A Highly Efficient and Stable Photocatalyst; N-Doped ZnO/CNT Composite Thin Film Synthesized via Simple Sol-Gel Drop Coating Method

The thin film of N-doped ZnO/CNT nanocomposite was successfully fabricated on soda lime glass substrate by a simple sol-gel drop-coating method. The structural, morphological, chemical, and optical properties of as prepared samples were characterized by a variety of tools such as X-ray Diffraction (XRD), Field Emission Scanning Electron Microscopy (FE-SEM), Fourier Transform Infrared spectroscopy (FT-IR), and UV-visible spectroscopy. The hexagonal crystalline structure was confirmed from XRD measurement without any other impurity phase detection in samples. The N-doped ZnO/CNT composite showed excellent photo-catalytic activity towards cationic methylene blue (MB) dye degradation with 100% removal rate under UV light irradiation as compared to N-doped ZnO (65%) and pure ZnO (47.36%). The convincing performance has also been observed for the case of visible light irradiation. The enhancement of that photocatalytic activity might be due to narrowing the band gap as well as the reduction of electron–hole pair recombination in ZnO matrix with the incorporation of dopant nitrogen and CNT. It is assumed from the obtained results that N-doped ZnO/CNT nanocomposite thin film can be employed as an economically achievable and ecofriendly method to degrade dye with UV and visible light irradiation. Additionally, density functional theory (DFT) calculations were applied to explore the effect of N-doping on electronic structure of ZnO. The computational study has supported the experimental results of significant band gap contraction, which leads to the maximum absorption towards higher wavelength and no appreciable change of lattice parameters after doping. A conceivable photocatalytic mechanism of N-doped ZnO/CNT nanocomposite has been proposed as well.


Introduction
Organic pollutants like colored dyes, antibiotics, and colorless organic matters in aquatic environments are now considered as threats for the natural environment and human health [1][2][3]. Especially, colored dyes are mainly heterocyclic organic compounds, which have N=N bonds and sulfonic group in their molecular structure. These dyes are extensively used in the various industries such as the textile industry. The disposal of dyes into water bodies can constitute widespread damages to human health, aquatic life, and the food chain because of its noxious effects [4][5][6]. The dyes are highly colored toxic organic compounds, which can affect the photosynthesis action of entire ecosystem by reducing sunlight penetration and dissolve oxygen [7,8]. This indicates the necessity of treatment of water containing dye before disposal into the environment. A wide number of dye composite thin film and found the enhanced visible light photodegradation over MB dye. Despite there being a huge number of works on composite photocatalysts, no work has been reported on N-doped ZnO/CNT composite thin film.
In this research, the N-doped ZnO/CNT thin film on glass substrate has been fabricated by a simple sol-gel drop coating method. The influence of dopant nitrogen and sensitizer CNTs on the structural behavior of the ZnO was analyzed by XRD, FESEM, FT-IR, DFT, and UV-visible spectroscopy. The prepared nanocomposite thin film demonstrated improved photocatalytic efficiency for the degradation of MB dye under UV and visible light irradiation as compared to pure ZnO and N-doped ZnO thin films.

