Methylene Blue Dye Photocatalytic Degradation over Synthesised Fe3O4/AC/TiO2 Nano-Catalyst: Degradation and Reusability Studies

In this study, activated carbon (AC) from coconut shell, as a widely available agricultural waste, was synthesised in a simple one-step procedure and used to produce a magnetic Fe3O4/AC/TiO2 nano-catalyst for the degradation of methylene blue (MB) dye under UV light. Scanning electron microscopy revealed that TiO2 nanoparticles, with an average particle size of 45 to 62 nm, covered the surface of the AC porous structure without a reunion of its structure, which according to the TGA results enhanced the stability of the photocatalyst at high temperatures. The photocatalytic activities of synthesised AC, commercial TiO2, Fe3O4/AC, and Fe3O4/AC/TiO2 were compared, with Fe3O4/AC/TiO2 (1:2) exhibiting the highest catalytic activity (98%). Furthermore, evaluation of the recovery and reusability of the photocatalysts after treatment revealed that seven treatment cycles were possible without a significant reduction in the removal efficiency.


Introduction
Numerous techniques for removing a synthetic dye from aqueous effluents to reduce their effects on the environment have been designed and developed, including adsorption, Fenton-like reactions, activated sludge, membranes, reductive degradations using zero-valent iron, and photo-catalysis. Among these methods, methods based on membranes and adsorption have shown high removal efficiencies (>90%) [1] but have several disadvantages, as these methods only can trap and retain the impurities, not destroy them. Moreover, the used adsorbents and rejected water from membrane processes are categorised as secondary contaminants [2].
Photocatalysis, in which the clean, safe, and inexhaustibly abundant energy of the sun can be harnessed, is a major advance in this field. Dyes such as synthetic organic compounds and common pollutants in wastewaters are widely used in many industries, such as textiles, cosmetics, food, printing, plastics, and leather. Disposal of dye pollutants into the water sources inhibits sunlight increasing its surface area, and enhancing the stability of the catalyst during the regeneration process. Such a mechanistic understanding is very important for the controlled growth of TiO 2 /AC/Fe 3 O 4 , which may be used in many applications.

Preparation of AC and Fe 3 O 4 /AC
A coconut shell was used to synthesise AC in a simple step. The coconut shell was washed with distilled water to remove impurities, then oven-dried at 70 • C for 10 h. The dried coconut shell was ground and placed in a tube furnace at 800 • C for 2 h, with a heating rate of 10 • C/min under a purified nitrogen flow. The prepared AC was washed several times to remove the smell and dried in an oven at 70 • C for 1 day for further experiments.
Ferrous and ferric phosphate salts in an alkaline aqueous solution were used for the fabrication of Fe 3 O 4 /AC nanoparticles via in situ chemical co-precipitation. Briefly, 5 g of AC, 6.66 g of FeCl 3 ·6H 2 O (0.06 M), and 3.66 g of FeSO 4 ·7H 2 O (0.04 M) were dissolved in 200 mL of distilled water and stirred vigorously with a mechanical stirrer on the hot plate and heated at 85 • C ± 1 • C for 1 h. Then, a KOH solution (20 M) was added dropwise into the prepared mixture while stirring with a magnetic stirrer until the pH reached 10-11. The mixture was stirred for 1 h to precipitate the hydrated iron oxide and cooled to room temperature. A strong magnet was used to separate the black precipitate, which was repeatedly rinsed with deionised water seven times and finally dried at 75 • C overnight in an oven.

Preparation of Fe 3 O 4 /AC/TiO 2
The desired amount of the prepared and dried magnetic AC (1:1, 1:2, 1:4, and 1:8 Fe 3 O 4 /AC:TiO 2 molar ratio) was obtained as follows: first, TiO 2 nanoparticles were dissolved in ethanol and homogenised by ultrasonication for 10 min. The obtained Fe 3 O 4 /AC was added to the solution and mixed on the hot plate at 110 • C for 1 h (during the heating, most of the solution evaporated) before the mixture was calcined at 400 • C in the furnace (without washing). The calcined catalysts were washed with distilled water seven times and oven-dried at 100 • C for 24 h.

