All chemicals were of analytical grade and used without any further purification, including: Copper(II) nitrate trihydrate (Cu(NO₃)₂∙3H₂O, 99.8%, AR-China); Zinc nitrate hexahydrate (Zn(NO₃)₂∙6H₂O, 98%, AR-China); Hexamethylenetetramine (HMTA, C₆H₁₂N₄, 95%, AR-China). All solutions were prepared with double distilled water.

A mixed solution of 0.1 M Zn(NO₃)₂ and 0.1 M HMTA was uniformly mixed in a volume ratio of 1:4 on a magnetic stirrer at constant speed for 15 minutes. To doped Cu²⁺ into the above-mixed solution, a solution of Cu(NO₃)₂ in the molar ratio of 0, 2%, 5% and 7% (calculated compared to the number of moles of Zn²⁺) was slowly added to the mixed solution. The mixed solution was continuously stirred for another 5 minutes and then transferred to a 100 mL Teflon to carry out the hydrothermal reaction at 90˚C for 6 hours. The precipitate obtained after hydrothermal process was filtered and washed several times with double distilled water until the pH ≈ 7. Next, the clean precipitate was dried at 80˚C for 24 hours. The final product was a fine white or blue powder depending on the doped Cu ion content. Synthetic powder product samples were denoted as pure ZnO, 2%Cu-ZnO, 5%Cu-ZnO and 7%Cu-ZnO samples with Cu²⁺ doped content of 0, 2, 5 and 7 mol.% (compared to the mole’s number of Zn²⁺), respectively. Figure 1 showed images of pure ZnO, 2%Cu-ZnO, 5%Cu-ZnO and 7%Cu-ZnO samples after hydrothermal process were filtered and washed and preparation for the next step was to dry at 80˚C for 24 hours to obtain the Cu²⁺ doped ZnO samples, respectively.

X-ray diffraction (XRD) patterns were measured on a X’pert Pro (PANalytical) MPD using Cu-K_α radiation (λ = 1.54065 Å, scan speed: 0.03˚/2s, 2θ » 25˚ - 75˚). Raman shift spectra were measured on MicroRaman LABRAM-1B (laser wavelength: 633 nm, laser power: 6.25 mW, Leica NPLAN L50x/0.50 BD microscope). Scanning electron microscopy/X-ray energy scattering spectra (SEM/EDX) were measured on a HITACHI TM4000 Plus. Reflectance spectra were measured on a Jasco V-750, DRUV-Vis (using a 60 mm integrated sphere of ISV-922, scan speed: 200 nm/min). Liquid UV-Vis absorption spectra were measured on an Agilent 8453.

The photocatalytic activity of the prepared samples was evaluated by degradation the methylene blue (MB) solution. 50 mg of the nanopowder catalyst was suspended in a pyrex containing 60 mL of 10 ppm methylene blue aqueous solution. Then, the mixture was stirred for 60 min in dark to obtain the adsorption-desorption equilibrium. The reaction mixture was exposed to ultraviolet light by a SH-HB2-200W Highbay lamp with central wavelength emission at 370 nm. After every 10 minutes of irradation, 5 ml of the solution was taken out, filtered, centrifuged to get the solution that had separated the catalyst powder and measured the absorbance of those selected solution on a UV-vis spectrophotometer (VARRIAN Carry100) at the maximum wavelength of MB solution (λ = 660 nm). MB degradation was determined according to the formula below.

D ( % ) = C 0 − C C 0 × 100 (1)

where D(%) is the degradation of MB, %; C₀ is the initial concentration of MB (mM) (t = 0); C is the concentration of MB (mM) at time t (minutes).

Define X-ray diffraction (XRD) patterns of pure ZnO, 2%Cu-ZnO, 5%Cu-ZnO and 7%Cu-ZnO samples are shown in Figure 2. On the XRD diagram, Figure 2 showed that the material samples all appear spectral peaks at 2θ diffraction angle positions of 31.7˚, 34.4˚, 36.3˚, 47.5˚, 56.6˚, 62.9˚, 67.9˚ and 69.1˚ correspond to the (100), (002), (101), (102), (110), (103), (112) and (201) lattice planes of the hexagonal wurtzite ZnO with space group (P63mc) (JCPDS card No. 89-7102) [18] [20] [28] .

Diffraction spectra showed that all samples were in single phase and there were no anomalous peaks related to Cu metal clusters or Cu oxides secondary phases. This showed that the Cu ions doped ZnO samples retain the hexagonal wurtzite structure of ZnO. Also, it was attributed to the good dispersion of Cu²⁺ into the ZnO lattice [17] [20] [28] [29] . This can be explained that Cu has been doped into the ZnO lattice and Cu²⁺ ions have replaced Zn²⁺ ions because the atomic radii of these two ions are nearly equal ( r Zn 2 + = 0.074 nm and r Cu 2 + = 0.072 nm) [17] [29] . The crystalline size of samples was calculated using the Scherrer formula [20] [28] :

d = 0.89 λ cos θ (2)

where d is the crystalline size (nm); λ is the wavelength of X-ray radiation (λ =

1.54065 Å); θ is the Bragg angle (radians); β is full width half maxima (FWHM) of (101) diffraction peak (radians). The calculated crystallite size of pure ZnO, 2%Cu-ZnO, 5%Cu-ZnO and 7%Cu-ZnO samples were about 64.5 nm, 22.6 nm, 21.8 nm and 20.4 nm, respectively. This can be observed that the crystallite size of samples decreases from 64.5 nm to 20.4 nm when Cu²⁺ content increases from 0 mol.% to 7 mol.%, indicating that the Cu dopant inhibited the crystallization of ZnO samples.

