Molybdenum impregnated g-C3N4 nanotubes as potentially active photocatalyst for renewable energy applications

Molybdenum (Mo) impregnated g-C3N4 (Mo-CN) nanotubes are fabricated via a thermal/hydrothermal process to augment photoelectrochemical properties during solar-driven water-splitting (SDWS) reactions. Graphitic-C3N4 is an attractive material for photocatalysis because of its suitable band energy, high thermal and chemical stability. The FE-SEM and HR-TEM comprehend the nanotube-like morphology of Mo-CN. The spectroscopic characterization revealed bandgap energy of 2.63 eV with high visible-light activity. The x-ray diffraction of pristine g-C3N4 and Mo-CN nanotubes discloses the formation of triazine-based nanocrystalline g-C3N4, which remains stable during hydrothermal impregnation of Mo. Furthermore, Mo-CN nanotubes possess high sp2-hybridized nitrogen content, and metallic/oxidized Mo nanoparticles (in a ratio of 1:2) are impregnated into g-C3N4. The XPS analysis confirms C, N, and Mo for known atomic and oxidation states in Mo-CN. Furthermore, high photocurrent efficiency (~ 5.5 mA/cm2) is observed from 5%-Mo-CN nanotubes. That displays efficient SDWS by 5%-Mo-CN nanotubes than other counterparts. Impedance spectroscopy illustrated the lowest charge transfer resistance (Rct) of 5%-Mo-CN nanotubes, which further confirms the fast electron transfer kinetics and efficient charge separation resulting in high photocurrent generation. Hence, 5%Mo-CN composite nanotubes can serve as a potential photocatalytic material for viable solar-driven water splitting.

www.nature.com/scientificreports/ of nanostructured and-or mesoporous g-C 3 N 4 materials 17 and doping with suitable metals or non-metals 1,[18][19][20] are proposed to address these problems 21 . Impregnation or doping metals into g-C 3 N 4 nanostructures are amongst the most suitable strategies to enhance the optical and photoelectrochemical properties of g-C 3 N 4 through the fabrication of innovative heterojunction nanostructures. Furthermore, metal doping can modify the electronic structure of semiconductors and their textural properties, thus improving their photocatalytic activity 21,22 . For instance, the photoelectrochemical performance of g-C 3 N 4 nanomaterials is significantly improved by Fe and Ti doping 11,[23][24][25] . These studies revealed increased surface area, narrower bandgap, well-aligned band structure, and photodegradation because of the enhanced optical absorption and faster rate of charge carrier transfer 11,[23][24][25] . On the other hand, bearing some outstanding characteristics mentioned above, such g-C 3 N 4 photocatalysts showed poor carrier properties, short hole diffusion length, excitation span, weak career mobility, and shallow light penetration depth resulting in decreased water oxidation on exposure to visible light 26,27 . These issues can be addressed by tuning morphology by surface modifications, reducing bang gap energy and overpotentials for enhanced photocurrent density during SDWS. In SDWS, reduced bang gap is important, as the redox reaction occurs in the electronic structure of the photocatalyst, i.e., illustrated by a conduction band (CB) and a valence band (VB) discerned by bandgap energy (E g ). Under visible light, the photocatalyst is excited by photons with energy equal to or greater than the bandgap energy (hv ≥ E g ). Thus, electrons receiving higher energy from the photons are pushed from VB to CB. In turn, the electrons and holes are transferred to the surface of the photocatalyst undergo reduction or oxidation 28 . These characteristics can be achieved either by growing nanosheets and nanotubes or by incorporating metals and corresponding cocatalysts into the host materials or both. As a result, one can alter electronic band structures for better photocatalytic performance and improved electron and hole transport 26 .
Considering the exceptional electrical, optical, and catalytic properties of Molybdenum (Mo) among transition metals for advanced energy applications, comprehensive studies investigate its photocatalytic reccital [29][30][31][32][33][34] . Mo-based materials exhibited excellent electrical conductivity, enhanced charge carrier mobility, and a variable(~ 1.2-2.2 eV) bandgap energy 21,[35][36][37][38] . Guo et al. 39 reported that Mo in the form of Mo(IV) when incorporated into the host materials, not only introduces localized electron-trapping states at the bottom of the conduction band but also elevates the Fermi level towards the defect level, which endows the doped system with enhanced n-type characteristic and the defect state with strong electron-trapping ability. Moreover, a nonuniform distribution of charge density is formed for the Mo-doped materials, facilitating the separation of photoexcited charge carriers. Therefore, the Mo-doped materials exhibit remarkably enhanced photocatalytic activity, making Mo a suitable material to enhance the photocatalytic water splitting nature of g-C 3 N 4 . Different strategies are known to prepare 1-D or 2-D Mo-based nanomaterials, such as mechanical exfoliation 40 , chemical exfoliation 41 , chemical vapor deposition (CVD) 42 hydro, and solvothermal 30,43,44 methods, etc.