Catalyst Characterisation
X-ray diffraction (XRD) was conducted using a Bruker D8 Advance X-Ray diffractometer operating at (40 kV, 35 mA) under Cu-Kα radiation (λ = 0.154 nm). Scanning electron microscopy (SEM) was performed with an FE-SEM, JEO JSM 7600-F (JOEL Ltd., Tokyo, Japan) instrument equipped with an EDX. Thermogravimetric analysis (TGA) was conducted on a Mettler Toledo TGA/SDTA 851e (Mettler Toledo Corporation, Zurich, Switzerland) and heated from room temperature to 1000 • C at a heating rate of 10 • C/min in air. The Raman spectra were collected by a Renishaw model 1000 Raman microscope (Gloucestershire, UK) using an excitation wavelength of 514 nm in an ambient environment.

Photo-Catalytic Activity for MB Degradation
MB as the case organic pollution was selected for the photo-degradation experiment to investigate the UV-assisted degradation of the MB solution by synthesised AC, Fe 3 O 4 /AC /TiO 2 , Fe 3 O 4 /AC, and TiO 2 at room temperature using a 1000-W UV lamp which emitted light with a wavelength of 664 nm. All the reactions were performed in magnetically stirred glass vessels located at a distance of 5 cm from the UV lamp and open at the top. Next, 0.1 g of the catalyst was added to 100 mL of the MB dye (100 mg/L, pH 11), then the mixture was sonicated in a water bath for 30 min to ensure MB adsorption equilibrium on the catalyst surface (120 rpm in the dark place). The solution was irradiated to degrade MB in the dark to prevent the impact of the outer light. The MB concentration of all the solutions was measured using a UV-VIS spectrophotometer (Shimadzu UV-2700 UV-Vis, Shimadzu, Kyoto, Tokyo) by measuring the absorbance of the solution (λ = 664 nm). Moreover, the effect of the initial pH in the range of 10-13 on the photo-catalyst activity of Fe 3 O 4 /AC/TiO 2 1:2 was investigated. The effect of H 2 O 2 on the degradation of the MB dye over time was investigated by adding 0.1 g of 40 mM H 2 O 2 to the dye solution as an oxidising agent. All the experiments were performed in triplicate. The removal efficiency (%) of MB was calculated using the following equation: where C 0 is the initial concentration of MB and C t is the MB concentration at different irradiation times. Furthermore, the reusability potential and the stability of the catalyst were investigated for seven cycles.

XRD Measurements of Fe 3 O 4 /AC and Fe 3 O 4 /AC/TiO 2
The structural properties of the prepared composites were characterised by X-ray diffraction, with the typical XRD patterns for the synthesised AC, TiO 2 nanoparticles, Fe 3 O 4 /AC, and Fe 3 O 4 /AC/TiO 2 in different ratios shown in Figure 1. In Figure 1a, the peaks at 24.8 • and 42.5 • denote the carbonaceous structures in the AC [21]; Figure 1b shows the XRD spectrum of TiO 2 nanoparticles, and Figure 1c shows the XRD spectrum of

Surface Morphology Analysis of Fe3O4/AC/TiO2 and Fe3O4/AC
The morphology of the synthesised coconut shell AC was observed by SEM (Figure 2a), showing the AC porous structure and heterogeneous surface. The pores observed (size: several micrometres) acted as channels for the adsorbents entering the adsorbate entrance. Figure 2b shows the presence of white aggregates of the metal oxide in the pores of the tile-like AC structures. a b was approximately 14, 18.7, 18.3, and 18.9 nm, respectively, indicating that increasing the ratio of TiO 2 from (1:1) to (1:2) decreased the crystallite size, which then increased when the TiO 2 ratio was increased from (1:2) to (1:4). In addition, increasing the ratio from (1:1) to (1:4) decreased the presence of the (311) peak, which was attributed to the elimination of the magnetic properties of the samples upon the increase in the TiO 2 ratio. However, increasing the TiO 2 ratio had no significant effects on the particle size. Moreover, TiO 2 -coated carbon-based materials (d-f) showed the same diffraction peaks as TiO 2 , with a small difference in the peak width corresponding to the increase in the crystallite size.