XRD Analysis
The XRD patterns were used to determine the effects of the addition of dopant on the crystal phase, lattice strain, lattice dislocation density, and crystallinity of ZnO thin films. Figure 1 shows the XRD patterns of (a) pure ZnO, (b) N-doped ZnO, and (c) N-doped ZnO/CNT thin films, respectively. All the prepared thin films with high crystallinity revealed a hexagonal wurtzite crystal structure, which mostly matches well with the reported data (JCPDS No. 01-075-0576) [35]. The peaks centered at an angle (2θ) of 31 (Figure 1b,c), representing a small crystal size of these thin films. The most intense peak (36.29 • ) of the XRD pattern of bare ZnO thin films has been little shifted to higher 2θ values for the N-doped and CNT composite of that materials. This confirms the doping of ZnO with nitrogen and composite with CNT. In addition, the XRD data were employed to evaluate average size of the crystallites, D from the peak half-width β, using the Scherrer's equation [21] (Equation (1)): where k, λ, and θ are the shape factor of the particle, wavelength, and the incident angle of the X-rays, respectively. According to this Scherrer equation, the average crystallite size for ZnO, N-doped ZnO, and N-doped ZnO/CNTs is 29.07 nm, 27.13 nm, and 14.28 nm, respectively. The crystalline sizes have been reduced significantly after addition of dopant nitrogen as well as CNT compared to that of pure ZnO. These results indicate that only doping of nitrogen into ZnO matrix plays a role to reduce the particle size in N-doped ZnO. Moreover, CNT inhibits the growth of ZnO crystallites drastically in N-doped ZnO/CNT nanocomposite thin film [8].
The lattice strain induced due to the crystal imperfection and distortion in the pure ZnO thin film, N-doped ZnO, and N-doped ZnO/CNT was calculated by the Stokes-Wilson equation (Equation (2)): The dislocation density points towards the amount of crystallographic defect or irregularity present in the crystal that strongly influences the properties of the synthesized materials. The dislocation density was calculated by applying the Williamson-Smallman relation [41] (Equation (3)):  The lattice strain induced due to the crystal imperfection and distortion in the p ZnO thin film, N-doped ZnO, and N-doped ZnO/CNT was calculated by the Stokesson equation (Equation (2)): The dislocation density points towards the amount of crystallographic defect o regularity present in the crystal that strongly influences the properties of the synthes materials. The dislocation density was calculated by applying the Williamson-Small relation [41] (Equation (3)): The average crystallite size, lattice strain, and dislocation density results are sum rized in Table 1. The lattice constants a & c are related to the inter-planar spacing of the atomic pl d for hexagonal wurzite structure and can be calculated from the XRD spectra by the lowing equation [41] (Equation (4)): The average crystallite size, lattice strain, and dislocation density results are summarized in Table 1. The lattice constants a & c are related to the inter-planar spacing of the atomic planes d for hexagonal wurzite structure and can be calculated from the XRD spectra by the following equation [41] (Equation (4)): The Miller indices h, k, & l of (100) plane of hexagonal structure is used to calculate the lattice constant a with the help of the following equation [40] (Equation (5)): The lattice constant c is calculated from the reflection (002) plane by using the following equation [42] (Equation (6)): Finally, the unit cell volume of hexagonal pure ZnO, N-doped ZnO, and N-doped ZnO/CNT thin films is calculated by the following equation [43] (Equation (7)): where a & c are the lattice constants of the prepared thin films and the obtained results of the lattice constants and the unit cell volume of synthesized thin films are summarized in Table 2. The lattice parameters value is the indication of positive (extensive) or negative (compressive) stress of crystallites if the value is lower or higher than the bulk materials [40]. The value of c is less than the value of bulk ZnO implies that all the thin films of ZnO crystallites are the state of compressive stress. The lattice constants a = b and c of hexagonal unit cell exhibit lower value compared with the value of bulk ZnO. It can be seen from Table 2 that the lower unit cell volume of all the prepared thin films compared to the standard is a result of higher stress.

√3 sin
The lattice constant c is calculated from the reflection (002) plane by using the following equation [42] (Equation (6)): = sin (6) Finally, the unit cell volume of hexagonal pure ZnO, N-doped ZnO, and N-doped ZnO/CNT thin films is calculated by the following equation [43] (Equation (7)): where a & c are the lattice constants of the prepared thin films and the obtained results of the lattice constants and the unit cell volume of synthesized thin films are summarized in Table 2. The lattice parameters value is the indication of positive (extensive) or negative (compressive) stress of crystallites if the value is lower or higher than the bulk materials [40]. The value of c is less than the value of bulk ZnO implies that all the thin films of ZnO crystallites are the state of compressive stress. The lattice constants a = b and c of hexagonal unit cell exhibit lower value compared with the value of bulk ZnO. It can be seen from Table 2 that the lower unit cell volume of all the prepared thin films compared to the standard is a result of higher stress.  Figure 2 shows the FESEM images of as synthesized (a) pure ZnO, (b) N-doped ZnO, and N-doped ZnO/CNT thin films calcined at 500 °C. The surface morphology of pure ZnO thin film reveals the irregular spherical shape of particles with some agglomeration while N-doped ZnO shows a smooth surface with small spherical grains distributed as uniform, compact, and adherent to substrate. However, the morphology of N-doped ZnO/CNT is formed in a more consistent spherical shape with a lower degree of agglomeration. The comparatively homogeneous morphology (Figure 2c) might be due to the inclusion of functionalized CNT that contains the polar functional groups on its surface that impede the agglomeration of nanoparticles. However, it looks very difficult to figure out the CNT structure from Figure 3c with that low resolution image. In addition, the uniform and small size of the catalyst particle can contribute to a greater adsorption of dye on surface and more light absorption in this photocatalytic system. With the inclusion of dopant nitrogen and CNT, the grain size decreases from 30 to 14 nm. The unevenness

