Cobalt chloride (CoCl₂), ammonium hydroxide (NH₄OH, 25 - 28%), selenium powder (Se, 99%), Sodium sulfite (Na_2-SO₃), and ethyl alcohol (94%) were purchased from Duksan Pure Chemical Co. Ltd., Korea. Rhodamine B (Rh. B), purchased from Texchem Korea Co. Ltd., was used as a standard industrial dye, All chemicals were used without further purification. All dilutions were carried out using distilled water.

Cobalt selenide was successfully prepared through an ultra-sonication technique. In the synthesis procedure, 2 g of anhydrous Sodium sulfite (Na₂SO₃) and 0.5 g of crude selenium (Se) powder were mixed together in 200 mL of ethylene glycol and stirred under vigorous magnetic stirring. To obtain selenium salt, the solution was stirred continually until the mixture was homogeneous. After that, 0.6 mmol of cobalt acetate was mixed with the above prepared solution and stirred for 20 minutes to obtain a stable solution. Finally, the homogeneous solution was treated by ultra-sonication for 3 h, with a digital sonicator. The CoSe₂ precipitates were obtained using 47 mm Whatman filter paper: precipitates were then heat treated for 5 h at 90°C.

Graphene oxide (GO) (obtained by Hummer’s-Offeman method¹⁹⁾) and a 0.9 molar solution of CoSe₂ particles were dispersed in 200 ml of ethylene glycol at a ratio of 1 : 1 and vigorously stirred for 30 min at 50°C. After ultra-sonication, the graphene oxide was reduced to graphene nanosheet. CoSe₂ compounds were decorated on the graphene sheet to generate CoSe₂-graphene composites. After filtering with 47 mm Whatman filter paper with a pore size of 0.7 mm, the prepared solution was allowed to cool and settle at room temperature. The resultant powder was washed with distilled water 5 times and then transferred into a dry oven for 5 h at 90°C in a furnace. CoSe₂-graphene-TiO₂ nanocomposites were prepared following the same procedure described above. 150 mg of GO and similar amount of titanium precursor (TNB) were dissolved in 200 mL of ethylene glycol for 30 minutes to obtain a homogenous solution; this was denoted as solution A. In the second step, 1.5 g of Na₂SO₃ and 0.2 g of Se were mixed in 200 mL of ethylene glycol and stirred until a homogenous solution was obtained; this was denoted as solution B. Finally the solutions marked A and B were mixed together under stirring for 30 min and then sonicated at room temperature using a controllable serial-ultrasonic apparatus (Ultrasonic Processor, VCX 750, Korea). The mixture solution was allowed to cool to room temperature and filtered with 47 mm Whatman filter paper at pore size of 0.7 mm. The resultant powder was washed 3 times and transferred to dry even for 5 h at 90°C before heat treatment at 500°C for 2 h. The prepared samples were marked CoSe₂, CoSe₂-G, and CoSe₂-g-TiO₂, respectively.

The adsorption and degradation activity of the as-prepared CoSe₂-G-TiO₂ were evaluated under visible light (xenon 8 W, λ > 420 nm) using RhB as a standard dye. In the experimental process, 30 mg of the CoSe₂-G-TiO₂ catalytic sample was dissolved in a 100 ml solution of RhB (2.0 × 10⁻⁶ mol/L) and vigorously stirred for 2 h in a dark box until solution attained absorption-desorption equilibrium. Before switching on the light source a sample was collected from the solution and put in a centrifuge at 10000 rpm for the elimination of solid materials. During the degradation process, using a pipette, 2 mg mL (CoSe₂-G-TiO₂) solution was collected every 30 min and centrifuged for 10 min to remove any suspended solid. A spectrophotometer (TU1810, Puxi Company, China) was used to analyze the withdrawn sample determine the UV-vis spectra. All the samples were irradiated for 180 min to compare their catalytic performance.