Surface Morphology Analysis of Fe 3 O 4 /AC/TiO 2 and Fe 3 O 4 /AC
The morphology of the synthesised coconut shell AC was observed by SEM (Figure 2a), showing the AC porous structure and heterogeneous surface. The pores observed (size: several micrometres) acted as channels for the adsorbents entering the adsorbate entrance. Figure 2b shows the presence of white aggregates of the metal oxide in the pores of the tile-like AC structures.

Surface Morphology Analysis of Fe3O4/AC/TiO2 and Fe3O4/AC
The morphology of the synthesised coconut shell AC was observed by SEM (Figure 2a), showing the AC porous structure and heterogeneous surface. The pores observed (size: several micrometres) acted as channels for the adsorbents entering the adsorbate entrance. Figure 2b shows the presence of white aggregates of the metal oxide in the pores of the tile-like AC structures. The SEM results of the TiO2-coated Fe3O4/AC samples are presented in Figure 3a-c. The TiO2 coatings were expected to modify the morphology of carbon, resulting in almost spherical particles. The ImageJ digital processing software was used to analyse the particle size and particle size distributions are shown in Figure 3d-f. The SEM analysis showed rough surfaces lightly produced on AC due to the loading of the TiO2 nanoparticles on the AC surface after synthesis. The TiO2 nanoparticles completely covered the AC surface without a reunion of its structure, even if it was not homogeneous. Figure 3 shows that TiO2 caused an increase in the average particle size of carbon, the average particle size was 50.22, 62.342, and 45.31 nm for sample 1:1, sample 1:2, and sample 1:4, respectively. As shown in Figure 3, the particle size increased with an increase in the TiO2 ratio from 1:1 to 1:2, then decreased with an increase in the ratio from 1:2 to 1:4, which was in good agreement a b  Figure 3a-c. The TiO 2 coatings were expected to modify the morphology of carbon, resulting in almost spherical particles. The ImageJ digital processing software was used to analyse the particle size and particle size distributions are shown in Figure 3d-f. The SEM analysis showed rough surfaces lightly produced on AC due to the loading of the TiO 2 nanoparticles on the AC surface after synthesis. The TiO 2 nanoparticles completely covered the AC surface without a reunion of its structure, even if it was not homogeneous. Figure 3 shows that TiO 2 caused an increase in the average particle size of carbon, the average particle size was 50.22, 62.342, and 45.31 nm for sample 1:1, sample 1:2, and sample 1:4, respectively. As shown in Figure 3, the particle size increased with an increase in the TiO 2 ratio from 1:1 to 1:2, then decreased with an increase in the ratio from 1:2 to 1:4, which was in good agreement with the obtained particle size from XRD. In Figure 3a,b, the TiO 2 particles covered the Fe 3 O 4 /AC surface but were aggregated in Figure 3c.
surface but were aggregated in Figure 3c.
The EDX spectra of the synthesised AC, Fe3O4/AC, and Fe3O4/AC/TiO2 in different ratios are shown in Figure 4a-e. Figure 4b indicates the presence of C, Fe, and O; Figure 4b shows the presence of potassium in the AC, which could be attributed to the addition of the KOH solution increasing the pH to 10-11. In Figure 4c-e, the peak ratio of Ti to AC and Fe confirmed the ratio of the used TiO2 and Fe3O4/AC. The presence of Au was attributed to the gold coating for SEM characterisation.  The TGA profiles in air for the samples of Fe3O4/AC and Fe3O4/AC/TiO2 in the ratios of 1:1, 1:2, and 1:4 are shown in Figure 5. Figure 5a shows that there was a one-step weight loss process. Fe3O4/AC was stable in air up to 538 °C, with the main weight loss occurring from 538 °C to 650 °C, and 59.8% weight loss until 1000 °C. The TGA measurements shown in Figure 5a indicate that the organic phases decomposed at temperatures below 650 °C. The TGA curves in Figure 5b Increasing the ratio of TiO2 caused a decrease in the weight loss percentage, indicating that a significant loss of carbon from AC (in Fe3O4/AC/TiO2) did not occur and that carbon might be inserted within the TiO2 structure, thereby confirming that TiO2-loaded AC materials had better adsorption capacity and photocatalyst ability at higher temperatures. Meanwhile, no exothermic peaks were observed in either of the TGA curves (Figure 5b-d) at around 450 °C, indicating that there was no brookite transformation to the anatase phase in the Fe3O4/AC/TiO2 samples [23,24], so the brookite phase became more stable during the calcination of the samples at 400 °C. Therefore, the Fe3O4/AC/TiO2 composite was thermally stable, hence suitable for the practical applications.  