= sin
Finally, the unit cell volume of hexagonal pure ZnO, N-doped ZnO, and ZnO/CNT thin films is calculated by the following equation [43] (Equation (7)): where a & c are the lattice constants of the prepared thin films and the obtained the lattice constants and the unit cell volume of synthesized thin films are summ Table 2. The lattice parameters value is the indication of positive (extensive) or (compressive) stress of crystallites if the value is lower or higher than the bulk [40]. The value of c is less than the value of bulk ZnO implies that all the thin film crystallites are the state of compressive stress. The lattice constants a = b and c o nal unit cell exhibit lower value compared with the value of bulk ZnO. It can be s Table 2 that the lower unit cell volume of all the prepared thin films compar standard is a result of higher stress.  Figure 2 shows the FESEM images of as synthesized (a) pure ZnO, (b) N-do and N-doped ZnO/CNT thin films calcined at 500 °C. The surface morpholog ZnO thin film reveals the irregular spherical shape of particles with some agglo while N-doped ZnO shows a smooth surface with small spherical grains distr uniform, compact, and adherent to substrate. However, the morphology of ZnO/CNT is formed in a more consistent spherical shape with a lower degree o eration. The comparatively homogeneous morphology ( Figure 2c) might be d inclusion of functionalized CNT that contains the polar functional groups on i that impede the agglomeration of nanoparticles. However, it looks very difficult out the CNT structure from Figure 3c with that low resolution image. In add uniform and small size of the catalyst particle can contribute to a greater adso dye on surface and more light absorption in this photocatalytic system. With the of dopant nitrogen and CNT, the grain size decreases from 30 to 14 nm. The un

= sin
Finally, the unit cell volume of hexagonal ZnO/CNT thin films is calculated by the follow where a & c are the lattice constants of the prepa the lattice constants and the unit cell volume of Table 2. The lattice parameters value is the ind (compressive) stress of crystallites if the value i [40]. The value of c is less than the value of bulk crystallites are the state of compressive stress. T nal unit cell exhibit lower value compared with Table 2 that the lower unit cell volume of all standard is a result of higher stress.  Figure 2 shows the FESEM images of as syn and N-doped ZnO/CNT thin films calcined at ZnO thin film reveals the irregular spherical sh while N-doped ZnO shows a smooth surface w uniform, compact, and adherent to substrate. ZnO/CNT is formed in a more consistent spher eration. The comparatively homogeneous mor inclusion of functionalized CNT that contains t that impede the agglomeration of nanoparticles out the CNT structure from Figure 3c with th uniform and small size of the catalyst particle dye on surface and more light absorption in this of dopant nitrogen and CNT, the grain size dec  Figure 2 shows the FESEM images of as synthesized (a) pure ZnO, (b) N-doped ZnO, and N-doped ZnO/CNT thin films calcined at 500 • C. The surface morphology of pure ZnO thin film reveals the irregular spherical shape of particles with some agglomeration while N-doped ZnO shows a smooth surface with small spherical grains distributed as uniform, compact, and adherent to substrate. However, the morphology of N-doped ZnO/CNT is formed in a more consistent spherical shape with a lower degree of agglomeration. The comparatively homogeneous morphology (Figure 2c) might be due to the inclusion of functionalized CNT that contains the polar functional groups on its surface that impede the agglomeration of nanoparticles. However, it looks very difficult to figure out the CNT structure from Figure 3c with that low resolution image. In addition, the uniform and small size of the catalyst particle can contribute to a greater adsorption of dye on surface and more light absorption in this photocatalytic system. With the inclusion of dopant nitrogen and CNT, the grain size decreases from 30 to 14 nm. The unevenness and grain size decrease because of the doping nitrogen as well as CNT that ascribe the enhancement in crystalline quality. and grain size decrease because of the doping nitrogen as well as CNT that ascribe the enhancement in crystalline quality.

FT-IR Analysis
The FT-IR spectrum illustrates the information of a particular compound about its functional groups, molecular geometry, and inter/intramolecular interactions. Figure 3

FT-IR Analysis
The FT-IR spectrum illustrates the information of a particular compound about its functional groups, molecular geometry, and inter/intramolecular interactions. Figure 3 displays the FT-IR spectra of as prepared (a) pure ZnO, (b) N-doped ZnO, and (c) N-doped ZnO/CNT nanocomposite thin film, respectively, and those were performed in the wave number range 400-4000 cm −1 using the KBr method at room temperature. All FT-IR spectra display a broad absorption band at around 3400-3600 cm −1 due to the presence of the O-H stretching mode of the hydroxyl groups. The absorption band around 2800-2900 cm −1 is identified as a stretching mode for the C-H bond of binder TEA. The band at 1650 cm −1 characterizes the H-O-H bending vibration of water. The bands in the low wave number region, 426 cm −1 , correspond to the vibration modes of Zn-O [41]. It is assumed that the bands appearing at 1109 cm −1 and 1020 cm −1 are due to stretching vibrations of Zn-O-Si and Si-O-Si that appear from the glass substrate [44].