The general chemical equation of the Cu²⁺ doped ZnO synthesis reaction could be described by the following mechanism [15] [19] [21] :

Z n ( N O 3 ) 2 ↔ H 2 O Z n 2 + + 2 N O 3 − (3)

C 6 H 12 N 4 + 6 H 2 O ↔ H 2 O 4 N H 3 + 4 H C O O H (4)

N H 3 + H 2 O ↔ H 2 O N H 4 + + O H − (5)

Z n 2 + + 2 O H − ↔ H 2 O Z n ( O H ) 2 ↓ (6)

Z n ( O H ) 2 ↓ ↔ M e t h o d H y d r o t h e r m a l Z n O ↓ +   H 2 O (7)

Z n ( N O 3 ) 2 + C u ( N O 3 ) 2 ↔ H y d r o t h e r m a l   m e t h o d H M T A Z n 1 − x C u x O + g a s e o u s   p r o d u c t s (8)

Figure 3 showed the Raman spectra of pure ZnO, 2%Cu-ZnO, 5%Cu-ZnO and 7%Cu-ZnO samples. The Raman spectra showed that the fabricated samples all had Raman wave vibration positions corresponding to the wave vibration positions of the hexagonal wurtzite structure ZnO with space group P63mc [13] [18] . The peaks that were observed at 335 cm⁻¹ and 390 cm⁻¹ were assigned to the E_2(high) − E_2(low) and A_1(TO) modes of the hexagonal wurrtzite ZnO, respectively. The peak at 440 cm⁻¹ was assigned to the E_1(TO) mode of the hexagonal wurrtzite

ZnO. The peak at 573 cm⁻¹ was assigned to the E_1(LO) mode of the hexagonal wurrtzite ZnO [13] [18] .

It could be seen that the wave vibration positions as wel as wave intensity of Raman spectra of the samples were almost similar, with no significant difference. It is possible that the doping amount of Cu²⁺ in the ZnO sample is small (≤7 mol.%). In addition, in the Raman spectra of pure ZnO and doped ZnO samples, there were no wave peaks corresponding to the vibration modes of CuO, Cu₂O or Cu. This could be explained similarly to above because the amount of Cu²⁺ doped into the lattice cell of ZnO is small (≤7 mol.%).

SEM images with different magnifications of pure ZnO, 2%Cu-ZnO, 5%Cu-ZnO and 7%Cu-ZnO samples were shown in Figure 4. The SEM images showed that the pure ZnO sample exhibit a nanorod structure with the thickness size of about 250 - 300 nm. The 2%Cu-ZnO sample exhibited a nanorods structure with the smaler thickness size (of about 20 - 30 nm). The 5%Cu-ZnO and 7%Cu-ZnO samples exhibited a tube structure in the meniscus form at the ends of the tube like a porosity meniscus tube with the thickness size of about 100 - 200 nm. Thus, it could be seen that when the content of doped Cu²⁺ in the ZnO lattice increased: it had the effect of preventing the growth of crystal grains. The crystals of the material samples grew in the form of rods with decreasing crystal size and increasing porosity.

Figure 5 showed the plot between absorbance and wavelength for the pure ZnO,

2%Cu-ZnO, 5%Cu-ZnO and 7%Cu-ZnO samples. The inset showed the photographs of the prepared pure ZnO, 2%Cu-ZnO, 5%Cu-ZnO and 7%Cu-ZnO samples of the experiment. The Kubelka-Munk (K-M) model was used to determine the band gap of the prepared samples based on equation below:

F ( R ) = ( 1 − R ) 2 2 R (9)

where F(R) was K-M function and R was reflectance [18] [20] . The band gap was found to be 3.31 eV, 3.17 eV, 3.16 eV and 3.15 eV for the pure ZnO, 2%Cu-ZnO, 5%Cu-ZnO and 7%Cu-ZnO samples, respectively. Here, it could be seen the band gap decreased gradually as the doping copper content increased gradually into the lattice of ZnO. This difference in the band gap was due to the difference in crystal shape and size of the nondoped and doped ZnO samples [18] [20] .

Figure 6 was the plot to study the photocatalytic activity of ZnO samples for the degradation of methylene blue (MB) under UV irradiation.

The plots in Figure 6 showed that the 2%Cu-ZnO sample was the most for degradation of the MB dye solution, and achieved the decomposition efficiency of about 97%; the 5%Cu-ZnO sample achieved the decomposition efficiency of about 96%; the 7%Cu-ZnO sample had a decomposition efficiency of about 92.6% and the ZnO pure sample had the lowest decomposition efficiency of about 20%. It could be seen that the different content of Cu dopant into the pure ZnO lattice changed the shape and size of the crystal particles of the prepared

samples and led to an improvement in the photocatalytic performance for degradation of MB dye solution. The highest MB degradation efficiency was obtained in the ZnO sample with a copper doping concentration of 2 mol.%.