Figure 5. Figure 5a shows that there was a one-step weight loss process. Fe 3 O 4 /AC was stable in air up to 538 • C, with the main weight loss occurring from 538 • C to 650 • C, and 59.8% weight loss until 1000 • C. The TGA measurements shown in Figure 5a indicate that the organic phases decomposed at temperatures below 650 • C. The TGA curves in Figure 5b 4), respectively. Increasing the ratio of TiO 2 caused a decrease in the weight loss percentage, indicating that a significant loss of carbon from AC (in Fe 3 O 4 /AC/TiO 2 ) did not occur and that carbon might be inserted within the TiO 2 structure, thereby confirming that TiO 2 -loaded AC materials had better adsorption capacity and photocatalyst ability at higher temperatures. Meanwhile, no exothermic peaks were observed in either of the TGA curves (Figure 5b-d) at around 450 • C, indicating that there was no brookite transformation to the anatase phase in the Fe 3 O 4 /AC/TiO 2 samples [23,24], so the brookite phase became more stable during the calcination of the samples at 400 • C. Therefore, the Fe 3 O 4 /AC/TiO 2 composite was thermally stable, hence suitable for the practical applications.  The Raman spectra of the obtained samples were collected in the range of 0-2000 cm −1 and are presented in Figure 6. Figure 6a shows two diffraction peaks around 1325 cm −1 (G band) and 1582 cm −1 (D band). The D band of the carbon material structure was associated with defects and became active when the crystallinity decreased, and the G band corresponded to the stretching vibrations with the basal graphene layers [25]. The characteristic diffraction peak around 670 cm −1 corresponded to Fe 3 O 4 and demonstrated the magnetic property of the obtained sample. The combination of Raman and XRD findings confirmed that the Fe 3 O 4 composite was formed during the synthesis. Figure 6b-d shows a comparison of the Raman spectra of the Fe 3 O 4 /AC/TiO 2 composites in the ratio of 1:1, 1:2, and 1:4, respectively. A well-resolved TiO 2 Raman peak was observed at~149 cm −1 and was attributed to the main E g anatase vibration mode for all three samples. The three diffraction peaks observed at around 396, 515, and 639 cm −1 indicated the major species of anatase crystallites [24]. There were two broad and weak peaks (in all the three Fe 3 O 4 /AC /TiO 2 samples) at~1300 cm −1 and 1600 cm −1 , which were assigned to the ill-organised graphite and E 2g mode in graphite, respectively (Figure 6e) [26]. The Raman spectra of the obtained samples were collected in the range of 0-2000 cm −1 and are presented in Figure 6. Figure 6a shows two diffraction peaks around 1325 cm −1 (G band) and 1582 cm −1 (D band). The D band of the carbon material structure was associated with defects and became active when the crystallinity decreased, and the G band corresponded to the stretching vibrations with the basal graphene layers [25]. The characteristic diffraction peak around 670 cm −1 corresponded to Fe3O4 and demonstrated the magnetic property of the obtained sample. The combination of Raman and XRD findings confirmed that the Fe3O4 composite was formed during the synthesis. Figure 6bd shows a comparison of the Raman spectra of the Fe3O4/AC/TiO2 composites in the ratio of 1:1, 1:2, and 1:4, respectively. A well-resolved TiO2 Raman peak was observed at ~149 cm −1 and was attributed to the main Eg anatase vibration mode for all three samples. The three diffraction peaks observed at around 396, 515, and 639 cm −1 indicated the major species of anatase crystallites [24]. There were two broad and weak peaks (in all the three Fe3O4/AC /TiO2 samples) at ~1300 cm −1 and 1600 cm −1 , which were assigned to the ill-organised graphite and E2g mode in graphite, respectively (Figure 6e) [26].

