A Novel Au@Cu2O-Ag Ternary Nanocomposite with Highly Efficient Catalytic Performance: Towards Rapid Reduction of Methyl Orange Under Dark Condition

Au@Cu2O core-shell nanocomposites (NCs) were synthesized by reducing copper nitrate on Au colloids with hydrazine. The thickness of the Cu2O shells could be varied by adjusting the molar ratios of Au: Cu. The results showed that the thickness of Cu2O shells played a crucial role in the catalytic activity of Au@Cu2O NCs under dark condition. The Au@Cu2O-Ag ternary NCs were further prepared by a simple galvanic replacement reaction method. Moreover, the surface features were revealed by TEM, XRD, XPS, and UV–Vis techniques. Compared with Au@Cu2O NCs, the ternary Au@Cu2O-Ag NCs had an excellent catalytic performance. The degradation of methyl orange (MO) catalyzed by Au@Cu2O-Ag NCs was achieved within 4 min. The mechanism study proved that the synergistic effects of Au@Cu2O-Ag NCs and sodium borohydride facilitated the degradation of MO. Hence, the designed Au@Cu2O-Ag NCs with high catalytic efficiency and good stability are expected to be the ideal environmental nanocatalysts for the degradation of dye pollutants in wastewater.


Introduction
Synthetic organic dyes and natural pigments with considerable coloring capacity are widely applied in paper, plastics, paints, cosmetics, food, textile, mining, and printing industries [1][2][3]. Azo dyes account for 60-70% of all dyes [4]. However, the direct discharge of wastewater containing azo dyes not only destroys the ecological environment but also poses a serious threat to human life and health, as reported in previous literature [4]. For example, when individuals are exposed to methyl orange (MO), a typical representative of azo dyes, it has been shown to cause physical discomfort such as increased heart rate, vomiting, and shock [5,6]. Thus, removal and degradation of azo dyes are critically important to prevent environmental pollution and protect human health [7,8].
Nowadays, a series of post/pre-treatment methods have been proposed to remove the azo dyes from the wastewater [1,[9][10][11]. Porous/activated carbons have been applied as commercial adsorbents for a galvanic replacement method and Ag nanocrystals were attached to the surfaces of Au@Cu 2 O to achieve Au@Cu 2 O-Ag NCs. Au@Cu 2 O-Ag NCs were also applied to the catalytic reduction of MO under dark condition. Furthermore, the corresponding catalytic mechanism was studied. This work introduces a new catalyst for the rapid and efficient wastewater treatment in the absence of light.

Synthesis of Au@Cu 2 O NCs
Step I: Au colloid solution was prepared as described in the literature [40]. 100 mL of 0.01% HAuCl 4 ·4H 2 O solution and 4 mL of 1% C 6 H 5 Na 3 O 7 ·2H 2 O solution were heated under reflux for 10 min to achieve the Au colloid solution.
Step II: 0.5 g of PVP was dissolved in 50 mL of 0.005 M Cu(NO 3 ) 2 solution under stirring at 450 rpm. Then, a certain amount of (1, 3, 5, 7, 9, and 11 mL) Au colloid solution was added to the mixture. Subsequently, N 2 H 4 ·H 2 O (34 µL, 17.5 wt%) was immediately introduced, and the mixture was stirred for 2 min. Then, the product obtained by centrifugation was washed with anhydrous ethanol and deionized water. Finally, Au@Cu 2 O NCs with different shell thickness were acquired, which were named as AC-1 NCs, AC-3 NCs, AC-5 NCs, AC-7 NCs, AC-9 NCs, and AC-11 NCs, respectively. The same procedure was used to prepare Cu 2 O nanocrystals but the Au colloid solution was not added.

Synthesis of Au@Cu 2 O-Ag NCs
The AC-1 NCs with the thickest Cu 2 O shell were chosen to deposit Ag nanocrystals on their surfaces. The AC-1 NCs were dispersed in 35 mL of deionized water. 200 µL of 0.004 M AgNO 3 solution was then added to the solution with stirring. After stirring for 10 min, the prepared Au@Cu 2 O-Ag NCs were washed several times with anhydrous ethanol and deionized water. Au@Cu 2 O-Ag NCs were named as AC-1-Ag NCs.

