Ferrous chloride (FeCl₂·4H₂O), stannous chloride (SnCl₂·2H₂O), Sodium hydroxide (NaOH), Hydrochloric acid (HCl), iodine (I₂), starch, sodium thiosulfate (Na₂S₂O₃), glacial acetic acid (CH₃COOH), acetone, and methylene blue dye (C₁₆H₁₈N₃SCl, 3H₂O) were used of analytical grade. A methylene blue stock solution (10 mmol/L) (Mw = 373.9 g/mol) was prepared and used in adsorption studies.

Papaya seeds were selected from agricultural waste from Mukalla/Hadhramaut/Yemen, where they are discarded in large quantities without effective use.

The collected papaya seeds were washed with water, dried at room temperature for 2 hours, then ground with a hand pestle and sieved to obtain a powder with a particle size of 0.5 mm, as shown in Figure 1. Then the powder was soaked in sodium hydroxide solution (30%) for 24 hours. The solution was then filtered, and the powder was placed in a furnace and heated to 450°C for two hours. The sample was then washed with 0.01 N hydrochloric acid until the filtrate became neutral. After that, the sample was dried at 110°C for 2 hours. In the present study, sodium hydroxide (NaOH) was used in the preparation of activated carbons because it is cost-effective and easy to remove after activation with good performance [19]. The surface area was estimated mathematically from iodine number and methylene blue number [20]. Some of the prepared activated carbon and papaya seed properties are shown in Table 1.

The three chloride solutions (SnCl₂, FeCl₂, SnCl₂ + FeCl₂) were prepared using Acetic acid as the solvent [21]. Adding 2 g of SnCl₂ or FeCl₂ to 8 ml of deionized water and 4 ml of acetic acid and stirring for one hour. Then the solution was stirred for one hour at 90°C. This solution was put in the furnace for 1 hour at 400°C. For FeO₂-SnO₂ bimetallic, 1 g of FeCl₂ was mixed with 1 g of SnCl₂ powder and added to 8 ml, followed by the same previous preparation method [12].

Bimetallic Fe₂O₃-SnO₂ nanoparticles were prepared by thermal decomposition using metal chlorides [21]. Each of the three obtained transparent solutions was loaded onto prepared activated carbon from papaya seeds, at a 20% metal/activated carbon, and stirred for 1 hour at 110°C until dried. Finally, it was placed in an oven for 2 hours at 400°C to obtain oxide nanoparticles. The reactions can be described as below [22]:

The L₉ array was used to determine the best parameters for maximum adsorption. A set of three-factor experiments with three levels is shown in Table 2. The signal-to-noise (S/N) ratio was calculated from the equation below [17,18]. The larger the value of (S/N), the better.

Adsorption isotherms were determined by stirring 50 mL of dye solutions at different concentrations with 0.5 g of adsorbent at ambient temperature until equilibrium was reached. Then the samples were measured by a UV-Vis Spectrometer at λ = 665 nm. The uptake qt was calculated from the equation below:

Where: C₀ and C_t are the concentrations before and after dye adsorption, respectively; V (L) is the volume; and W (g) is the weight of the adsorbent.

To determine the ash content, 0.5 g of the adsorbent was weighed in a pre-weighted ceramic crucible. After that, the sample was heated to 750°C for 3 hours. Ash percent was calculated as below:

To determine moisture content, 0.25 gram of adsorbent was placed into the crucible in the oven at 110°C. Then the sample was allowed to cool. The moisture content was calculated as below:

W₁ is the original adsorbent weight (g), W₂ the crucible weight (g), W₃ the crucible with ash weight (g), W₄ the crucible with dried adsorbent weight (g).

Iodine number (IN) (mg/g) measures the level of carbon activation. A higher iodine number means higher degree of activation and higher micropore content. To determine the iodine number, 0.05 g of the adsorbent was titrated with sodium thiosulfate. The iodine number was calculated by the equation below:

V_b (ml) and V_s (ml) are volumes of sodium thiosulfate for blank and sample (ml), respectively, M (g). is the mass of adsorbent.