Herein, we present a straightforward approach to prepare Molybdenum (Mo) impregnated g-C 3 N 4 nanotubes via a thermal and hydrothermal route. The surface morphology and elemental composition are characterized via FE-SEM, HR-TEM, XRD, and XPS. Which analyses reveal the nanotubes-like morphology of molybdenumdoped g-C 3 N 4 (Mo-CN), high purity, and crystalline structure. Furthermore, high nitrogen content and reduced bandgap prove decisive factors for improved optical and photoelectrochemical properties of Mo-CN nanotubes. Photoelectrochemical (PEC) measurements of Mo-CN nanotubes exhibit good photocurrent generation with excellent stability under 1 Sun solar irradiation source, low charge transfer resistance (R ct ), fast electron transfer kinetics, and efficient charge separation. Thus, it supports the hydrothermally fabricated Mo-CN nanotubes with great potential for efficient solar-driven water splitting.

Materials and methods
Materials. High purity, analytical grade chemicals, solvents, and reagents were obtained and used as received without further purifications unless otherwise indicated. Ammonium heptamolybdate tetrahydrate, melamine, ethanol, and acetone were purchased from Millipore Sigma and were used as received. Deionized water was used for all experiments and solutions, including solutions used in photoelectrochemical (PEC) measurements.

Synthesis of molybdenum-impregnated graphitic carbon nitride (Mo-CN) nanotubes.
The synthesis of Molybdenum impregnated carbon nitride (Mo-g-C 3 N 4 ) nanotubes is carried out as follows. It includes the preparation of bulk graphitic carbon nitride (g-C 3 N 4 ) from melamine. First, melamine was annealed in the air for 4 h at 500 °C in a furnace until a yellowish powder (g-C 3 N 4 ) is obtained. Next, the bulk g-C 3 N 4 was exfoliated to prepare g-C 3 N 4 nanosheets by sonication for 2 h in 70% ethanol. A Morphological transformation strategy was followed with slight modification 45 . Afterward, the as-synthesized g-C 3 N 4 nanosheets powder was slowly heated at the rate of 10 °C/min up to 300 °C and held at this temperature for 60 min, and then it was transferred into an ice-water bath. Next, the as-prepared g-C 3 N 4 nanotubes were collected by filtration and vacuum dried at 120 °C for 4 h. In the next step, Molybdenum was impregnated into g-C 3 N 4 nanotubes by the hydrothermal approach. Several studies support the metals doping into g-C 3 N 4 via thermal/hydrothermal treatments 39,[46][47][48] . For a hydrothermal reaction, 5% and 15% aliquots (w/w ratio) of ammonium heptamolybdate tetrahydrate and as prepared g-C 3 N 4 nanotubes powder were mixed in water, respectively. Next, each reaction mixture was sonicated at 60 °C for 1 h. Subsequently, the reaction mixtures were transferred into stainless-steel autoclaves containing Teflon vessels. The hydrothermal reaction proceeded for 24 h at 180 °C. Then, each reaction mixture was centrifuged at 4000 rpm for 5 min. Finally, the products 5% Mo-g-C 3 N 4 and 15% Mo-g-C 3 N 4 were collected, washed with deionized water thrice before drying in a vacuum oven for 2 h at 150 °C. Figure 1 shows a schematic of the formation of Mo impregnated g-C 3 N 4 nanotubes. www.nature.com/scientificreports/ Characterization. X-ray diffraction (XRD) was used to characterize Mo-CN nanotubes via a benchtop MiniFlex X-ray diffraction (mini-XRD) instrument from Rigaku with Cu Kα radiation at 40 kV and 15 mA. XRD patterns were recorded in the range of 10-70° (2θ) at a scanning rate of 3° min -1 . The structural composition and crystalline phases of Mo-CN nanotubes were determined from the XRD library database. Also, the structural composition and elemental speciation of Mo-CN nanotubes were verified by x-ray photoelectron spectroscopy (XPS) using PHI 5000 Versa Probe II spectrometer (UlVAC-PHI), employing Al Kα as the incident radiation source. The C1s (E = 284.5 eV) served as the internal standard. The nanotubes-like morphology of Mo-CN was observed under TESCAN Lyra 3 field emission dual beam (electron/focused ion beam) system combined high-end field-emission scanning electron microscope (FE-SEM). JEOL JEM-2100F transmission electron microscope was used to acquire high-resolution transmission electron micrographs (HR-TEM) at 200 kV. Optical properties of Mo-CN nanotubes were measured by diffused reflectance spectroscopy on Agilent Cary 5000 high-performance UV-Vis-NIR spectrophotometer containing praying mantis accessory with alignment tools and powder cell sample cups. These materials are also tested for photoluminescence studies carried out with fluorolog-3 Imaging Spectrophotometer at an excitation wavelength of 350 nm and a slit width of 2 nm. In contrast, PL emission spectra were acquired by Spectrofluorimeter (JASCO, FP-8500) and FTIR 6700 Nicolet™ Fourier transform infrared (FTIR) spectrometer recorded the vibrational modes in the materials.