Catalytic Activity Measurement
All the catalytic activity measurements were performed in the absence of light. 2.5 mg of catalyst was added to 2.5 mL of deionized water. The catalyst was uniformly dispersed in deionized water by ultrasonication. 100 mL of 10 mg/L MO solution and 9 mL of 3.78 mg/L NaBH 4 solution were added to the beaker. The catalyst was added for catalytic reduction and magnetic stirring was maintained during the reaction to ensure uniform dispersion. During the catalytic reduction, 2 mL of solution was sampled from the suspended catalyst nanoparticles for measurement at certain time intervals. The catalytic process was monitored by measuring the change in absorbance of MO at 464 nm by ultraviolet-visible (UV-Vis) absorption spectroscopy. To investigate the repeatability, the samples were separated by centrifugation and then washed several times. Then, the same procedure for catalytic reduction of MO solution was repeated for four cycles. Figure 1 shows the preparation process and catalytic application for degradation of MO of AC-1-Ag NCs with assistance of NaBH 4 .

Characterization
Transmission electron microscopy (TEM) images were obtained by using a Hitachi H-800 transmission electron microscope (JEOL Ltd., Tokyo, Japan) operating at an accelerating voltage of 200 kV. UV-Vis absorbance spectra were recorded on a Shimadzu UV 3600 spectrophotometer (Shimadzu Corporation, Tokyo, Japan) in the range of 350-800 nm. X-ray powder diffraction (XRD) analysis was conducted using a Rigaku D/MAX 3C X-ray diffractometer (Rigaku Corporation, Tokyo, Japan) with Cu Kα radiation (λ = 1.5418 Å). X-ray photoelectron spectra (XPS) analysis was performed using a Thermo Scientific ESACLAB 250 Xi A1440 system (Thermo Fisher Scientific, Waltham, MA, USA).  Figure 2 shows the TEM images of Au@Cu2O NCs with different Cu2O shell thicknesses. It is shown in Figure S1 that the average diameter of the Au core with quasi-spherical shape is about 13 nm. When the added volume of Au colloid solution is 1.00, 3.00, 5.00, and 7.00 mL, as depicted in Figure 2a-d, the Cu2O shell thickness of AC-1 NCs, AC-3 NCs, AC-5 NCs, and AC-7 NCs is about 82, 71, 53, and 30 nm, respectively. The surfaces of Au nanocrystals are coated by Cu2O to form a core-shell structure because the citrate ligands existing on the surfaces of the Au nanocrystals act as the binder and promote the contact of Cu 2+ ions with the Au surfaces during the reduction process. The reduction of Cu 2+ ions and the growth of Cu2O take place on the surfaces of Au cores [41]. As a result, Au@Cu2O NCs with core-shell structures are achieved. All of the four samples exhibit a relatively uniform spherical morphology and have good dispersion, which ensures that most of core-shell nanocrystals are involved in catalytic reactions. In addition, as different volumes of Au colloid solution are introduced into the Cu 2+ reaction mixture with a fixed initial concentration, with the increase of molar ratio of Au to Cu 2+ , the Cu2O shell thickness attached to Au cores decreases. Similar results were found by other research groups [42,43]. Notably, when the volume of Au colloid solution is further increased to 9.00 mL, Cu2O shells on the surfaces of Au nanocrystals almost disappear. As for the sample of AC-11 NCs, this abnormal phenomenon is more evident. A large number of pure Au nanocrystals without Cu2O shells are observed, as depicted in Figure 2e,f. This is most probably because Au nanocrystals reach the supersaturation point so that there is not enough number of Cu 2+ ions in the reaction system to form the Cu2O shells on the surfaces of Au nanocrystals [26].

