Oxygen Functionalized Graphitic Carbon Nitride for Photocatalytic Degradation of Dye

Photocatalyst such as graphitic carbon nitride (g-C3N4) is being studied intensively due to its ability in photocatalysis. g-C3N4 is a metal-free semiconductor photocatalyst with a bandgap of approximately 2.7 eV which contributes to its good visible light harvesting ability. In this work, bulk g-C3N4 was produced via pyrolysis of melamine in a muffle furnace. Functionalized g-C3N4 with improved properties was synthesized via modified Hummers method. The powdered form of functionalized g-C3N4 were characterized using SEM and EDX to identify its physiological properties. The result showed that the introduction of oxygen into g-C3N4 is proven by the increased content of oxygen in the functionalized g-C3N4 upon oxidation using Hummers method. Besides, exfoliation of g-C3N4 to smaller particle size observed from the SEM images. Then, the phototcatalytic performances of the bulk g-C3N4 and functionalized g-C3N4 were evaluated by degrading of Methylene Blue (MB) dyes under LED light irradiation. The result revealed that the bulk g-C3N4 has a higher efficiency in removal of dyes (56.40 % in 150 minutes) than the functionalized g-C3N4 (22.60 % in 150 minutes) which indicates that it has a better photocatalytic degradation ability, which possibly due to the destruction of compound structure under strong acid treatment.


Introduction
Graphitic carbon nitride (g-C3N4) is a metal-free semiconductor photocatalyst with a bandgap of approximately 2.7 eV.According to Zhu et al. (2014), g-C3N4 is stable under atmospheric condition and it also possesses basic surface sites that potentially made it a good catalyst [1].g-C3N4 is widely studied for photocatalysis due to its visible light harvesting ability.Photocatalysis is the use of the light energy to activate the photocatalyst by promoting electrons from valence band to conduction band, resulting in generation of positively charged hole [2].The interaction of the holes and water will lead to the degradation of pollutants as well as inactivation of microorganism.Besides, photocatalysis was also applied to catalyse the reactions including water splitting and carbon dioxide reduction.The criteria that affect the catalytic performance of the photocatalyst includes light harvesting, recombination rate, surface reaction and so on.There are several advantages of g-C3N4 as a photocatalyst, including excellent chemical and thermal stability, good visible light absorption ability, appropriate electronic structure and eco-friendly nature.Despite of having many benefits, g-C3N4 is found to be having a fast recombination rate of charge carriers and also having a lower specific surface area due to stacking of layered structure [3].To improve the photocatalytic ability of the photocatalyst, modification of the material by the application of strategies such as doping, morphology alteration, semiconductor coupling, shape-control synthesis and more is much needed.Several studies show that by introducing electron-withdrawing group to g-C3N4, the photocatalytic ability of the photocatalyst can be improved [3,4].Hummers method is usually employed during the production of graphene oxide.Considering the similar properties of g-C3N4 and graphene, the oxidation of g-C3N4 could also be done through modified Hummers method, in which bulk g-C3N4 could be exfoliated to g-C3N4 nanosheets and functional groups including carbonyl group (C=O) could be introduced to the surface of the compound [4,5].On the other hand, the research gap of this study is the extended adsorption duration of the photocatalyst.The mixture of photocatalyst and MB dye solution was stirred overnight to achieve adsorption-desorption equilibrium.Previous studies done by other researches was found to be conducting the adsorption step for a duration of 30 mins to 1 h [3,6].To further confirm the reaction is dominated by photocatalytic reaction, the adsorption duration can be extended to ensure the photocatalytic efficiency is not contributed by adsorption.
With increasing population in the world, water pollution has become a major concern due to rapid growth in industrialization.Dyes are one of the major water contaminants that bring a negative impact towards the environment.Methylene Blue (MB) dye which is phenothiazine derivative, is commonly used in textile and paper industry [7].MB dyes are highly toxic and carcinogenic.Photocatalysis is a promising method used in degrading the harmful dyes materials.In this work, in order to evaluate the photocatalysis performance of the photocatalyst, photocatalytic degradation of MB dye was conducted.

Synthesis of bulk g-C3N4
Melamine powder (Reagent Grade, 99 %, Sigma-Aldrich) was placed inside a crucible with lid covered and directly heated in a programmable furnace at 550 ⁰C for 4 hours in the furnace.After that, the product was grounded into fine powder form.Yellow colour bulk g-C3N4 in powder form was obtained as the product.

Synthesis of functionalized g-C3N4 via modified Hummers method
Bulk g-C3N4 was allowed to react with concentrated sulphuric acid (Reagent Grade, 98 %, Sigma-Aldrich) at 60 ⁰C for 1 h.Then, the content was placed in an ice bath to keep the temperature low.On the other hand, potassium permanganate (KMnO4) was added slowly into the mixture.After that, the mixture was heated to 30 ⁰C for 30 min.Next, deionized water was added very slowly and carefully into the mixture.Hydrogen peroxide (H2O2) solution was added into the mixture until a milky white colour product was obtained.The product was washed and centrifuged for multiple times prior to vacuum drying at 60 ⁰C.

Characterization
The surface morphologies and elemental composition of bulk g-C3N4 and functionalized g-C3N4 were studied and recorded by using scanning electron microscopy (SEM, Hitachi S-3400N) and energy dispersive X-ray spectroscopy (EDX, Hitachi S-3400N) Prior to morphology studies and elemental analysis, the samples were coated with a layer of gold to improve the conductivity of the sample.

