Enhanced Fenton-like catalytic performance of N-doped graphene quantum dots incorporated CuCo 2 O 4

In the present study, a novel nanocomposite based on CuCo 2 O 4 and N-doped graphene quantum dots (N-GQDs) as an iron-free heterogeneous Fenton-like catalyst was prepared by a two‒step solvothermal method. Surface morphology, chemical structure, crystal phase, surface area and the pore size distribution of the synthesized nanocomposite were characterized with field‒emission scanning electron microscopy (FE‒SEM), e nergy dispersive X-ray spectroscopy (EDS), Fourier Transform infrared spectroscopy (FT-IR), X-ray di ﬀ raction (XRD), X-ray photoelectron spectroscopy (XPS), high resolution transmittance electron microscopy (HRTEH) and N 2 adsorption-desorption analysis. The results showed that N-GQDs play a significant role in the enhancement of catalytic degradation of methylene blue (MB) as a model contaminant. CuCo 2 O 4 /N-GQDs nanocomposite exhibited higher degradation rate of about 47 times higher than that of pristine CuCo 2 O 4 . This promotion can be attributed to the higher surface area of CuCo 2 O 4 /N-GQDs nanocomposite compared to pristine CuCo 2 O 4 as well as the synergistic effects between N-GQDs and CuCo 2 O 4 . The effects of pH, catalyst mass and the initial concentration of H 2 O 2 on the catalytic degradation of MB were investigated using Box–Behnken design. Finally, the dominant reactive oxygen species generated in the degradation process were identified and the possible mechanism for the catalytic degradation of MB was purposed.


Introduction
One of the most common sources of pollutants discharged by different textile industries into the wastewater are organic dyes. Discharges of these organic pollutants with severe color, high toxicity and low biodegradability properties into wastewater without adequate treatment can cause serious health and environmental problems. There are a variety of processes available, depending on the kind of dye, for the removal of organic dyes by physical and chemical treatment technologies including coagulation, adsorption and membrane process. However, these conventional methods cannot be used effectively to remove dyes from wastewater. An efficient group of methods that have been successfully used to remove synthetic dyes from wastewater is advanced oxidation processes (AOPs).
Fenton method, one of the most popular advanced oxidation processes, is a catalytic process that utilizes hydrogen peroxide as an oxidant and dissolved ferrous salts as a homogeneous catalyst.
Iron-free heterogeneous Fenton-like systems for activating H2O2 in AOP based have been proposed 11 . Among all the catalysts, copper-and cobalt-containing catalysts have attracted more attention recently, because of high electrical conductivity and high electrochemical activity of copper and cobalt, respectively. However, Cu/Co-based catalysts exhibit weak catalytic activities, which limit their practical applications in Fenton-like processes. Therefore, one the most important challenges in Fenton-like processes is the construction of catalysts with high activity, better stability and reusability.
Graphene quantum dots (GQDs) are a new class of zero dimensional carbon materials that exhibit properties of graphene and carbon dots. Combining the structure of graphene with carbogenic origin and quantum nature of carbon dots causes GQDs to show novel chemical/physical properties. In the recent years, carbon materials such as carbon dots 24 , graphene, graphitic carbon nitride 26 and carbon quantum dots have been introduced into metal oxides to improve the catalytic, photocatalytic or photoelectrocatalytic performances of nanocomposites. These results suggest that GQDs play a significant role in the improvement of catalytic performance, and the combination of GQDs -metal oxide systems can be regarded as an effective strategy to construct more efficient catalysts.

Catalytic activity tests
The Catalytic activity of the catalysts for the degradation of MB was evaluated in a batch system. Typically, 50.0 mL of MB solution (5.0 mg/L) was added to a glass thermostated reactor (25.0 °C) together with 0.2 mL of 30% H2O2. Then, 5.0 mg of catalyst (CuCo2O4 or CuCo2O4/N-GQDs) was added to the solution and the mixture was stirred. To avoid any influence of ambient light, all reactions were carried out in a dark situation. The absorbance spectra were recorded in the wavelength range of 400-750 nm every 5 min until 35 min.