UV-Visible Absorption Spectra Analysis
The photocatalytic activity of photocatalyst depends on the light absorption capacity. UV-visible absorption spectra have been measured to determine the optical properties of the prepared thin films and the obtained results are shown in Figure 4a. The pure ZnO has an absorption maximum at 362 nm, whereas the peaks at 368 nm and 366 nm are obtained for the N-doped ZnO/CNT and N-doped ZnO, respectively. The presence of absorption peak indicates the electronic transition from valence band to conduction band of ZnO nanoparticles [45]. It can be seen that N-doping has an influence on the absorption spectrum into UV to the visible region (red shift). The similar results are also found in the references for the redshift in the absorption edge [30]. The band gap of prepared thin films has been calculated with help of the following Tauc equation (Equation (8)): where Eg is the band gap energy, hν is the photon energy, and α refers to the absorption coefficient, by plotting the value αhν along the y-axis and hν along the x-axis followed by the extrapolating the liner portion of the curve to hν axis (αhν) 2  αhν= (hν − Eg)1/2 (8) where Eg is the band gap energy, hν is the photon energy, and α refers to the absorption coefficient, by plotting the value αhν along the y-axis and hν along the x-axis followed by the extrapolating the liner portion of the curve to hν axis (αhν) 2

Density Functional Theory (DFT) Study
The lattice parameters of optimized Zn8O8 a = b = 3.127Å and c = 5.032 Å are in good agreement with experimental values ( Table 2). After nitrogen doping, the lattice parame-

Density Functional Theory (DFT) Study
The lattice parameters of optimized Zn 8 O 8 a = b = 3.127Å and c = 5.032 Å are in good agreement with experimental values (Table 2). After nitrogen doping, the lattice parameters of N-doped Zn 8 O 8 (a = b = 3.124 Å and c = 5.098 Å) changes 0.003 Å and 0.066 Å for a and c, respectively. Theoretically calculated lattice parameters deviate 3.3% from those that are experimentally measured. To determine the effect of nitrogen doping in ZnO, band structure and total density of states (TDOS) of ZnO and N-doped ZnO have been calculated as shown in Figure 5. The band structure shows the band gap of ZnO is 2.18 eV, which is lower than that experimentally obtained. This deviation is originated because the DFT calculation is concerned with only the ground state [46]. After nitrogen doping, the band gap reduces to 1.71 eV, which is relevant for UV-visible absorption spectrum. To justify this, we calculated the total density of states (TDOS). The peak of the valence band maximum (VBM) of ZnO is located below the Fermi level. However, nitrogen doping in ZnO attributes the upward shift of VBM to reduce the band gap. There is almost no effect of nitrogen doping on conduction band minimum (CBM) of ZnO.

Photocatalytic Activity Evaluation
The photocatalytic performance of pure ZnO, N-doped ZnO, and N-doped ZnO/CNT thin films in degradation of MB dye was carried out under UV and visible light irradiation and results are shown in Figure 6. To confirm the photolysis and adsorption onto thin film, the adsorption as well as photodegradation experiments were conducted separately under visible light and UV light irradiation for 4 h. For both studies, no appreciable changes were observed due to adsorption and self photodegradation of the MB dye. About 100% of MB is degraded by an N-doped ZnO/CNT nanocomposite thin film photocatalyst within 4 h under UV light irradiation, while the efficiency is 63.69% under visible light irradiation. From Figure 6, it was noticed that about 60% of MB was removed after 60 min by UV light irradiation for N-doped ZnO/CNT, whereas only 16.66% and 28.57% for pure ZnO and N-doped ZnO photocatalysts. In case of visible light irradiation, the effectiveness of pure ZnO has 23.41% degradation efficiency, much less compared with N-doped ZnO (34.62%) and N-doped ZnO/CNT (63.69%). These results clarify that

