Hollow Dodecahedra Graphene Oxide- Cuprous Oxide Nanocomposites With Effective Photocatalytic and Bactericidal Activity

In this study, a kind of graphene oxide-cuprous oxide (GO-Cu2O) nanocomposites was fabricated with different morphologies to serve as a photocatalytic material for the degradation of organic/inorganic dyes under visible light and the bactericidal effect against pathogenic bacteria. The GO-Cu2O was prepared with solid cube and hollow dodecahedra morphologies through in-situ synthesis, and characterized by scanning electron microscopy (SEM), transmission electron microscope (TEM), X-ray diffraction (XRD), Raman, Ultraviolet and visible spectrophotometry (UV/vis), and Fourier transform infrared spectroscopy. In comparison with cubic GO-Cu2O, the absorption and degradation efficiency of the GO-Cu2O dodecahedra (GCD) composite in Methyl orange (MO), Rhodamine B (RhB), and phenol was higher owning to the more active sites for the simultaneous dye and light absorption of hollow structure. The antibacterial effect of the GO-Cu2O dodecahedra was examined by the flat colony counting method with an excellent bactericidal effect against pathogenic bacteria. The possible mechanism for the preparation of GCD possessing the enhancement of the visible-light photocatalytic and antibacterial efficiencies were also investigated.


INTRODUCTION
Water pollution causes great damage to ecosystems, human health, as well as the sustainable economic and social development because the pollutant complex, along with bacteria, cause difficulty in decontamination by conventional water treatment processes (Schwarzenbach et al., 2010;Wang and Yang, 2016). Hence, developing an effective and facile way to degrade pollutants has become an active area in environmental research. Recently, inorganic nanomaterials have attracted numerous attentions because of their controllable shapes and sizes, as well as their effective photocatalytic activities, such as those in metal oxide semiconductors (e.g. TiO 2 , ZnO) or narrow band gap semiconductors (e.g. Ag 3 PO 4 ) (Shao et al., 2019;Zhang et al., 2020).
Although these inorganic materials exhibit promising photocatalytic activities, there are still several problems that need to be overcome, for instance, relatively poor lightharvesting abilities in the visible region, the use of toxic or harmful chemicals, or poor charge separation and transport. Cuprous oxide (Cu 2 O) is a promising metal oxide material in the application of photocatalysis because it is a p-type semiconductor (C hole > C electron ) with a small band gap (E g 2.17 eV). Recently, numerous efforts have been devoted to synthesize Cu 2 O with different morphologies such as nanowires, octahedra, cuboctahedra, and nanocubes (Hua et al., 2011;Deng et al., 2012;Hou et al., 2013;Hong et al., 2014). Among them, a hollow structure has intrinsic advantages in photocatalysis applications Xiao et al. (2019), such as enhancing light harvesting Li et al. (2007), Wang et al. (2012), Wu et al. (2013), Dinh et al. (2014), Qi et al. (2014), promoting the synergistic effects of light scattering and localized surface plasmon resonance (LSPR) Zhang et al. (2014), Shi et al. (2016), reducing charge recombination Marschall (2014), Li et al. (2015), and accelerating surface reactions due to a high surface area (Sun et al., 2003;Vaughn and Schaak, 2012;Wang et al., 2017). Furthermore, due to a large carbon sheet structure, graphene oxide (GO) was also introduced in a hybrid with Cu 2 O to effectively increase the adsorption sites and improve the transfer of electrons between the materials for the inhibition of hole and electron recombination. The addition of carbon based 2D materials during the in-situ preparation of the nanoparticles may improve the photocatalytic performance effectively (Khan et al., 2015;Khan et al., 2016a;Khan et al., 2016b;Khan et al., 2017;Ahmed and Haider, 2018;Khan et al., 2018). Lee and his group synthesized Ag-Cu 2 O together with graphene oxide for the enhanced photocatalytic performance (Sharma et al., 2018). SunilMeti et al. reported zinc oxide nanocomposites wrapped with reduced graphene oxide to enhance the photocatalytic activity (Meti et al., 2018).
Herein, the nontoxic and novel visible-light-driven GO-Cu 2 O composite was used as an inspiring photocatalytic material to address the aforementioned problems in water pollution. In this work, aqueous solutions of copper salt, alkali, surfactant, and reductant were used to prepare hollow dodecahedral Cu 2 O through an in-situ synthesis process. The GO sheet was added through an electrostatic reaction of negatively charged GO and Cu ions and leaving a final modified GO-Cu 2 O hollow dodecahedral (GCD) structure. The solid cubic GO-Cu 2 O structures were also fabricated by systematically changing the reductant amount for comparison. After the successful preparation of the crystal components, the structures were characterized and confirmed. The photocatalytic performance of the GO-Cu 2 O in different dyes under visible light were investigated and compared, while the antibacterial performance was evaluated by a flat colony counting method and TEM. In general, the as-prepared GCD enhanced lightharvesting, separated the excited e − -h + pairs, and promoted charge transfer. The increased reactive oxygen species generated from visible irradiation made the oxidation of the organic pollutant and elimination of the bacteria possible, as shown in Scheme 1.
The characterizations, photocatalytic activity, and bactericidal activity tests were described in the supporting information.