The crystallinity of the CoSe₂-G-TiO₂ nanocomposite was investigated using monochromatic high intensity Cu Kα radiation (λ = 1.5406 A) in XRD (Shimadzu XD-D1). The surface morphology, including such aspects as the composition, and crystallization of the prepared simple, was studied using a scattering electron microscope (SEM) (JS M-5600 JEOL, Japan) and a transmission electron microscope (TEM) (JEOL, JEM-2010, and Japan). The change in absorbance was measured using a UV-Vis spectrophotometer (Optizen POP, Korea). Raman spectra of the prepared samples were studied using a spectrometer (Jasco, NRS-3100) with an excitation laser wavelength of 532.06 nm

We analyzed the crystallographic structure of the prepared sample through XRD techniques. Fig. 1 displays the XRD results for the CoSe₂-G-TiO₂ composites. The XRD pattern of CoSe₂-G has diffraction peaks at around 2θ of 27.5, 33.5, 36.6, 37, 41.5, 43.6, 47.5, 51.2, 58.8, 61.2 and 65.2^°, which can be indexed to the characteristic peaks of (011), (101), (111), (200), (210), (121), (211), (002), (031), (022) and (320) plane reflections with lattice parameter a = 4.850, b = 5.827, c = 3.628, and space group Pnnm (58) (JCPDS PDF#00-053-0449). The XRD pattern of TiO₂ has diffraction peaks at around 2θ of 25.5, 38.5, 48.6, 56.5, 64.6, 68.2 and 75.5^°, which can be indexed to the characteristic peaks of (101), (112), (200), (211), (204), (116), (220) and (216), which correspond to the anatase crystal phase (JCPDS PDF#00-021-1272). Moreover, TiO₂ (101) and graphene (002) peaks are located at the same 2θ values. This demonstrates the successful formation of the CoSe₂-G-TiO₂ ternary nanocomposite. The surface morphology of the TiO₂-G, CoSe₂-G and Cu₂Se-G-TiO₂ nanocomposites were examined by scanning electron microscopy (SEM). The images in Fig. 2(a-c) show the overall structure of the nanocomposites; it can be clearly seen that the graphene has a plate- like structure and that the TiO₂, and CoSe₂ nanoparticles are amalgamated into a group of Clusters. Fig. 2(a) shows that the TiO₂ nanoparticles are unevenly distributed on the graphene sheet. On the other hand, Fig. 2(b) shows that the CoSe₂ partilcles have spherical shapes and are partially agglomerated on the graphene sheet. The attachment of nanoparticles is very useful to overcome the interaction between functionalities on the graphene surface. Fig. 2(c) shows the structure of the CoSe₂-G-TiO₂ ternary nanocomposite. In this figure, it can be clearly observers that the TiO₂ has dispersed on the graphene sheet and on the supporting the CoSe₂ nanoparticles. Further transmission electron microscope (TEM) imaging is used to confirm the shape of the TiO₂ and CoSe₂ nanoparticles. TEM analysis, carried out to determine the microscopic structure information of the CoSe₂-G-TiO₂ ternary nanocomposite, is shown in Fig. 3(a-c). To explain show the surface morphology and the overall history of the graphene sheet and the CoSe₂-TiO₂ nanoparticles, the images in Fig. 3(a-c) were taken at different magnifications. Fig. 3(a-b) displays 200 nm magnification images of the CoSe₂-G-TiO₂ ternary nanocomposite. In these images, CoSe₂ nanoparticles can be seen to have almost spherical form; there is also a tube type of TiO₂ partially agglomerated on the graphene sheet. Moreover, the graphene sheet provides a large plate like structure; there are also tube type TiO₂ supported CoSe₂ nanoparticles on the graphene sheet. These interactions between the graphene sheet and the TiO₂, and CoSe₂ nanoparticles increase the electrocatalytic performance.²⁰⁾ Fig. 3(c) shows bunches of tube type TiO₂ and CoSe₂ nanoparticles attached to the graphene surface. This is clear evidence of the good contact and interaction among the TiO₂ and CoSe₂ nanoparticles and graphene. Thus, we can say that ultra-sonication synthesis of a ternary nanocomposite is superior to other techniques, and will be favorable for the enhancement of the photo catalytic efficiency of the CoSe-G-TiO₂ ternary nanocomposite. The Raman spectra technique provides structural information for carbon-based materials (e.g., number of layers and crystal structure). Raman spectroscopy of graphene is usually used to characterize two prominent peaks: the D and the G band. The D band is induced by defects that usually occur at 1350 cm⁻¹, while the G band is induced by sp² carbon bands observed at 1575 cm⁻¹. The Raman spectra of the CoSe₂-G-TiO₂ nanocomposites had a D band at 1350 cm⁻¹ and a G band at 1589 cm⁻¹ (Fig. 4). These two peaks, the G band and the D-band, provide information on the nature of the carbon-carbon bonds and the formation of defects caused by the distortion of the sp² bonded carbon atoms.²¹⁾ The characteristics of the peaks around 100 cm⁻¹ to 250 cm⁻¹ are attributed to the CoSe₂ crystal; peaks observe these, located at 392, 561 and 640 cm⁻¹, correspond to the B1g(1), A1g + B1g (2), and Eg (2) modes of anatase TiO₂.^22,23) The relative intensity of the D band compared to that of the G band (I_D/I_G) is calculated and found to be about approximately 1.002. These results reveal that the introduction of TiO₂ still affects the ratio and leads to an increased number of defects.^24-26)