Results and Discussion
A selected area electron diffraction (SAED) pattern of AC-1 NCs is shown in Figure 3a Figure S1 that the average diameter of the Au core with quasi-spherical shape is about 13 nm. When the added volume of Au colloid solution is 1.00, 3.00, 5.00, and 7.00 mL, as depicted in Figure 2a-d, the Cu 2 O shell thickness of AC-1 NCs, AC-3 NCs, AC-5 NCs, and AC-7 NCs is about 82, 71, 53, and 30 nm, respectively. The surfaces of Au nanocrystals are coated by Cu 2 O to form a core-shell structure because the citrate ligands existing on the surfaces of the Au nanocrystals act as the binder and promote the contact of Cu 2+ ions with the Au surfaces during the reduction process. The reduction of Cu 2+ ions and the growth of Cu 2 O take place on the surfaces of Au cores [41]. As a result, Au@Cu 2 O NCs with core-shell structures are achieved. All of the four samples exhibit a relatively uniform spherical morphology and have good dispersion, which ensures that most of core-shell nanocrystals are involved in catalytic reactions. In addition, as different volumes of Au colloid solution are introduced into the Cu 2+ reaction mixture with a fixed initial concentration, with the increase of molar ratio of Au to Cu 2+ , the Cu 2 O shell thickness attached to Au cores decreases. Similar results were found by other research groups [42,43]. Notably, when the volume of Au colloid solution is further increased to 9.00 mL, Cu 2 O shells on the surfaces of Au nanocrystals almost disappear. As for the sample of AC-11 NCs, this abnormal phenomenon is more evident. A large number of pure Au nanocrystals without Cu 2 O shells are observed, as depicted in Figure 2e,f. This is most probably because Au nanocrystals reach the supersaturation point so that there is not enough number of Cu 2+ ions in the reaction system to form the Cu 2 O shells on the surfaces of Au nanocrystals [26].

Results and Discussion
A selected area electron diffraction (SAED) pattern of AC-1 NCs is shown in Figure 3a.  [44,45]. Both of the SAED and XRD results confirm that the prepared AC-1 NCs consist of Cu 2 O and Au. Figure 3b shows the high-angle annular dark-field scanning TEM (HAADF-STEM) images and the corresponding energy-dispersive spectroscopy (EDS) elemental mapping images of AC-1 NCs. It is easily observed that the outer Cu 2 O completely cover the Au cores because the image contrast is brighter for the heavier element. In addition, EDS mapping images of Au, Cu, and O elements show that the Au element is located in the center of the core-shell structure, and Cu and O elements are uniformly distributed on the surfaces of Au nanocrystals [46]. are shown in Figure S2. As for AC-1 NCs, apart from the characteristic diffraction peaks of Cu2O (JCPDS card no. 05-0667), the diffraction peaks appearing at 38.18, 64.57, and 77.54° are corresponding to (111), (220), and (311) of Au (JCPDS card no. 04-0784) [44,45]. Both of the SAED and XRD results confirm that the prepared AC-1 NCs consist of Cu2O and Au. Figure 3b shows the high-angle annular dark-field scanning TEM (HAADF-STEM) images and the corresponding energy-dispersive spectroscopy (EDS) elemental mapping images of AC-1 NCs. It is easily observed that the outer Cu2O completely cover the Au cores because the image contrast is brighter for the heavier element. In addition, EDS mapping images of Au, Cu, and O elements show that the Au element is located in the center of the core-shell structure, and Cu and O elements are uniformly distributed on the surfaces of Au nanocrystals [46].   Figure 4 shows the UV-Vis absorption spectra of Au colloid solution, Cu2O nanocrystals, and Au@Cu2O NCs with different shell thicknesses. The pure Au colloid solution exhibits a prominent SPR absorption peak at 518 nm. A strong absorption peak of pure Cu2O nanocrystals appears at 464 nm. As for Au@Cu2O NCs, two broad absorption peaks are observed. The absorption peak located at 526-584 nm may be attributed to plasmon resonance of Au cores, while the peak at 406-432 nm is ascribed to Cu2O shells. The SPR peak of Au@Cu2O NCs shows a significant redshift compared with that of pure Au colloid solution. With the increase of the Cu2O thickness, the redshift