Photocatalyic degradation of MB dye under visible light irradiation
30 mg of photocatalyst was added into 30 mL of MB dye solution (10 ppm).The mixture was stirred overnight to achieve adsorption-desorption equilibrium.Then, the photocatalytic test was conducted under visible light irradiation by using a white light LED bulb (20 W).The samples were drawn periodically to record their UV-vis spectra on the spectrophotometer.The degradation efficiency of MB dye can be obtained by using equation as shown below: where  is the concentration of MB dye and  is absorbance of MB dye solution at 664 nm at different time intervals.
The kinetic degradation of MB dye was studied according to the pseudo-first order model as shown in the equation below: where  is the concentration of MB dye and  (min -1 ) is the apparent reaction rate constant and  (min) is the time taken of the reaction.

Results and Discussion
Figure 1 shows the synthesis route of functionalized g-C3N4.Yellow colour bulk g-C3N4 was successfully produced through pyrolysis of melamine.After oxidation via Hummers method, g-C3N4 turned into white colour.This is probably due to the reduction in the size of the π-conjugated domains that resulted in the increase of the band energy gap between valence band and conduction band [4].Table 1 shows the elemental analysis of the bulk g-C3N4 and functionalized g-C3N4.The adding of oxygenated functional group into functionalized g-C3N4 was confirmed because the weight percentage of oxygen in functionalized g-C3N4 increased to 11.99 wt%.Theoretically, 4.85 wt% oxygen was detected in bulk g-C3N4 because oxygen in the air came in contact with melamine during pyrolysis.Figure 2 shows the SEM images of bulk g-C3N4 and functionalized g-C3N4.Agglomeration of layered structures was observed for bulk g-C3N4 whereas there was less agglomeration or layer stacking observed in functionalized g-C3N4.Besides, the lateral dimension of the particles of functionalized g-C3N4 is also smaller.The particle size for bulk g-C3N4 is approximately 30 µm while the particle sizes of functionalized g-C3N4 is ranging from 10 µm to 100 nm.This is due to the exfoliation of the bulk g-C3N4 into smaller size by the adding of strong acid.By having a smaller particle size, the photogenerated electrons and holes can reach the solid-liquid interface faster, reducing the rate of recombination of the photogenerated electron-hole pairs [8].Next, photocatalytic degradation of MB was conducted to evaluate the photocatalytic activity of the photocatalyst.Figure 3 demonstrates the UV-Vis absorption spectra of photocatalytic degradation of MB dye for 150 min whereas figure 4 illustrates the comparison of degradation efficiency of MB dye by using bulk g-C3N4 and functionalized g-C3N4.Prior to the photocatalytic reaction, the reactant species would need to be adsorbed onto the surface of the catalyst.Thus, the adsorption ability of the photocatalyst was said to be having a direct impact to the overall photocatalytic efficiency of the photocatalyst.The adsorption percentage of the bulk and functionalized g-C3N4 were found to be 26.40 % and 15.51 % respectively.The adsorption performance of bulk g-C3N4 was better despite having larger particle sizes and stacking structure.This could be owing to the pH of the photocatalyst suspension, in which the functionalized g-C3N4 is more acidic due to the presence of unremoved strong acid after washing.Recent studies have suggested that at lower pH, the adsorption and photocatalysis performance of g-C3N4 are found to be lower due to coulombic repulsions between the similarly charged photocatalyst and dye molecule during the reaction [9,10].Based on figure 4, the bulk g-C3N4 exhibits higher photocatalytic activity which is around 56.40 % in 150 minutes than functionalized g-C3N4 with degradation efficiency at around 22.60 % in 150 minutes.A blank test without photocatalyst was also conducted for comparison.Based on the result, negligible amounts of dye were degraded without addition of photocatalyst under visible light irradiation.Table 2 shows the comparison study of the a) b) degradation of MB dye by using g-C3N4.Both the bulk g-C3N4 and functionalized g-C3N4 produced in this study had showed a lower photocatalytic efficiency as compared to the literature, in which the photocatalyst possessed a MB dye (50 ppm) degradation efficiency of approximately 45 % in 180 minutes [11].Figure 5 shows the kinetic curves for the photocatalytic degradation of MB dye over bulk g-C3N4 and functionalized g-C3N4.The reaction was fitted to the pseudo-first order model with a regression coefficient (R 2 ) of 0.8854 for reaction over bulk g-C3N4 and 0.8528 for reaction over functionalized g-C3N4.The apparent reaction kinetic constant for the bulk and functionalized g-C3N4 were 0.0055 min -1 and 0.0017 min -1 respectively.Hence, in overall, the reaction rate of bulk g-C3N4 was faster, indicating that bulk g-C3N4 is a better photocatalyst than the functionalized g-C3N4.The unexpected result could be owing to the destruction of the structure of g-C3N4 under strong acid treatment, halting its photocatalytic activity.Through Hummers method treatment, the planar atomic structure of the photocatalyst might be destroyed as the hydrogen bonding between layers of polymeric melon units were not strong enough to withstand the strong oxidation effect from the reaction [12].

Figure 4 .
Figure 4. Comparison of degradation efficiency of MB dye by using bulk g-C3N4 and functionalized g-C3N4.

Figure 5 .
Figure 5. Kinetic curves of photocatalytic degradation of MB dye by using bulk g-C3N4 and functionalized g-C3N4.

Table 2 .
Comparison study of degradation of MB dye by using g-C3N4.