Photo-Catalytic Activity
The synthesised catalyst activities were evaluated based on the photo-degradation of the MB aqueous solution (100 mg/L, pH 11) under UV irradiation at 664 nm. A blank experiment with the MB dye solution and with no photocatalyst was performed for comparison. After 30 min of irradiation in the absence of the photocatalyst, no evident MB degradation was observed. The pure photocatalytic removal efficiencies of the commercial TiO2 nanoparticles (in 30 min and 60 min) were 23% and 31%, respectively. The adsorption efficiencies of the synthesised coconut shell AC were 68% and 76.2% (in 30 min and 60 min) and the Fe3O4/AC/TiO2 photocatalyst samples presented a high removal efficiency (ranging from ~66% (1:4) to 98% (1:2)). TiO2 loaded on AC presented a very high photocatalytic degradation efficiency compared with pure TiO2, synthesised AC, and Fe3O4/AC (Figure 7). The concentration of the MB aqueous solution decreased significantly, in the two-step physical-chemical phenomenon of (1) adsorption by AC and (2) photocatalytic decomposition by TiO2 and Fe3O4. The magnetic AC played the role of an adsorbent at the ratios of 1:1, 1:2, and 1:4 (Fe3O4/AC and TiO2), with most AC channels not dominated by the TiO2 [10]. Figure 7 shows a very low degradation percentage of the photocatalyst samples of Fe3O4/AC and TiO2 in the ratio of 1:4, which may be attributed to the aggregation of TiO2 on AC, as shown in Figure 3c. The adsorption mainly occurred on the surface of the catalysts, and AC played the main role of an adsorbent, with