of SPR peak becomes more obvious, and the SPR peak intensity becomes weaker. A possible explanation is that are shown in Figure S2. As for AC-1 NCs, apart from the characteristic diffraction peaks of Cu2O (JCPDS card no. 05-0667), the diffraction peaks appearing at 38.18, 64.57, and 77.54° are corresponding to (111), (220), and (311) of Au (JCPDS card no. 04-0784) [44,45]. Both of the SAED and XRD results confirm that the prepared AC-1 NCs consist of Cu2O and Au. Figure 3b shows the high-angle annular dark-field scanning TEM (HAADF-STEM) images and the corresponding energy-dispersive spectroscopy (EDS) elemental mapping images of AC-1 NCs. It is easily observed that the outer Cu2O completely cover the Au cores because the image contrast is brighter for the heavier element. In addition, EDS mapping images of Au, Cu, and O elements show that the Au element is located in the center of the core-shell structure, and Cu and O elements are uniformly distributed on the surfaces of Au nanocrystals [46].   Figure 4 shows the UV-Vis absorption spectra of Au colloid solution, Cu2O nanocrystals, and Au@Cu2O NCs with different shell thicknesses. The pure Au colloid solution exhibits a prominent SPR absorption peak at 518 nm. A strong absorption peak of pure Cu2O nanocrystals appears at 464 nm. As for Au@Cu2O NCs, two broad absorption peaks are observed. The absorption peak located at 526-584 nm may be attributed to plasmon resonance of Au cores, while the peak at 406-432 nm is ascribed to Cu2O shells. The SPR peak of Au@Cu2O NCs shows a significant redshift compared with that of pure Au colloid solution. With the increase of the Cu2O thickness, the redshift of SPR peak becomes more obvious, and the SPR peak intensity becomes weaker. A possible explanation is that  Figure 4 shows the UV-Vis absorption spectra of Au colloid solution, Cu 2 O nanocrystals, and Au@Cu 2 O NCs with different shell thicknesses. The pure Au colloid solution exhibits a prominent SPR absorption peak at 518 nm. A strong absorption peak of pure Cu 2 O nanocrystals appears at 464 nm. As for Au@Cu 2 O NCs, two broad absorption peaks are observed. The absorption peak located at 526-584 nm may be attributed to plasmon resonance of Au cores, while the peak at 406-432 nm is ascribed to Cu 2 O shells. The SPR peak of Au@Cu 2 O NCs shows a significant redshift compared with that of pure Au colloid solution. With the increase of the Cu 2 O thickness, the redshift of SPR peak becomes more obvious, and the SPR peak intensity becomes weaker. A possible explanation is that Cu 2 O shells have larger refractive index compared to that of the solvent [26]. Apart from the SPR peak, it is found that there is a blueshift in the wavelength centered at 406-432 nm for Au@Cu 2 O NCs compared with the absorption peak of pure Cu 2 O nanocrystals due to the interband transition in Cu 2 O and the scattering of Cu 2 O shells [47]. It is worth emphasizing that the absorption peak of AC-9 NCs and AC-11 NCs resulting from Cu 2 O shells is almost invisible because the Cu 2 O shells are not formed on the surfaces of these two samples. The inset of Figure 4 is an optical picture of the Au@Cu 2 O NCs. The color of the solution changes from yellow to dark purple with the decrease of the Cu 2 O shell thickness. Nanomaterials 2020, 10, x FOR PEER REVIEW 6 of 14 Cu2O shells have larger refractive index compared to that of the solvent [26]. Apart from the SPR peak, it is found that there is a blueshift in the wavelength centered at 406-432 nm for Au@Cu2O NCs compared with the absorption peak of pure Cu2O nanocrystals due to the interband transition in Cu2O and the scattering of Cu2O shells [47]. It is worth emphasizing that the absorption peak of AC-9 NCs and AC-11 NCs resulting from Cu2O shells is almost invisible because the Cu2O shells are not formed on the surfaces of these two samples. The inset of Figure 4 is an optical picture of the Au@Cu2O NCs. The color of the solution changes from yellow to dark purple with the decrease of the Cu2O shell thickness. XPS is employed to characterize the Cu oxidation state of Au@Cu2O NCs [48]. The high resolution XPS scans of Au@Cu2O NCs are presented in Figure 5. Cu 2p spectrum of