Results and discussion
FT-IR spectroscopy. FTIR spectra of pristine g-C 3 N 4 , 5%Mo-CN, and 15%Mo-CN nanotubes are presented  www.nature.com/scientificreports/ C-N(-C)-C (fully condensed) units, respectively. Furthermore, a sharp peak around 810 cm -1 corresponds to the characteristic breathing mode vibrations of triazine units, which further confirms the formation of the g-C 3 N 4 structure, as demonstrated in Fig. 1. These peak assignments agree with the pertinent literature [49][50][51][52] . Furthermore, the broad transmittance peaks exhibited in the range of 3000-3300 cm -1 are attributed to the N-H vibrations, i.e., ν(N-H) stretching. The existence of ν(N-H) band in 5%Mo-CN and 15%Mo-CN samples suggests that g-C 3 N 4 nanotubes remain protonated during the impregnation of Mo metal nanoparticles, which also substantiates the stability of the triazine-based g-C 3 N 4 structure. However, significant differences in the FTIR spectra of pristine and Mo-impregnated g-C 3 N 4 nanotubes can be observed in the peak shifts and intensity changes at 1580, 1650, and 3125 cm −1 , which correspond to stretching and scissoring vibrations of -NH/-NH 2 groups [49][50][51][52] . These changes in the FTIR spectra of Mo-CN nanotubes are attributed to the deposition of Mo nanoparticles on g-C 3 N 4 surfaces, which results in strong metal coordination with g-C 3 N 4 nanotubes. It is envisaged that such interactions enhance the potential photoelectrochemical performance.
Surface morphology. FE-SEM studies the surface morphology of pristine g-C3N4, 5%Mo-CN, and 15%Mo-CN nanotubes, and the respective scanning electron micrographs are shown in Fig. 2. Both pristine g-C 3 N 4 and Mo-CN composite samples exhibit nanotubes-like surface morphology with variable length, thickness, and diameter. However, it is observed that the pristine g-C 3 N 4 nanotubes exhibit the finest surface morphology compared to Mo-CN samples. For example, the average diameter of g-C 3 N 4 nanotubes is 154 ± 28 nm, while their length varies from 500 nm to a few µm, as shown in Fig. 2a, b. Furthermore, the impregnation of Mo nanoparticles on g-C 3 N 4 nanotubes in a hydrothermal process (180 °C for 24 h) also leads to the thickening and growth of g-C 3 N 4 nanotubes. For instance, the average diameter of 5%Mo-CN nanotubes increased to 196 ± 43 nm. However, this effect is more pronounced at the higher concentration of ammonium molybdate heptahydrate, the precursor used for the fabrication of Mo nanoparticles, because 15%Mo-CN nanotubes exhibit a significant increase in the thickness, length, and diameter compared to 5%Mo-CN or pristine g-C 3 N 4 samples. As can be seen in Fig. 2e, f, 15%Mo-CN nanotubes are 5 µm long with relatively smooth walls, and their diameter is 500 ± 75 nm. Thus, ammonium molybdate heptahydrate concentration significantly influences the morphology of Mo-CN nanotubes. X-ray diffraction. XRD investigates the structural purity and crystalline phase analysis of pristine g-C 3 N 4 and Mo-CN nanotubes. Figure 4 shows the XRD patterns of the pristine g-C 3 N 4 , 5%Mo-CN, and 15%Mo-CN nanotubes. The characteristic sharp peak at 27.5° and a hump at 13° 2θ (Cu Kα radiation) corresponding to the (002) and (100) are the reflections of crystalline g-C 3 N 4 layered material that further reveals the formation of triazine-based g-C 3 N 4 structure. That is in good agreement with the ICDD reference database for g-C 3 N 4 , JCPDS 87-1526 21 . Moreover, Suter et al. 53 estimated the XRD patterns of planar-/buckled-layer configurations of triazine-based g-C 3 N 4 from DFT calculations and concluded that the planar graphitic layers with AB-stacking and interlayer spacing of ~ 3.24 Å exhibit the main peak around ~ 27° 2θ with a minor sharp reflection at 24.4° 2θ. We observed similar patterns for pristine g-C 3 N 4 and 5%Mo-CN nanotubes with reflections at 27.5° and 25.2° 2θ and an interlayer spacing of 3.24 Å. In the case of 15%Mo-CN nanotubes, only the main reflection at 27.5° 2θ is observed with no shoulder around ~ 24.4° 2θ, which may be