The photocalytic activity of the CoSe₂-graphene nanocomposite using RhB as an organic dye under visible irradiation was investigated and results are shown in Fig. 5(a-c). Fig. 5 shows the adsorption capability of the CoSe₂-G-TiO₂ ternary nanocomposite. Fig. 5(a) shows a decline in the intensity of the electronic absorption spectra (λ_max) with different intervals of time, which demonstrates the degradation of RhB in the presence of CoSe₂-G-TiO₂ catalysts. The λ_max value of CoSe₂-G-TiO₂ was found to be 553 nm; the sample had degraded by approximately 85.2% at the end of 180 min. In the degradation experiment depends following factors, (1), the absorption by the CoSe₂-G-TiO₂ photocatalyst, (2), the fast charge transfer route and (3), graphene.

First two steps, to decompose the organic pollutants. While graphene acts an adsorption support material, and the dye absorbs molecules on the surface of graphene via π-π interaction, Fig. 5(a) shows that the concentration of RhB changes at different intervals of time, which shows the good absorption efficiency of RhB in the CoSe₂-G-TiO₂ nanocomposites. To achieve adsorption-desorption equilibrium, the prepared sample was kept in the dark for 30 min: after obtaining the adsorption-desorption equilibrium, the solution was kept in a closed box and visible light was turned on. The solution was irradiated for 30 min: every 30 min, the sample was withdrawn from the chamber for further syntheses. Then, the sample was centrifuged and the dye concentration was determined using a UV -visible spectrometer. The photo catalytic performances of the different products were evaluated under visible light irradiation. Fig. 5(b) shows the photocatalytic degradation of the different samples with different intervals of time. Fig. 5(b) shows that the photocatalytic degradation performance of the CoSe₂-GTiO₂ nanocomposites is higher than those of TiO₂, CoSe₂, graphene, and TiO₂-G. With the increase of the time, the intensity of the characteristic absorption band of Rh B (553 nm) significantly decreased: after 180 min, approximately 85.2% of the organic dye had been degraded. Furthermore, the kinetics of the degradation can be expressed as

where K_appt is the apparent rate constant, and. The kinetic plot and the apparent rate constant (K_aap) are shown in Fig. 5(c). It can be clearly seen that, with coupling of CoSe₂ and pure TiO₂ and graphene, the photocatalysis rate increased from 4.3 × 10⁻³ to 6.7 × 10⁻³ min, which is more than six times higher than that of pure TiO₂.