Photocatalytic Activity Evaluation
The photocatalytic performance of pure ZnO, N-doped ZnO, and N-doped ZnO/CNT thin films in degradation of MB dye was carried out under UV and visible light irradiation and results are shown in Figure 6. To confirm the photolysis and adsorption onto thin film, the adsorption as well as photodegradation experiments were conducted separately under visible light and UV light irradiation for 4 h. For both studies, no appreciable changes were observed due to adsorption and self photodegradation of the MB dye. About 100% of MB is degraded by an N-doped ZnO/CNT nanocomposite thin film photocatalyst within 4 h under UV light irradiation, while the efficiency is 63.69% under visible light irradiation. From Figure 6, it was noticed that about 60% of MB was removed after 60 min by UV light irradiation for N-doped ZnO/CNT, whereas only 16.66% and 28.57% for pure ZnO and Ndoped ZnO photocatalysts. In case of visible light irradiation, the effectiveness of pure ZnO has 23.41% degradation efficiency, much less compared with N-doped ZnO (34.62%) and Ndoped ZnO/CNT (63.69%). These results clarify that N doping enhances the photocatalytic efficiency of plain ZnO by reducing the band gap. On the other hand, CNTs improved the interfacial electron transfer rate and lessened the recombination possibility between photo-induced electron hole pair (e − /h + ) more effectively than the ZnO photocatalyst alone. It could be summarized that the reduction of e − /h + recombination and generation of more • OH radicals in the N doped ZnO as well as N-doped ZnO/CNT samples play a vital role in the improved rate of photocatalytic degradation. The produced • OH radicals have strong enough reactivity to break different (C-C, C=C, and C=O) bonds in MB molecules lead to the formation of CO 2 and H 2 O at the end. However, many researchers have used numerous pollutants such as 4-cholorophenol, antibiotics, and natural organic matters, in order to examine the feasibility of the photocatalyst [2,47,48]. However, many researchers have used numerous pollutants such as 4-cholorophenol, antibiotics, and natural organic matters, in order to examine the feasibility of the photocatalyst [2,47,48]. Recent reports point out that several oxidants such as peroxymonosulfate, hydrogen peroxide (H2O2), peroxydisulfate, ClO3 − etc. in heterogeneous photocatalytic system increase the degradation efficiency either by lowering e − /h + recombination rate through accepting the conduction band electrons at the surface of ZnO photocatalyst or by providing additional oxygen atoms as an electron acceptor to form the superoxide radical ion (O2 •− ) [48]. Herein, the effect of varying dose of oxidant H2O2 (5-30 mmol) was studied in case of N-doped ZnO/CNT nanocomposite photocatalyst thin film for MB dye degradation under UV light irradiation. The results showed that MB dye degradation increases due to the addition of H2O2 and the MB dye degradation efficiency was 100% within 60 min under UV light irradiation for 5 mmol H2O2. The results of degradation efficiency were similar up to 25 mmol H2O2, whereas, for additional doses like above 25 mmol of H2O2, the degradation efficiency was not changed. The higher degree of degradation is due to the formation of highly reactive radical intermediates and the electron capture by oxidant H2O2. As a consequence, the addition of H2O2 to the heterogeneous system increases the concentration of • OH radicals.
N-doped ZnO/CNT (electron from conduction band) + H2O2 → N-dopedZnO/CNT + OH -+ OH • As H2O2 is an electron acceptor species, it does not only produce • OH radicals but also obstructs the electron hole recombination [48]. When the H2O2 concentration becomes high, the excess H2O2 consumes hydroxyl radicals as scavenger [49] (Equation (9)): H2O2 + • OH → HO2 • + H2O (9) When H2O2 was added, the maximum degradation was achieved in the first 60 min  As H 2 O 2 is an electron acceptor species, it does not only produce • OH radicals but also obstructs the electron hole recombination [48]. When the H 2 O 2 concentration becomes high, the excess H 2 O 2 consumes hydroxyl radicals as scavenger [49] (Equation (9)): When H 2 O 2 was added, the maximum degradation was achieved in the first 60 min only, whereas, without oxidants, it took 240 min for the 100% degradation. The influence of initial dye concentration on photocatalytic degradation is defined by the Langmuir-Hinshelwood kinetic model (Equation (10)) [50]: where r is the rate of reaction, k is the reaction rate constant (mgL −1 min −1 ), K is the observed equilibrium constant of the reactant (Lmg −1 ), and [MB] is the reactant concentration (mg L −1 ). When [MB] is very small, K[MB] is negligible with respect to unity and photocatalysis can be simplified to an apparent pseudo-first-order kinetics [51]. Therefore, Integration of Equation (11) gives where k obs is the observed pseudo-first-order rate constant (min −1 ). A plot of ln vs. irradiation time t provided a straight line with their rate constants at different initial concentrations of MB (Figure 7). The observed degradation rate and correlation coefficients are represented in Table 3. All correlation coefficients values suggest that the proposed kinetic model was in excellent agreement with our experimental data. N-doped ZnO/CNT under UV light irradiation shows the best photocatalytic performance with a rate constant much higher than those of N-doped ZnO and bare ZnO under visible as well as UV light irradiation. The same result was observed by Hadi et al. during the photo-catalytic degradation of MB dyes using Tungsten doped zinc oxide as photocatalysts [52].
where r is the rate of reaction, k is the reaction rate constant (mgL -1 min -1 ), K is the observed equilibrium constant of the reactant (Lmg -1 ), and [MB] is the reactant concentration (mg L -1 ). When [MB] is very small, K[MB] is negligible with respect to unity and photocatalysis can be simplified to an apparent pseudo-first-order kinetics [51]. Therefore, Integration of Equation (11) gives ] = (12) where kobs is the observed pseudo-first-order rate constant (min −1 irradiation time t provided a straight line with their rate constants at different initial concentrations of MB (Figure 7). The observed degradation rate and correlation coefficients are represented in Table 3. All correlation coefficients values suggest that the proposed kinetic model was in excellent agreement with our experimental data. N-doped ZnO/CNT under UV light irradiation shows the best photocatalytic performance with a rate constant much higher than those of N-doped ZnO and bare ZnO under visible as well as UV light irradiation. The same result was observed by Hadi et al. during the photo-catalytic degradation of MB dyes using Tungsten doped zinc oxide as photocatalysts [52].