Experimental design and optimization
In order to obtain the best reaction conditions for the degradation of MB, the impacts of the catalyst mass (mg), pH and the initial concentration of H2O2 (mol/L) on the catalytic performance of CuCo2O4/N-GQDs for the degradation of MB were investigated. A three-factor, three levels Box-Behnken design was employed in order to achieve the highest degradation reaction rate constant (kobs). Catalyst mass, pH and the initial concentration of H2O2 were investigated as independent variables and rate constant of the MB degradation was considered as the dependent variable. The experimental dominions of these factors were established as preliminary experiments. The variables and levels of the Box-Behnken design model are given in Table S1 (Supplementary Information). The polynomial equation generated by this experimental design is as follows: where, y is the estimated response (dependent variable); xi and xj the coded variables of the independent variables; β0 the constant, βi the linear coefficient, βii the quadratic coefficient, βij the interactive coefficient and ε is a random error.

Spectroscopy and morphology characterization of nanocatalysts
The N-GQDs were synthesized by a facile hydrothermal method at 180°C. N-GQDs were hydrothermally synthesized by using citric acid and urea as carbon and nitrogen sources, respectively. In the hydrothermal conditions, citric acid and urea species react with each other and citric acid amide molecules are formed. Then, citric acid amide molecules are self-assembled and nanosheet structures are formed. N-doped GQDs with nitrogen-and oxygen-containing groups can be formed by intramolecular dehydration and deammoniation of neighbor citric acid amide molecules.
The absorbance -and photoluminescence (PL) spectra of N-GQDs are shown in Fig. 1.
As it is clear, N-GQDs absorption spectra shows two clear absorption bands at 234 and 337 nm and a broad peak at around 600 nm. The origins of the peaks at 234 nm and 337 nm are due to π→ π* transition of C=C and n→π* transition of the C=O, respectively 28, 31 . The inset in Fig. 1 shows the photograph of N-GQDs solution under sun-light, indicating the intense absorption of light in visible region, consistent with the absorption spectra measurements. Similar to the most fluorescent carbon-based nanomaterials, the solution of N-GQDs emits blue light (λPL=450 nm) when excited with 360 nm (Fig. 1). The origins of the PL spectra of N-GQDs species is probably due to the effects of size and the surface functional groups in N-GQDs 32 .     The surface chemical composition of the synthesized CuCo2O4/N-GQDs nanocomposite was characterized by XPS as shown in Fig. 6. A full XPS spectrum (survey spectrum) of CuCo2O4/N-GQDs surface in Fig. 6a indicates the presence of Co, O, Cu, N and C elements in CuCo2O4/N-GQDs nanocomposite. Two major peaks with binding energies at 796.7 eV and 781.1 eV are observed for Co 2p curve at Fig. 6b, corresponding to the Co 2p1/2 and Co 2p3/2 spin-orbit peaks, respectively 29, 42 . The Co 2p3/2 spectrum was fitted with two spin orbit doublets at 781.36 eV assigned to Co 2+ components at tetrahedral sites and 779.9 eV assigned to Co 3+ components at octahedral sites, and three satellite peaks I, II and III assigned to Co 2+ and Co 3+ components and are all in agreement with the results reported in literature 29,43 . For Co 2p1/2 XPS spectrum, the similar fitting was also done. The ratio of Co 2+ to Co 3+ ions in CuCo2O4/N-GQDs nanocomposite was calculated from deconvoluted data and obtained to be 1.7 and is in agreement with literature 29, 42, . The Co 2p3/2-2p1/2 spin-orbit splitting characteristic of the Co 2+ and Co 3+ species are almost identical for both oxides and are in agreement with the results reported in literature 29,42,43 . Fig. 6c shows the Cu 2p XPS spectrum for CuCo2O4/N-GQDs nanocomposite. The Cu 2p spectrum is deconvoluted into several components due to Cu oxides and was fitted with three spin orbit doublets assigned to Cu + , Cu 2+ and Cu 3+ and two satellite peaks I and II. The components at 932.5 eV, 934.6 eV and 937.0 eV correspond to Cu + at the tetrahedral sites, Cu 2+ and Cu 3+ at the octahedral sites, respectively and are in agreement with the results reported in literature 42 . Cu + ions is seems to be due to the X-ray induced reduction of Cu 2+ ions 44 . The presence of paramagnetic Cu 2+ species is revealed by the intense satellite peaks at 943.9 and 954.2 eV. The most prominent peak in Cu 2p spectrum is at 934.6 eV, which indicates the majority of Cu 2+ at more energetically stable structure of octahedral sites 29,44 . Based on the deconvoluted Co and Cu curve regions, the ratio of Co 2+ /Co 3+ and Co 2+ /Cu 2+ ions in CuCo2O4/N-GQDs are 1.7 and 12.2, respectively which shows that Co 2+ ions are partially substituted by Cu 2+ ions.

