual mass concentration (mg/L) of the MB solution. V: the volume of the MB solution (= 100 mL). m_CA: the mass of the adsorbent used (g).

The adsorption or filtration device is illustrated in Figure 7 . In the context of the defluorination tests of hyperfluorinated brackish water, two experimental devices or single-column systems were designed; one filled with activated carbon based on filao bedding ( Figure 7(A) ), the other with zircon ( Figure 7(B) ). Each single-column device consists of a 5 - 10 L raw water reservoir, an inlet valve, a perforated connecting sleeve without a filter, an adsorbent column (zircon or activated carbon), a perforated connecting sleeve with a filter, and a filtration flow control valve as shown in Figure 7 .

After characterizing the raw water, a volume of 500 mL was introduced into the reservoir of each device. The inlet valve was opened, and the filtration flow rate was adjusted to 0.83 mL/s using the shut-off valve. The contact time between the water and the adsorbent was set at 10 minutes. The filtrates obtained were collected in beakers and then stored in refrigerated bottles before being sent to the laboratories for physicochemical analyses.

The analysis of the physicochemical parameters of raw water ( Table 1 ) compared to the standards and limits recommended for drinking water, reveals that the majority of the measured values exceed the accepted thresholds, thus confirming the need for prior treatment before consumption. However, the present study does not aim to directly make the analyzed water potable, but rather to evaluate the relative effectiveness of adsorption on zircon and adsorption on activated carbon based on filao litter for the removal of fluorides.

These results were obtained following analyses carried out with an X-ray fluorescence spectrometer, the Niton XLT900s, at the laboratory of the Applied Nuclear Technology Institute (ITNA) of the Cheikh Anta Diop University of Dakar (UCAD). Table 2 gives the composition of activated carbon and Table 3 that of zircon.

The results obtained highlight notable differences between activated carbon from filao litter and zircon in terms of fluoride removal. Chemical analysis showed that filao litter is particularly rich in calcium (66.5%), aluminum (12.25%) and iron (6.25%), elements known for their ability to interact with fluoride ions by precipitation or adsorption on metal hydroxides. In addition, its low density (0.25 g/L) and its porous structure after activation constitutes an advantage for the development of a large specific surface area, which favors adsorption. These characteristics explain the good performances observed during adsorption tests, consistent with the iodine indices close to those of commercial carbons (838 - 876 mg/g) [19] .

In contrast, zircon has a composition dominated by zirconium (80.83%) and silica (15.32%), with a much higher density (3.16 g/L). Zirconium is known for its high affinity towards fluoride ion through the formation of stable complexes (Zr-F), giving the material remarkable selectivity and efficiency for defluorination [21] . Although its structure is denser and less porous than that of activated carbon, the high zirconium concentration compensates for this limitation by providing many specific active sites.

These results are consistent with those reported in the literature, particularly on activated carbons obtained from bitter almond shell and olive seeds, which also exhibited interesting fluoride adsorption capacities. Similarly, the performance observed with zircon confirms the conclusions of previous studies highlighting the effectiveness of zirconium-based materials in the defluoridation of groundwater [22] [21] .

It therefore appears that activated carbon from filao litter constitutes a versatile and accessible adsorbent, capable of treating several pollutants thanks to its porosity and the diversity of its constituent elements, while zircon, more specific, proves to be particularly effective in the targeted elimination of fluoride. The choice between these two materials therefore depends on the context of use on the one hand, the overall treatment of water quality for activated carbon and on the other hand, the selective and optimized treatment of fluoride for zircon.

The characterization of activated carbon based on filao litter consisted of determining the moisture, volatile matter, ash, fixed carbon, iodine index and methylene blue index. The results obtained during this characterization are presented below. The characterization results are shown in Table 4 .

The results obtained on filao litter ( Table 4 ) are generally in agreement with those reported in the literature [23] . Indeed, the reference study on biomasses also showed low ash contents, low humidity levels and high proportions of volatile matter. This convergence confirms that filao litter, like other biomasses studied, constitutes a precursor suitable for the production of quality activated carbons, presenting both a good adsorption capacity and an interest for water treatment.

The iodine and methylene blue indices of zircon and coal are grouped in Table 5 .

The results show that both zircon and activated carbon prepared from filao litter have iodine values comparable to those of commercial activated carbons [19] [23] , indicating good efficiency in terms of adsorption capacity. However, activated carbon from filao litter develops a slightly higher performance than zircon, due to its porous structure more suited to the adsorption of methylene blue molecules. Zircon, although efficient, could release some mineral constituents into the treated water, which justifies the complementary use of activated carbon downstream.

In Figure 8 , we presented the results on the adsorption spectrum of methylene blue which allowed us to choose the best wavelength for the determinations of the methylene blue indices of coal and zircon.

For different wavelengths, we find that in the adsorption spectrum of methylene blue, the absorbance reaches its maximum at the values of 580, 600, 610, 620, 630 nm. Thus, we chose the wavelength 620 nm for the rest of the work.

In the following Figure 9 , we have illustrated the calibration line obtained during the determination of the methylene blue indices. The equation of the calibration line, representing the absorbance as a function of the concentration of Methylene Blue, with a regression coefficient R² = 0.998, can be considered as a good linear fit. This equation is used, subsequently, for the calculation of the concentration of a given solution.

The following device Figure 10 was adopted for the treatment as written in the methodology.

The results of the analysis of raw water, treated water and standards are presented in Table 6 .

Analysis of the results obtained highlights an initial fluoride concentration in the raw water from the Diohine borehole of 5.74 mg/L, a value well above the standard recommended by the WHO set at 1.5 mg/L. This high content confirms the need for defluoridation treatment before any consumption [24] . After treatment, two distinct behaviors are observed depending on the adsorbent used. Zircon allows a significant reduction in the fluoride concentration, bringing it down to 1.84 mg/L, which corresponds to a reduction of approximately 67.9%. Although this value remains slightly above the limit set by the WHO, it demonstrates the high efficiency of this material and its strong affinity for fluoride ions [22] [25] [26] . In comparison, activated carbon made from filao litter reduced the concentration to 2.29 mg/L, a reduction of 60.1%. This significant result, however, remains lower than that obtained with zircon and thus reflects a more limited effectiveness with respect to fluoride.

These observations indicate that zircon performs better than filao activated carbon for fluoride removal. However, neither adsorbent can achieve the recommended potability standards. On a practical level, activated carbon has the advantage of being sourced locally and is very accessible, making it an attractive alternative for decentralized and sustainable treatment. Zircon, on the other hand, appears to perform better but could pose economic or supply constraints. In this sense, process optimization, particularly through combined or sequential treatments (zircon-carbon, or coupling with other adsorbents such as activated alumina), could make it possible to achieve fluoride concentrations that comply with international standards.