Study of Photocatalyst Stability and Reusability
The long-term stability and recyclability of photocatalyst thin film are very crucial for MB dye degradation as well as large scale practical utilization in the industry. Therefore, the reusability of a prepared N-doped ZnO/CNT nanocomposite thin film was examined by five consecutive cycles for the degradation of MB dye under UV light illumination, and results are shown in Figure 8. After every test, the thin film composite photocatalyst was washed with distilled water and dried naturally at room temperature followed by the heat treatment at 500 • C for 1 h. After that, a thin film was applied for a second cycle MB dye degradation test like the first cycle experiment. The degradation efficiency of MB dye was reduced to about 5% after five successive cycles, which means that degradation efficiency reached 95% from 100% after five recycling times. These results definitely indicate that the fabricated N-doped ZnO/CNT nanocomposite thin film photocatalyst is highly stable and reusable. Furthermore, a comparison of the photocatalytic activity of the N-doped ZnO/CNT thin film photocatalyst with previously reported photocatalysts was listed in Table 4. From the Table 4, it is clear that degradation efficiency and reusability of N-doped ZnO/CNT nanocomposites are comparable to the results with earlier reported ZnO-based materials.

Study of Photocatalyst Stability and Reusability
The long-term stability and recyclability of photocatalyst thin film are very crucial for MB dye degradation as well as large scale practical utilization in the industry. Therefore, the reusability of a prepared N-doped ZnO/CNT nanocomposite thin film was examined by five consecutive cycles for the degradation of MB dye under UV light illumination, and results are shown in Figure 8. After every test, the thin film composite photocatalyst was washed with distilled water and dried naturally at room temperature followed by the heat treatment at 500 °C for 1 h. After that, a thin film was applied for a second cycle MB dye degradation test like the first cycle experiment. The degradation efficiency of MB dye was reduced to about 5% after five successive cycles, which means that degradation efficiency reached 95% from 100% after five recycling times. These results definitely indicate that the fabricated N-doped ZnO/CNT nanocomposite thin film photocatalyst is highly stable and reusable. Furthermore, a comparison of the photocatalytic activity of the N-doped ZnO/CNT thin film photocatalyst with previously reported photocatalysts was listed in Table 4. From the Table 4, it is clear that degradation efficiency and reusability of N-doped ZnO/CNT nanocomposites are comparable to the results with earlier reported ZnO-based materials.

Photocatalytic Degradation Mechanism
The excellent photocatalytic efficiency of synthesized N-doped ZnO/CNT composite supports the following degradation mechanism of MB dye under UV light irradiation as described by Chemical equations (1-7). The computational and theoretical studies confirm the incorporation of dopant N into the ZnO matrix and doesn't change the wurzite structure of ZnO; it only replaces the O atom with an N atom. Upon illumination of Ndoped ZnO/CNT composite semiconductor photocatalysts by the UV light having energy (hv>>E g ) greater than the band gap results in the formation of charge carriers' electronholes (e − /h + ). These electron-holes (e − /h + ) recombine quickly, thereby producing minute amounts of highly reactive oxidizing species such as hydroxyl radicals, hydrogen peroxide, and superoxides that take part in the photocatalytic degradation of MB dye. However, the addition of CNT with the N-doped ZnO forms a heterojunction interface that can reduce the electron-hole recombination rate: (1) N-doped ZnO/CNT composite + hv→ h vb + + e cb − (CNT) The improved degradation efficiency of N-doped ZnO/CNT composite is due to the synergistic effect of CNT into N-doped ZnO as shown in Figure 9.

Experimental Section
3.1. Materials   Figure 9. Schematic illustration of photocatalytic mechanism of the N-doped ZnO/CNT thin films under UV light irradiation.