Fig. 6
Fig. 6d shows XPS window of C 1s core level. This spectrum exhibits two main peaks centered at 284.8 (named as A) and 288.6 eV (named as B), which are attributed to carboncarbon and carbon-oxygen bonds, respectively. The A peak has been deconvoluted to two peaks with binding energy (BE) of 284.8 and 286.1 eV which are attributed to C-C and C=C bonds of N-GQDs, respectively. The peak B has been curve fitted by a peak located at BE value of 288.7 which is assigned to carbon-oxygen (C-O) bond. The presence of peak B (carbon-oxygen bonds) with remarkable intensity as compared to peak A (carbon-carbon bonds) with the ratio of 0.57 revealed that the chemical bonds exist between C atoms of N-GQDs and O atoms of CuCo2O4.
Chemical bond formation between CuCo2O4 and N-GQDs molecules can also be verified through investigation of O 1s XPS window (Fig. 6e).
The deconvolution peaks of the O 1s spectrum are also resolved into three components, centered at 529.8, 531.6 and 533.3 eV, respectively (Fig. 6e) As it is clear from the results ( Table 1), SBET of CuCo2O4/NGQDs nanocomposite (105.67 m 2 g -1 ) was significantly increased (~43%) by the incorporation of N-GQDs into the CuCo2O4 (with SBET of 74.29 m 2 g -1 ). The result of BJH analysis of CuCo2O4 in Fig. 7b shows that it possesses pores in the range of 1.70 to 10 nm, and strong peaks are observed at the pore diameters around 1.85, 4.61 and 6.05 nm. The pore-size distribution of CuCo2O4/N-GQDs was also obtained from the BJH analysis and showed that most of the pores were in the range of 1.21 to 20 nm, and strong peaks are observed at pore diameters around 1.21, 4.61 and 6.05, 9.22 and 18.93 nm.

Fenton catalytic activity of prepared catalysts
The catalytic activities of CuCo2O4 and CuCo2O4/N-GQDs catalysts were compared by plotting Ct/C0 ratio as a function of reaction time (Fig. 8). As it is clear, CuCo2O4/N-GQDs nanocomposite performed much more efficiently than pristine CuCo2O4. The degradation efficiency of CuCo2O4 in MB degradation by hydrogen peroxide after 35 min is below 10%.
Amazingly, when CuCo2O4/N-GQDs was used as a catalyst in the reaction mixture, 96.4 % of MB was degraded in 35 min. The observed rate constant of the catalytic degradation of MB (kobs, min -1 ) in the presence of the CuCo2O4/N-GQDs as catalyst is 0.094 min -1 (Fig. 8d), which is 47 times higher than that of CuCo2O4 (kobs=0.002 min -1 ). The higher catalytic activity of the CuCo2O4/N-GQDs nanocomposite compared to CuCo2O4 can be attributed to the higher surface area that provides more active sites for dye degradation. However, the increase in catalytic activity of CuCo2O4/N-GQDs is not only attributed to its higher surface area, because as discussed, SBET of CuCo2O4/NGQDs nanocomposite was increased about 43% (or 1.43 times higher) by the incorporation of N-GQDs into the CuCo2O4, while its catalytic activity is about 47 times higher than that of pristine CuCo2O4.

Fig. 8
The main reason for increase in the catalytic activity of CuCo2O4/N-GQDs nanocomposite compared to CuCo2O4 can be attributed to the enhanced electron transport kinetics between Ndoped GQDs and metal ions including Cu and Co. This leads to these metal ions to be tightly attached to the surface of N-GQDs, which enhances the electron transport kinetics between N- GQDs and metal ions. To understand the synergistic effects of N-GQDs and CuCo2O4, an in-situ experiment was designed.