Synthesis of Graphene Oxide-Cuprous Oxide Nanocomposites With Different Morphologies
In a typical synthesis, different volumes of water were used to obtain the final 1 L solution (Deng et al., 2012;Huang et al., 2012). Flasks containing CuCl 2 solution (0.1 M, 50 ml) and sodium dodecyl sulfate (8.7 g, SDS; 100%) were placed in a water bath and kept at 32°C. Then, NaOH (1.0 M, 18 ml) was added in dropwise with vigorous stirring. A Cu(OH) 2 precipitate was formed and the color of the solution changed from dark blue to light blue. Different amounts of NH 2 OHHCl (solid cube: 40 ml; hollow dodecahedra: 240 ml) were then poured within 5 s and left to cool down to RT. The final products were then dried and obtained after washing several times with DI water/ethanol. The GCD nanocomposites were synthesized via a similar route as the SCHEME 1 | Possible photocatalytic and bactericidal mechanism of GCD.
Frontiers in Chemistry | www.frontiersin.org September 2021 | Volume 9 | Article 755836 2 Cu 2 O, except that 40 mg GO was sonicated into the deionized water at the beginning.

RESULTS AND DISCUSSION
With the addition of the NH 2 OHHCl reductant during the in-situ process, the Cu(OH) 2 was the first nucleation seed and then the Cu 2 O nanocrystals grew on the surface to form various morphologies. The higher concentration of the NH 2 OHHCl increased the growth rate of the Cu 2 O and changed the morphology from a cube to a dodecahedron, while the HCl from the reductant etched the Cu 2 O to form the final hollow structure (Kuo and Huang, 2008).
The morphologies of the samples were characterized by SEM. Figure 1A showed that the synthesized CC possessed a cube morphology of about 100-500 nm in size. From the SEM image in Figure 1B, CD displayed a dodecahedron morphology, and the average diameter was about 300 nm, most of which were broken or full of holes. After the addition of GO to the samples, the ionic interaction between the Cu cations and GO − anions modified the Cu 2 O morphology. The SEM image of the GCC composites ( Figure 1C) highlighted the presence of cube-like Cu 2 O polyhedrons with a uniform size of less than 200 nm, most of which were wrapped by GO sheets. The addition of GO also initiated a minor change of the morphology. The GCD in Figure 1D illustrated that the dodecahedron-like particles have an average diameter of about 100-200 nm.
The solid cubes and hollow dodecahedra were also characterized by TEM. The size of the CC particles coated with the GO sheets ( Figure 2A) was estimated to be between 150-200 nm, which was consistent with the SEM image of the GCC particles in Figure 1C. Figure 2B was a typical TEM image of CD covered with a small quantity of GO sheets, from which the hollow structure and the presence of GO were clear.
After determining the composites' morphologies by SEM and TEM, the XRD spectra were utilized to confirm the prepared crystal components. According to the JCPDs, the peaks at (110), (111), (200), (220), (311), and (222) were assigned to the standard cubic structure Cu 2 O (No. 03-0898) (Wang et al., 2002;Zhang et al., 2007). No other peaks were detected, such as CuO and Cu cupric oxide, demonstrating the purity of the as-obtained products. There was no clear peak for GO observed in the XRD pattern, due to the small amount and low diffraction intensity. The presence of GO was confirmed by Raman and TEM. The (200) diffraction peak of the Cu 2 O cube was stronger than the other diffraction peaks, indicating a high proportion of (100) facets. Similar results were also confirmed for the (220) diffraction peak of the rhombic dodecahedral Cu 2 O nanocrystals, implying a high proportion of (110) crystal planes. Also, Raman spectroscopy was used to measure the vibrations of the sp 2hybridized carbon atoms for the confirmation of the presence of GO. Hence, it was observed from Figure 3B that the Raman spectra of the as-synthesized samples contained the D peak (1,343cm −1 , disorder-activated Raman mode) and G peak (1,588 cm −1 , sp 2 hybridized carbon) which were assigned to GO Yang et al., 2009; Krishnamoorthy    Figure 4A displays the changes in degradation for MO using GCD. It was observed from Figure 4A that after the ultrasonic mixing of the solution for 30 min, the solution was absorbed to some degree. The absorption capacity of pure CD was about 80%, which was higher than the GCD composites. The inset in Figure 4A indicated the degradation efficiency of the GCD increased as the amount of GO. When 40 mg of GO was introduced into the pure Cu 2 O, the corresponding composite showed the highest photocatalytic activities of near 100% within 60 min.
However, further addition of GO led to a decrease in the photocatalytic activity of the composites. The possible reason was that too much GO may fully wrap the GCD to eradicate the light irradiation.