Functionalization of CNT
Functionalization of CNT leads to the oxygenated polar functional groups on its surface. Therefore, oxygenated polar functionalities enhance the binding of CNT with ZnO through some physical/chemical interactions, for instance hydrogen bonding, van der Waals attraction as well as other bonds. In addition, the acid treatments (functionalization) also reduce the impurities of the amorphous carbon as well as unfold the ends and or break the tubes. In a typical functionalization, fresh MWCNT (1 g) was immersed into 40 mL of mixed solution (1:1 ratio) of concentrated sulfuric acid and nitric acid. [35]. Then, the combined solution was refluxed for 5 h at 60-70 • C in an oil bath to get dark-brown suspension. After completion of reaction, the prepared suspension was separated by centrifugation after cooling at room temperature. Thereafter, the product was washed with de-ionized water and ethanol until the pH of the filtrate become neutral, and dried in vacuum oven at 80 • C to use in the next step.

Fabrication of N-Doped ZnO/CNT Thin Film
N-doped ZnO/CNT thin film was fabricated by applying sol-gel drop coating protocol according to the literature with some modifications [57]. Typically, to prepare the N-doped ZnO/CNT, 0.7316 g of zinc acetate dihydrate and 0.4 mL of tri-ethylamine were first dissolved in 9.6 mL of anhydrous ethanol under stirring environment (300 rpm) for 4-5 min, followed by the addition of 5wt% of nitrogen (urea) (28 mg urea) and 5wt% of CNTs (5 mg) to the above solution. The resultant solution was sonicated (Powersonic 405, Hwashin Technology Co., Seoul, Korea) for 1 h to obtain a uniform dispersion. Then, the solution was preserved for 1 h maintaining temperature (55 ± 5) • C with continuous stirring at 300 rpm until a quite stable and transparent sol was obtained. Prior to the coating process, the stable N-doped ZnO/CNT sol was aged for 24 h and a commercial bench grinder (model: ST-150) was used to abrade glass slide having dimension of 10 mm × 60 mm × 1.5 mm. The abrasive glass slide was cleaned by using potassium dichromate and dichloromethane solution, respectively, followed by washing with alcohol and distilled water and then dried at 100 • C in an oven for further use. The N-doped ZnO/CNT nanocomposite thin film (heterostructure) was fabricated by a sol-gel drop coating technique, as follows: first, the 0.4 mL sol was used drop-wise on a rubbed glass substrate using a pipette. Subsequently, the sol supported on substrate was warmed for 20 min in an oven at 80 • C to eliminate the loosely bonded particles from the glass surface. Coated thin film could cover approximately 2 µm on the above estimated area of abrasive glass substrate. To obtain the thick film (~4 µm), the coating was repeated twice as well as calcined using a muffle furnace (JSMF-30T, Gongju, Korea) retaining temperature 500 • C for 2 h. Similarly, the N-doped ZnO thin film was prepared in the absence of CNT. Finally, pure ZnO thin film was prepared using the same procedure. A schematic flow diagram for the fabrication of N-doped ZnO/CNT nanocomposite thin film by sol-gel drop coating method is shown in Figure 10.
could cover approximately 2 μm on the above estimated area of abrasive glass substrate. To obtain the thick film (~4 μm), the coating was repeated twice as well as calcined using a muffle furnace (JSMF-30T, Gongju, Korea) retaining temperature 500 °C for 2 h. Similarly, the N-doped ZnO thin film was prepared in the absence of CNT. Finally, pure ZnO thin film was prepared using the same procedure. A schematic flow diagram for the fabrication of N-doped ZnO/CNT nanocomposite thin film by sol-gel drop coating method is shown in Figure 10.

Film Characterization
The crystallinity of the prepared thin film was investigated using the XPERT PRO Xray diffraction (XRD, Almelo, Netherlands) instrument with Cu-Kα monochromatic radiation source (λ = 1.5406 Å, V = 40 kV, I = 30 mA, RT) in the 2θ limit of 10-70°. The average crystallite sizes of synthesized thin films were estimated with the help of Scherrer equation. The surface morphology and microstructure of synthesized nanocomposite thin films were assessed by the Field Emission Scanning Electron Microscope (FESEM; JSM-

Film Characterization
The crystallinity of the prepared thin film was investigated using the XPERT PRO X-ray diffraction (XRD, Almelo, Netherlands) instrument with Cu-K α monochromatic radiation source (λ = 1.5406 Å, V = 40 kV, I = 30 mA, RT) in the 2θ limit of 10-70 • . The average crystallite sizes of synthesized thin films were estimated with the help of Scherrer equation. The surface morphology and microstructure of synthesized nanocomposite thin films were assessed by the Field Emission Scanning Electron Microscope (FESEM; JSM-7600F, Tokyo, Japan). The optical properties of samples were accomplished by double beam UV-Visible spectrophotometer (UV-1800, Shimadzu, Kyoto, Japan). The Fourier Transform Infrared (FT-IR) spectra of the prepared thin films were recorded in transmission mode on KBr pellets using a spectrometer (Shimadzu IR Prestige 21, Kyoto, Japan).