Photo-Catalytic Activity
The synthesised catalyst activities were evaluated based on the photo-degradation of the MB aqueous solution (100 mg/L, pH 11) under UV irradiation at 664 nm. A blank experiment with the MB dye solution and with no photocatalyst was performed for comparison. After 30 min of irradiation in the absence of the photocatalyst, no evident MB degradation was observed. The pure photocatalytic removal efficiencies of the commercial TiO 2 nanoparticles (in 30 min and 60 min) were 23% and 31%, respectively. The adsorption efficiencies of the synthesised coconut shell AC were 68% and 76.2% (in 30 min and 60 min) and the Fe 3 O 4 /AC/TiO 2 photocatalyst samples presented a high removal efficiency (ranging from~66% (1:4) to 98% (1:2)). TiO 2 loaded on AC presented a very high photocatalytic degradation efficiency compared with pure TiO 2 , synthesised AC, and Fe 3 O 4 /AC (Figure 7). The concentration of the MB aqueous solution decreased significantly, in the two-step physical-chemical phenomenon of (1) adsorption by AC and (2) photocatalytic decomposition by TiO 2 and Fe 3 O 4 . The magnetic AC played the role of an adsorbent at the ratios of 1:1, 1:2, and 1:4 (Fe 3 O 4 /AC and TiO 2 ), with most AC channels not dominated by the TiO 2 [10]. Figure 7 shows a very low degradation percentage of the photocatalyst samples of Fe 3 O 4 /AC and TiO 2 in the ratio of 1:4, which may be attributed to the aggregation of TiO 2 on AC, as shown in Figure 3c. The adsorption mainly occurred on the surface of the catalysts, and AC played the main role of an adsorbent, with the degradation subsequently occurring on TiO 2 [27]. The porous structure of AC with an appropriate content resulted in the dye molecules gathering around the TiO 2 nanoparticles at a low concentration of the MB solution. Therefore, the porous structure of AC was very important in facilitating the diffusion of the MB reactants and products on the TiO 2 active sites during the photocatalytic reaction, which improved the photocatalytic degradation process [7].
relative stability under irradiation. In the presence of catalysts, the MB-degradation efficiency is greatly improved [28,29].
The degradation performance of the Fe3O4/AC /TiO2 catalysts with different initial pH values (10-13) was investigated. Increasing the pH from 10 to 12 increased the degradation performance from ~91.4% to 98.6%, the degradation performance decreased at pH 13 to ~84% after 120 min. The maximum value of the catalytic activity was observed at pH 12 (~98.3%). These results indicated that the pH significantly affected the degradation efficiency of the catalyst, which was heavily dependent on the transformation of the surface properties and activities of the catalyst and the pollutant [30,31].
In addition, the presence of H2O2 during the MB photocatalytic degradation was crucial. In this study, the MB photocatalytic degradation was evaluated using Fe3O4/AC/TiO2 (best photocatalyst) with H2O2 and with free H2O2. Figure 9 shows that the presence of H2O2 accelerated the photocatalytic performance of the nanocomposites, which could be due to the production of the active hydroxyl radicals. Hydrogen peroxide, as a powerful oxidiser, promoted the photocatalytic performance of the nanocomposites [32]. As reported by Poulopoulos et al. [33], a combination of H2O2 with TiO2 is very effective in improving the photocatalytic performance.   The results show that: without the catalyst, no MB is degraded under UV-light, implying MB is relative stability under irradiation. In the presence of catalysts, the MB-degradation efficiency is greatly improved [28,29].

Reusability of Catalyst
Recovery and reusability are essential parameters for the selection of a cost-effective and feasible catalyst for pilot-scale remediation systems. The reusability performance of our best nanocomposite was investigated for seven cycles of MB photo-degradation using 100 mL of the MB dye (100 mg/L). Figure 10 shows the recyclability and stability of the Fe3O4/AC/TiO2 (1:2) catalyst after seven cycles during the 60-min reaction. The Fe3O4/AC/TiO2 catalyst could be recycled conveniently after the treatment via a strong magnet and reused; the results are depicted in Figure 10. The catalyst recovery and reusability processes are shown in Figure 11. After seven cycles, the photo-activity of Fe3O4/AC/TiO2 decreased from ~98% to ~93% (only 5%) after seven successive cycles with high efficiency. This slight decrease in the photocatalysis activity could be attributed to the following: (1) material losses might occur during the recovery step (washing and drying), which would lead to a lower dose in the subsequent cycle, thereby decreasing the surface catalytic activity and degrading the performance [34]. (2) The properties of magnetic nanoparticles, such as aggregation (this effect can reduce the effective surface area and decrease the number of active sites) and fouling, might The degradation performance of the Fe 3 O 4 /AC /TiO 2 catalysts with different initial pH values (10-13) was investigated. Increasing the pH from 10 to 12 increased the degradation performance from 91.4% to 98.6%, the degradation performance decreased at pH 13 to~84% after 120 min. The maximum value of the catalytic activity was observed at pH 12 (~98.3%). These results indicated that the pH significantly affected the degradation efficiency of the catalyst, which was heavily dependent on the transformation of the surface properties and activities of the catalyst and the pollutant [30,31].
In addition, the presence of H 2 O 2 during the MB photocatalytic degradation was crucial. In this study, the MB photocatalytic degradation was evaluated using Fe 3 O 4 /AC/TiO 2 (best photocatalyst) with H 2 O 2 and with free H 2 O 2 . Figure 9 shows that the presence of H 2 O 2 accelerated the photocatalytic performance of the nanocomposites, which could be due to the production of the active hydroxyl radicals. Hydrogen peroxide, as a powerful oxidiser, promoted the photocatalytic performance of the nanocomposites [32]. As reported by Poulopoulos et al. [33], a combination of H 2 O 2 with TiO 2 is very effective in improving the photocatalytic performance.