AC-1 NCs exhibits two contributions assigned to Cu2O instead of Cu (0) or Cu (II) [49,50], which are Cu 2p3/2 and Cu 2p1/2 at 931.1 and 951.1 eV, respectively [51,52]. The peak positions of Cu 2p3/2 and Cu 2p1/2 slightly shift towards the higher binding energy side with the decrease of Cu2O thickness owing to the variation in the chemical environment, which demonstrates an interaction between Au and Cu2O [47]. In addition, because the XPS spectra intensity is directly proportional to the concentration of atoms, the binding energy intensity of Cu 2p decreases when the Cu2O thickness on the surfaces of Au decreases. XPS is employed to characterize the Cu oxidation state of Au@Cu 2 O NCs [48]. The high resolution XPS scans of Au@Cu 2 O NCs are presented in Figure 5. Cu 2p spectrum of AC-1 NCs exhibits two contributions assigned to Cu 2 O instead of Cu (0) or Cu (II) [49,50], which are Cu 2p 3/2 and Cu 2p 1/2 at 931.1 and 951.1 eV, respectively [51,52]. The peak positions of Cu 2p 3/2 and Cu 2p 1/2 slightly shift towards the higher binding energy side with the decrease of Cu 2 O thickness owing to the variation in the chemical environment, which demonstrates an interaction between Au and Cu 2 O [47]. In addition, because the XPS spectra intensity is directly proportional to the concentration of atoms, the binding energy intensity of Cu 2p decreases when the Cu 2 O thickness on the surfaces of Au decreases. In order to evaluate the catalytic activity of Au@Cu2O NCs in dark conditions, a model reaction for the degradation of MO in the presence of excess NaBH4 is executed by monitoring the change of the UV-Vis absorption peak at λ max = 464 nm. Au nanocrystals, AC-1 NCs, AC-3 NCs, AC-5 NCs, and AC-7 NCs are used to catalyze the reduction of MO solution. Figure 6a illustrates the ln(C/C0) versus reaction time plots, and the degradation of MO solution is completed in 42, 50, 48, 46, and 45 min, respectively. The catalytic activity of Au nanocrystals is higher than that of Au@Cu2O NCs. Unfortunately, Au nanocrystals often tend to agglomerate in order to minimize their surface energy, which restricts their real application [25]. The subtle difference in reduction rate of MO by Au@Cu2O NCs depends upon the differences of thickness of Cu2O shells and size of catalysts due to the size effect [53]. Therefore, the Cu2O shells on the surfaces of Au nanocrystals play a crucial role in catalyzing the degradation of MO. Since the stability is an important index in appraising the quality of catalysts, the stability of Au@Cu2O NCs is also evaluated [54]. Four-run recycle degradation experiments in Figure 6b demonstrate that the four samples can be used for at least four successive cycles. AC-1 NCs have the highest catalytic stability. UV-Vis absorption spectra for reduction of MO catalyzed by AC-1 NCs are displayed in Figure 6c. The absorption peak at 464 nm for MO shows a downward trend and disappears within 50 min. The pseudo-first-order kinetics equation is applied to evaluate the rate constant of the reaction. The rate coefficients are obtained by using the following equation: where C is the concentration of MO at reaction time t, C0 is the initial concentration of MO at t = 0 and k is the rate constant. As shown in Figure 6d, the relationship of ln(C/C0) versus reaction time indicates that the reduction of MO catalyzed by AC-1 NCs follows the pseudo-first-order kinetics. The correlation coefficient R 2 and the rate constant k are 0.991 and 0.299 min −1 , respectively. In order to evaluate the catalytic activity of Au@Cu 2 O NCs in dark conditions, a model reaction for the degradation of MO in the presence of excess NaBH 4 is executed by monitoring the change of the UV-Vis absorption peak at λ max = 464 nm. Au nanocrystals, AC-1 NCs, AC-3 NCs, AC-5 NCs, and AC-7 NCs are used to catalyze the reduction of MO solution. Figure 6a illustrates the ln(C/C 0 ) versus reaction time plots, and the degradation of MO solution is completed in 42, 50, 48, 46, and 45 min, respectively. The catalytic activity of Au nanocrystals is higher than that of Au@Cu 2 O NCs. Unfortunately, Au nanocrystals often tend to agglomerate in order to minimize their surface energy, which restricts