Photo-Catalytic Assessment
The photo-catalytic degradation efficiency of as-fabricated nanocomposite thin film was determined by measuring the photodegradation of an aqueous solution of MB dye as a target pollutant under visible light and UV light irradiation. The experimental setup is described as follows: firstly, to evaluate photo-catalytic activity of the prepared thin films under visible light irradiation, 200 mL aqueous solution of MB (5 mg/L) and a thin film photocatalyst were taken in a beaker. Then, the beaker containing catalyst was placed in a reactor consisting with a cycled cooling water system to avoid any thermal reaction. In each study, the aqueous solution of MB containing catalyst was initially stirred vigorously for 60 min in dark to attain adsorption-desorption equilibrium before exposure to visible light (200W tungsten lamp). At given time intervals, the solution was taken from the reaction container with a pipette and the change of residual concentration of MB dye solution was monitored by UV-visible spectrophotometer (UV-1800, Shimadzu, Kyoto, Japan) at a characteristics wavelength of 664 nm. The photocatalytic activity of the fabricated thin films over MB dye was analyzed with the help of the following Equation (13): Degradation Efficiency (%) = (C 0 − C t )/C 0 × 100 (13) where C 0 is the initial concentration of MB before light irradiation and C t is the concentration after light irradiation at time t. Similarly, the photodegradation efficiency of thin films photocatalysts under UV light irradiation was carried out by the above-mentioned method using the 60 W low pressure mercury lamp. The effect of oxidant H 2 O 2 addition on photodegradation efficiency over MB dye under UV light illumination and thin films reusability were also investigated.

Computational Methods
To investigate the nitrogen doping effect on ZnO, we studied the electronic properties of ZnO and N-doped ZnO. Consequently, a hexagonal wurtzite structure of ZnO with the space group P6 3 mc and C 4 6v symmetry was modeled. The size of supercell 2 × 2 × 1 was considered for 16-atom Zn 8 O 8 . We replaced one oxygen atom by a nitrogen to get 6.25% N-doped ZnO (Zn 8 N 1 O 7 ). The cell volume and atomic relaxation for both Zn 8 O 8 and Zn 8 N 1 O 7 were carried out until the forces on each atom were below 0.01 eVÅ • −1 . The total energy of the system was converged within 10 −5 eV. The plane waves were expanded up to a cutoff energy of 450 eV. The Brillouin zone was considered using 4×4×1 k-point Monkhorst-Pack mesh. The exchange-correlation interaction in the generalized gradient approximation (GGA)+U method with Perdew, Burke, and Ernzerh of (PBE) function was used for geometry optimization and energy calculation [58]. The effective Hubbard U values 10 and 6 were used for Zn-3d and O-2p, respectively [59]. We adopted the same U value of O for N atom in Zn 8 N 1 O 7 . The valence electron configurations Zn3d 10 4s 2 , O2s 2 2p 4 and N2s 2 2p 3 were considered for pseudo potential construction. All density functional theory (DFT) based calculations were performed using the projector augmented wave (PAW) pseudo potentials, as implemented in the VASP code [60].

Conclusions
A thin film of N-doped ZnO/CNT nanocomposite was fabricated by a low cost sol-gel drop coating protocol and used to achieve photo-catalytic degradation of MB dye under visible and UV light irradiation, respectively. Experimental results of the XRD and FESEM studies indicated that ZnO was formed in nano scale successfully without any impurities in the hexagonal wurzite crystal phase. Findings of experimental and computational studies of lattice parameters of prepared thin films with the addition of nitrogen and CNT into the ZnO crystal do not show any significant differences, which indicate no change of crystal lattice of ZnO. Nearly 100% degradation of a 5 mgL −1 MB dye solution at neutral pH was achieved by an N-doped ZnO/CNT photocatalyst after 4 h of irradiation by UV light. The H 2 O 2 oxidant effect revealed the 100% MB dye degradation efficiency for an N-doped ZnO/CNT nanocomposite thin film within 60 min under UV light irradiation. The Ndoped ZnO/CNT thin film may possibly be reused, which means that the photo-catalytic degradation process could be functioned at a fairly low cost. Overall, the findings in this report can be beneficial and supportive in designing up a scalable and practical way for industrial wastewater treatment.