Chemical functional groups were identified by an FT-IR spectrophotometer 4600- Jasco/Japan using KBr pellets, where a small amount of adsorbent was mixed with milligrams of KBr and manually pressed under 8 tons. The frequency range was measured between (400-4000 cm^-1). Muffle Furnace model (FB1310M) from Thermolyne/U.S.A with a thermostat to control and program temperature up to 1300°C (1573 K) was used to prepare the adsorbents, dry the samples, and measure the moisture and ash content. To study the morphology of papaya seeds, prepared activated carbon and Fe₂O₃-SnO₂/AC surfaces, scanning electron microscopy (SEM) from the FEI company model Quanta 200 was used.

Figure 2 shows FT-IR spectra of papaya seeds and the prepared activated carbon, illustrating the chemical functional groups. O–H stretching vibration band at 3400 cm^-1 is from hydroxyl groups, phenols, or alcohols, and the position of this band at a lower wavenumber may be due to adsorbed water from the surroundings. In contrast, the C–H aliphatic stretching vibration bands appear at 2924 cm^-1 and at 2854 cm^-1 [23]. The peak in the region of ≈1650 cm^-1 has been observed by C=C stretching absorption while C-O stretching peak at 1100 cm^-1 may be belong to phenols and esters groups at the surface of adsorbent [24,25]. The presence of hydroxyl groups, carbonyl groups, and aromatic compounds in adsorbents, which play important roles in the adsorption process, provides evidence of the lignocellulosic structure of papaya seeds. Two bands at 1745 and 1634 have changed into a strong sharp peak at 1556 cm^-1 ascribed to C=O conjugated with C=C stretching vibration, like a highly conjugated C=O stretching in carboxylic groups. This indicates an increase in oxygen content on the surface after activation. This result was observed in other materials such as Tunisian olive-waste cakes [26] and lignocellulose.

UV-Vis absorption spectrum for the transparent Fe₂O₃-SnO₂, Fe₂O₃ and SnO₂ solutions (Figure 3) shows a absorption narrow peak between 278-295 nm. UV-Vis absorption spectrum of SnO₂ nm has an absorption peak at 295nm, which corresponds to a band gap of 4.1 eV, indicating a nanoparticle size corresponding to a bulk Bohr exciton radius equal to 2.8 nm. The Fe₂O₃ UV-Vis absorption has narrow band at 280 nm, the absorption spectra Fe₂O₃/SnO₂ showed a blue shift and the peak was found to be at 278 nm calculated optical band gap of ~4.48 eV [27].

The SEM images of the papaya seed surface, shown in Figure 4, indicate that the papaya powder particles are spherical, with a particle size of about 3 µm, while the prepared activated carbon is cracked and scattered, with a smaller particle size compared to papaya powder particles. In agreement with the data from the ultraviolet spectrum. The SEM images for SnO₂ and Fe₂O₃ Figure 5 show that the nanoparticle size was about 80 nm. The SEM image of the Fe₂O₃-SnO₂/AC surface Figure 6 indicates the nanostructures of Fe₂O₃-SnO₂/AC, which shows a good dispersion of Fe₂O₃ and SnO₂ on the surface of activated carbon [13].

There are three types of pore size: micropore (less than 2 nm), mesopore (2-50 nm), and macropore (more than 50 nm). The porous volume and the surface area were calculated from iodine number and methylene blue number [20]. The results were put in Table 3, which shows that the surface areas and pore volumes were increased by chemical activation of the papaya seeds because chemical activation led to the development of micropores in the prepared activated carbon, increasing the porosity and surface area. The nanocatalyst 20% Fe₂O₃-SnO₂/AC has less porosity than the activated carbon because the metals close the pores. However, they haven't affected it so much, and the Fe₂O₃-SnO₂/AC 20% nanocatalyst still has a high surface area of 594 cm²/g with a large micropore volume of 0.291 cm³/g and a large total pore volume of 0.616 cm³/g, this indicates that the nanoparticles don’t agglomerate and don’t block the pores. They don’t crystallize outside of the pores.