Implication of active species in Fenton process
The most common active species in Fenton-like processes are hydroxyl radical ( Based on these evidences, TBA and CHCl3 were used as scavengers of • OH and • O2−, respectively. As shown in Fig. 9a, the MB degradation efficiency in 35 min reached above 96% in the absence of the scavengers. When TBA was added (0.5 mmol/L and 3.0 mmol/L) to the reaction vessel including MB, the degradation efficiencies within 35 min decreased to 45.0% and 35.0%, respectively. Also, when KI was added to the reaction vessel, the degradation efficiencies decreased from 96.4% to 37.0% and 26.0%, in the presence of KI from 0 to 0.5 mmol/L and 3 mmol/L, respectively (Fig.9b). These results demonstrate that MB was mainly degraded by the attack of • HO radicals ( • OHfree and • OHads) and • HO species (especially • OHads) is the dominant reactive oxygen species in the CuCo2O4/N-GQDs system.

Fig. 9
To investigate the role of superoxide radical anions during the MB degradation, chloroform (CHCl3) was applied as an • O2− scavenger. As can be seen from Fig. 9c, the degradation efficiency of MB during the reaction time (35 min) was decreased from 96.4% to 86% and 52% as the CHCl3 concentration was increased from 0 to 0.5 mmol/L and 3.0 mmol/L, respectively.
The decrease in the degradation efficiency of MB with increasing CHCl3 suggested that, although • O2− was formed in the system, but its role in promoting the degradation of MB is limited relative to hydroxyl radicals, • HO.
To further confirm the existence of • HO radicals, hydroxyl radicals generated on a CuCo2O4/N-GQDs nanocomposite were identified by a photoluminescence (PL) spectroscopy using terephthalic acid (TA) as a probe molecule, which reacted with ·OH radicals to produce highly fluorescent product, 2-hydroxyterephthalic acid (TAOH). The maximum PL intensity of TAOH at around 425 nm and is proportion to the amount of produced ·OH radicals. Fig. S1 (Supplementary Information) shows the PL spectra of the TA solution after reaction with H2O2 for 2 and 5 min over CuCo2O4/N-GQDs. As shown in Fig. S1 (Supplementary Information), PL intensity increased with increasing reaction time, which indicated that the ·OH radicals formed gradually in the presence of nanocatalyst and these radicals are the main active species in the degradation process of MB.

Optimization of the degradation of MB by CuCo2O4/N-GQDs
The optimization step of the MB degradation onto CuCo2O4/N-GQDs was performed using a Box-Behnken design. According to the reports of previous studies, the pseudo first-order kinetic model is used extensively in the heterogeneous Fenton-like processes. The pseudo first-order  ( 3) where, as it was noted in Table 1, x1, x2 and x3 denote coded values of the pH of experiment, catalyst mass and initial concentration of H2O2, respectively.
Results of analysis of variance (  Fig. 11 illustrates the response surface plots for the relationship between pH, catalyst mass and initial concentration of H2O2 on the degradation of MB onto CuCo2O4/N-GQDs. In these plots, the response is mapped against two experimental variables while the remaining are fixed at their central levels. Fig. 11a shows that the degradation rate constant of MB increases with increase in catalyst mass and H2O2 concentration. The increase in kobs is due to the increase in hydroxyl radical formation when H2O2 concentration and catalyst amount is increased. Fig.   11b presents the influence of the initial concentration of H2O2 and the pH values on the degradation rate constant of MB. The results showed that the degradation rate constant of MB increases with increase in pH and initial concentration of H2O2.
In contrast to iron-based catalysts which higher degradation efficiencies is achieved at acidic conditions, the proposed catalyst work in neutral and alkaline conditions. The results clearly reveal that Fenton reaction catalyzed by CuCo2O4/N-GQDs shows a remarkable advantage, because a large volume of pollutants which are discharged in environment have neutral or basic pH values. The similar results were also observed for other cobalt-based catalysts.

Conclusions
In conclusion, this work describes the synthesis of CuCo2O4/N-GQDs nanocomposite by two-