To further understand the photocatalytic process, the visible light photocatalytic experiments shown in Figures 4B-D were investigated using MO, RhB, and phenol as target dyes. As shown in Figure 4B, the curve of MO using the GCD was the lowest, whereas the best performance for degrading MO and phenol ( Figures 4B, D) was the pure Cu 2 O cube. The efficiencies of absorption and degradation of MO, RhB, and phenol ( Figures 4B-D) using the GCD composite were approximately 100%, 75%, and 5% respectively, within 120 min. While the efficiencies of MO, RhB, and phenol using GCC were 50%, 60%, and 6%. Hence, it was suggested that the GCD composite exhibited excellent photocatalytic performance for colored organic dyes. However, the performance of colorless dyes was very poor. This demonstrated that the photocatalytic effect of the GO-Cu 2 O nanoparticles was mainly due to the adsorption of the colored dyes instead of degradation of organic dyes, which was the limitation of the synthesized photocatalytic materials.
Using the experimental results, a possible photocatalytic mechanism was deduced. Generally, the mechanism of the GO-Cu 2 O catalytic activity was similar to other photocatalytic materials. The mechanism consisted of the absorption and degradation of the dyes and the absorption  Frontiers in Chemistry | www.frontiersin.org September 2021 | Volume 9 | Article 755836 was greater than the degradation in the photocatalytic performance. The possible photocatalytic absorption mechanism was in large part due to the structure of the Cu 2 O. Due to the advantage of a hollow structure, the GCD nanocomposites had the highest absorption. In addition, GO sheets offered more active adsorption sites, which also improved the adsorption of dyes. After the investigation of the photocatalytic performance of GCD, the flat colony counting method was also introduced to examine the specific bactericidal effect (Dahle et al., 2004;Yang et al., 2015;Choudhry, 2016). As shown in Figure 5, the bacterial number was calculated after the addition of GCD at 0, 24, and 48 h. Specifically, the original concentration of the pathogenic bacteria was approximately 1*10 7 CFU/ml. After the incubation of the bacteria for a period of 24 and 48 h without the GCD, an obvious increase was observed in the E. coli and S. typhi at 1*10 7.5 CFU/ml. The amount of S. aureous and P. aeruginosa remained the same. However, with the treatment of GCD at 24 h, the number of bacteria decreased from 1*10 7.5 to 1*10 4.5 CFU/ml with an antibacterial rate of more than 99.9%. After the incubation time was prolonged to 48 h, the number of the S. aureous was reduced to 1*10 2.5 CFU/ml, and the specific bactericidal rate for all the bacteria was summarized in Table 1. Considering the bactericidal experiment, it was concluded that the prepared GCD had a broad-spectrum antibacterial activity toward pathogenic bacteria. To discuss the mechanism of the bactericidal effect, the morphological changes of E. coli and S. aureus were investigated by TEM before/after the addition of GCD. Figure 6 (a,b) showed the original E. coli rod-like structures with a size of 1 μm × 0.5 μm, while the S. aureus has a spherical structure with a diameter of 0.5 μm in the absence of GCD. However, after the treatment of samples that involved the light irradiation of the GCD, ROS was generated to interact with the membrane, making the cytoplasm flow out of the bacteria, and finally killed them. Also, other research considered that the presence of Cu ions in the GCD reacted with the oxygen to produce ROS through the Fenton reaction for further photocatalytic and bactericidal performance (Touati, 2000;El Saeed et al., 2016;Shen et al., 2020). Generally, GCD acted as a kind of bifunctional material for applications in photocatalysis and bactericide.

CONCLUSION
In summary, an effective in-situ synthesis to produce different morphologies of GO-Cu 2 O was demonstrated. After characterizing with SEM, TEM, XRD, Raman spectroscopy, UV/vis spectroscopy, and FT-IR spectroscopy, the degradation performance of Cu 2 O and GO-Cu 2 O for different dyes under visible light was measured. Antibacterial experiments were also investigated against pathogenic bacteria. The presence of GO along with the hollow structure created a synergistic effect that increased the photo harvesting and facilitated the electron transfer to generate more ROS for the enhancement of the photocatalytic and bactericidal performances. Herein, this work offers new insights into the facile synthesis of GO-based nanocomposites for the applications of photocatalytic degradation and sterilization of wastewater pollutants using visible light.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

AUTHOR CONTRIBUTIONS
ZS conceived of the presented idea, YY and HS carried out the experiment, JZ carried out some experiments and analyzed the data during the revision of the manuscript, XT and YL verified the analytical methods, ZJ, PW, JQ were involved in planning and supervised the work.