their real application [25]. The subtle difference in reduction rate of MO by Au@Cu 2 O NCs depends upon the differences of thickness of Cu 2 O shells and size of catalysts due to the size effect [53]. Therefore, the Cu 2 O shells on the surfaces of Au nanocrystals play a crucial role in catalyzing the degradation of MO. Since the stability is an important index in appraising the quality of catalysts, the stability of Au@Cu 2 O NCs is also evaluated [54]. Four-run recycle degradation experiments in Figure 6b demonstrate that the four samples can be used for at least four successive cycles. AC-1 NCs have the highest catalytic stability. UV-Vis absorption spectra for reduction of MO catalyzed by AC-1 NCs are displayed in Figure 6c. The absorption peak at 464 nm for MO shows a downward trend and disappears within 50 min. The pseudo-first-order kinetics equation is applied to evaluate the rate constant of the reaction. The rate coefficients are obtained by using the following equation: where C is the concentration of MO at reaction time t, C 0 is the initial concentration of MO at t = 0 and k is the rate constant. As shown in Figure 6d, the relationship of ln(C/C 0 ) versus reaction time indicates that the reduction of MO catalyzed by AC-1 NCs follows the pseudo-first-order kinetics. The correlation coefficient R 2 and the rate constant k are 0.991 and 0.299 min −1 , respectively. To enhance the catalytic performance, Ag nanocrystals are further deposited onto the surfaces of AC-1 NCs. Compared with AC-1 NCs, the TEM image of AC-1-Ag NCs in Figure 7a shows that there is no appreciable morphological change after coating with Ag nanocrystals. However, the SAED pattern of AC-1-Ag NCs in Figure 7b indicates that new diffraction rings appear, which are assigned to (111), (200), (220), and (311) crystal planes of Ag. XRD pattern of AC-1-Ag NCs in Figure  S2 also confirms the existence of Ag nanocrystals. Four additional diffraction peaks located at 37.96, 44.2, 64.58, and 77.26° match the corresponding planes of Ag (JCPDS card no. 04-0783) [55][56][57]. HAADF-STEM and the corresponding EDS mapping images of AC-1-Ag NCs are presented in Figure 7c. It is observed that Ag elements are uniformly dispersed onto the surfaces of AC-1 NCs. High resolution XPS scans of Ag 3d for AC-1-Ag NCs are fitted by XPSPEAK41 software, as shown in Figure S3a. Binding energy peaks located at 368.1 and 374.1 eV with a spin-orbit splitting of 6.0 eV are ascribed to Ag 3d5/2 and Ag 3d3/2, which are consistent with the standard reference XPS spectrum of metallic Ag [58][59][60]. Figure S3b depicts UV-Vis absorption spectra of AC-1 NCs and AC-1-Ag NCs. By comparison, AC-1-Ag NCs have a strong and broad absorption peak between 310 and 480 nm, which is mainly caused by the SPR of Ag [61][62][63]. To enhance the catalytic performance, Ag nanocrystals are further deposited onto the surfaces of AC-1 NCs. Compared with AC-1 NCs, the TEM image of AC-1-Ag NCs in Figure 7a shows that there is no appreciable morphological change after coating with Ag nanocrystals. However, the SAED pattern of AC-1-Ag NCs in Figure 7b indicates that new diffraction rings appear, which are assigned to (111), (200), (220), and (311) crystal planes of Ag. XRD pattern of AC-1-Ag NCs in Figure S2 also confirms the existence of Ag nanocrystals. Four additional diffraction peaks located at 37.96, 44.2, 64.58, and 77.26 • match the corresponding planes of Ag (JCPDS card no. 04-0783) [55][56][57]. HAADF-STEM and the corresponding EDS mapping images of AC-1-Ag NCs are presented in Figure 7c. It is observed that Ag elements are uniformly dispersed onto the surfaces of AC-1 NCs. High resolution XPS scans of Ag 3d for AC-1-Ag NCs are fitted by XPSPEAK41 software, as shown in Figure S3a. Binding energy peaks located at 368.1 and 374.1 eV with a spin-orbit splitting of 6.0 eV are ascribed to Ag 3d 5/2 and Ag 3d 3/2 , which are consistent with the standard reference XPS spectrum of metallic Ag [58][59][60]. Figure  S3b depicts UV-Vis absorption spectra of AC-1 NCs and AC-1-Ag NCs. By comparison, AC-1-Ag NCs have a strong and broad absorption peak between 310 and 480 nm, which is mainly caused by the SPR of Ag [61][62][63].