Reusability of Catalyst
Recovery and reusability are essential parameters for the selection of a cost-effective and feasible catalyst for pilot-scale remediation systems. The reusability performance of our best nanocomposite was investigated for seven cycles of MB photo-degradation using 100 mL of the MB dye (100 mg/L).

Reusability of Catalyst
Recovery and reusability are essential parameters for the selection of a cost-effective and feasible catalyst for pilot-scale remediation systems. The reusability performance of our best nanocomposite was investigated for seven cycles of MB photo-degradation using 100 mL of the MB dye (100 mg/L). Figure 10 shows the recyclability and stability of the Fe 3 O 4 /AC/TiO 2 (1:2) catalyst after seven cycles during the 60-min reaction. The Fe 3 O 4 /AC/TiO 2 catalyst could be recycled conveniently after the treatment via a strong magnet and reused; the results are depicted in Figure 10. The catalyst recovery and reusability processes are shown in Figure 11. After seven cycles, the photo-activity of Fe 3 O 4 /AC/TiO 2 decreased from~98% to~93% (only 5%) after seven successive cycles with high efficiency. This slight decrease in the photocatalysis activity could be attributed to the following: (1) material losses might occur during the recovery step (washing and drying), which would lead to a lower dose in the subsequent cycle, thereby decreasing the surface catalytic activity and degrading the performance [34].
(2) The properties of magnetic nanoparticles, such as aggregation (this effect can reduce the effective surface area and decrease the number of active sites) and fouling, might change during the seven cycles [35]. (3) The adsorptive catalytic surface activity of the catalyst gradually decreased because of the obstruction of the pores and the active sites by catechol and its intermediates after each cycle [30,36]. In general, several cycles can be conducted using the same material with almost the same pollutant degradation efficacy [37,38].
Nanomaterials 2020, 10, x FOR PEER REVIEW 12 of 15 change during the seven cycles [35]. (3) The adsorptive catalytic surface activity of the catalyst gradually decreased because of the obstruction of the pores and the active sites by catechol and its intermediates after each cycle [30,36]. In general, several cycles can be conducted using the same material with almost the same pollutant degradation efficacy [37,38].

Conclusions
Coconut shell AC and magnetic Fe3O4/AC/TiO2 have been synthesized in this work. The samples were characterized via different techniques. All the synthesised samples showed higher degradation efficiency under UV light than commercial TiO2, in particular, Fe3O4/AC/TiO2 (1:2) presented the highest degradation rate of 98% in 60 min. The simple synthesised coconut shell AC showed high adsorption efficiency (76.2%) and could be used with the TiO2 photo-catalyst to enhance the TiO2

Conclusions
Coconut shell AC and magnetic Fe 3 O 4 /AC/TiO 2 have been synthesized in this work. The samples were characterized via different techniques. All the synthesised samples showed higher degradation efficiency under UV light than commercial TiO 2 , in particular, Fe 3 O 4 /AC/TiO 2 (1:2) presented the highest degradation rate of 98% in 60 min. The simple synthesised coconut shell AC showed high adsorption efficiency (76.2%) and could be used with the TiO 2 photo-catalyst to enhance the TiO 2 photocatalytic activity (22% higher). The prepared magnetic photocatalysts presented good magnetic separation efficiency, which made the recovery and reusability affordable and simple. In conclusion, this paper presents an environmentally friendly and economical alternative for the photo-catalyst degradation of MB dye using the widely available agricultural waste for the production of TiO 2 -loaded carbonaceous materials.