The catalytic activity of AC-1-Ag NCs in the reduction of MO is also investigated under dark condition. The UV-Vis absorption spectra and the corresponding ln(C/C 0 ) versus reaction time plots of MO solution catalyzed by AC-1-Ag NCs are shown in Figure 8a,b. The complete degradation of MO solution can be realized within only 4 min. The linear relationship between logarithm of absorbance ln(C/C 0 ) and reaction time (t) indicates that the reduction of MO by AC-1-Ag NCs follows pseudo-first-order kinetics. The rate constant of MO reduction is as high as 1.081 min −1 . The above results show that the catalytic rate of AC-1-Ag NCs is significantly improved compared with that of AC-1 NCs owing to the presence of Ag nanocrystals on the surfaces of AC-1 NCs. The reusability of AC-1-Ag NCs for catalytic reduction of MO is also been evaluated, as shown in Figure 8c. The catalytic reactivity of AC-1-Ag NCs decreases slightly after four cycles, indicating that AC-1-Ag NCs are highly reusable. To better understand the possible mechanism for the reduction of MO by AC-1-Ag NCs with the help of excess NaBH 4 , MO is reduced by AC-1-Ag NCs without the addition of NaBH 4 and by using NaBH 4 alone. The catalytic activity of AC-1-Ag NCs in the reduction of MO is also investigated under dark condition. The UV-Vis absorption spectra and the corresponding ln(C/C0) versus reaction time plots of MO solution catalyzed by AC-1-Ag NCs are shown in Figure 8a,b. The complete degradation of MO solution can be realized within only 4 min. The linear relationship between logarithm of absorbance ln(C/C0) and reaction time (t) indicates that the reduction of MO by AC-1-Ag NCs follows pseudo-first-order kinetics. The rate constant of MO reduction is as high as 1.081 min −1 . The above results show that the catalytic rate of AC-1-Ag NCs is significantly improved compared with that of AC-1 NCs owing to the presence of Ag nanocrystals on the surfaces of AC-1 NCs. The reusability of AC-1-Ag NCs for catalytic reduction of MO is also been evaluated, as shown in Figure  8c. The catalytic reactivity of AC-1-Ag NCs decreases slightly after four cycles, indicating that AC-1-Ag NCs are highly reusable. To better understand the possible mechanism for the reduction of MO by AC-1-Ag NCs with the help of excess NaBH4, MO is reduced by AC-1-Ag NCs without the addition of NaBH4 and by using NaBH4 alone. Figure 8d demonstrates negligible degradation of MO by adding only AC-1-Ag NCs and only NaBH4. From this point of view, both the catalysts and the electron donors (BH4 − ions) originated from NaBH4 are essential to reduce MO. This strongly confirms the prominent catalytic effects of AC-1-Ag NCs in the presence of an excess amount of NaBH4 on the reduction of MO. In fact, the kinetic barrier between acceptor molecules (MO) and BH4 − ions is too high to allow the reduction reaction of MO to proceed [64]. However, the addition of

Conclusions
To sum up, we synthesized Au@Cu2O-Ag NCs by a simple galvanic replacement reaction method. The Au@Cu2O NCs with controllable shell thickness could be prepared by changing the amount of Au colloid solution. The results of the TEM images indicated that the shell thickness of Au@Cu2O NCs could be changed. It had been also demonstrated that Ag nanocrystals could be successfully compounded on Au@Cu2O NCs. Compared with Au@Cu2O NCs, the catalytic efficiency of Au@Cu2O-Ag NCs was significantly improved. Au@Cu2O-Ag NCs were capable of rapidly reducing MO within 4 min. In addition, Au@Cu2O-Ag NCs had good reusability and high stability. Therefore, the nanocatalysts we designed have great potential to remove organic dye pollutants from wastewater by catalytic degradation.
Supplementary Materials: The following are available online at www.mdpi.com/xxx/s1. Figure S1: TEM image of the Au nanocrystals, the insert is the particle size distribution. Figure S2: XRD patterns of Cu2O nanocrystals, AC-1 NCs, and AC-1-Ag NCs. Figure S3: High resolution XPS scans of Ag 3d for AC-1-Ag NCs (a) and UV-Vis absorbance spectra (b) of AC-1 NCs and AC-1-Ag NCs. Figure S4

Conflicts of Interest:
The authors declare no conflict of interest.

Conclusions
To sum up, we synthesized Au@Cu 2 O-Ag NCs by a simple galvanic replacement reaction method. The Au@Cu 2 O NCs with controllable shell thickness could be prepared by changing the amount of Au colloid solution. The results of the TEM images indicated that the shell thickness of Au@Cu 2 O NCs could be changed. It had been also demonstrated that Ag nanocrystals could be successfully compounded on Au@Cu 2 O NCs. Compared with Au@Cu 2 O NCs, the catalytic efficiency of Au@Cu 2 O-Ag NCs was significantly improved. Au@Cu 2 O-Ag NCs were capable of rapidly reducing MO within 4 min. In addition, Au@Cu 2 O-Ag NCs had good reusability and high stability. Therefore, the nanocatalysts we designed have great potential to remove organic dye pollutants from wastewater by catalytic degradation.