attributed to the modification of the layered structure during the hydrothermal impregnation of Mo nanoparticles. In addition, the degree of crystallization of Mo-doped g-C 3 N 4 catalysts reduces apparently with increasing Mo concentration, also illustrated by Wang et al. 21 for fabrication of Mo dopped g-C 3 N 4 . Nonetheless, our results are consistent with the x-ray data of  X-ray photoelectron spectroscopy. XPS is used to confirm the formation of Mo-CN composite nanotubes and elemental states. Figure 5 shows the XPS survey scan and the characteristic core-level spectra of Mo3d, C1s, N1s, and O1s as a function of the respective binding energy values, which agree with the relevant literature 21,37,55 . The C1s peak at 284.5 ± 0.1 eV is used as a charge reference for the XPS core-level spectra and is usually attributed to the carbon-containing contaminations or adventitious carbons 56 . In the C1s core-level spectrum, as shown in Fig. 5c Three components are identified in the N1s core-level spectrum, as shown in Fig. 5d. These components are distinguished as: (a) a peak at 398.8 ± 0.1 eV (FWHM = 1.9 eV) corresponding to the sp 2 -hybridized aromatic C=N-C, (b) a peak at 400.6 ± 0.1 eV (FWHM = 2.3 eV) attributed to the ternary C-N(-C)-C (fully condensed) units, and (c) a weak peak at 404.3 ± 0.1 eV (FWHM = 3.0 eV) corresponding to the C-NH-C (partially condensed) units 58 . The O1s core-level spectrum reveals three components, as shown in Fig. 5e, which are identified as metal (Mo(IV)) oxides with a characteristic peak at at 530.2 ± 0.1 eV (FWHM = 1.8 eV), metal (Mo(IV)) hydroxides showing a peak at 531.2 ± 0.1 eV (FWHM = 1.8 eV), and adsorbed water appearing at 532.7 ± 0.1 eV (FWHM = 1.8 eV) 59,60 . These results not only confirm the impregnation of Mo nanoparticles on g-C 3 N 4 nanotubes and the formation of Mo-CN composites but reveal the presence of both metallic and oxidic Mo nanoparticles. Chemical speciation of the Mo3d core-level spectrum demonstrates a ratio of 1:2 between the metallic Mo nanoparticles and Mo(IV) oxide/hydroxide species. On the other hand, the characteristic triazine-based structure of g-C 3 N 4 nanotubes is confirmed from the core-level C1s and N1s spectra, which align well the previously published reports 36,38,49,[56][57][58]61 .
UV-visible spectroscopy. UV-visible spectroscopy is performed to study the optical properties of pristine g-C 3 N 4 and Mo-CN composite nanotubes. Figure 6 shows the UV-vis diffused reflectance spectra of different samples. To calculate the bandgap, we have used Kubelka-Munk (K-M) function and Tauc plots 62-64 . The K-M Photoluminescence spectroscopy. The room temperature photoluminescence spectra for the Mo Impregnated g-C 3 N 4 are recorded at an excitation wavelength of 350 nm, as shown in Fig. 6d. The outstanding photocatalytic performance of the 5%Mo-impregnated catalysts can be attributed to its nanotubes-like structure and narrow bandgap, which allows it to harvest light more efficiently. In Mo-CN nanotubes, Molybdenum may act as the photogenerated electron target to reduce the recombination of photogenerated electron-hole pairs. This observation and the separation efficiency of the photogenerated electrons and holes are confirmed by photoluminescence is also supported by impedance analysis (Fig. 8). The maxima of the PL peak are observed around ~ 450 nm, which lies in the visible light region. On the other hand, the pristine g-C 3 N 4 exhibits a strong emission peak at about ~ 450 nm at ambient temperature. It is a known fact that higher fluorescence intensity means more recombination of electron-hole pairs and lower photocatalytic activities 65 . The carrier dynamics of pristine g-C 3 N 4 and 5%Mo-CN and 15% Mo-CN are presented in Fig. 6d. In contrast, all the samples exhibited similar emission peaks in the range of − 450 nm. The emission intensities of 5% Mo-CN and 15% Mo-CN were lower than that of pristine g-C 3 N 4 21,66 . This is indicating the enhanced charge separation, and transfer in Mo impregnated CN nanotubes. The lowest peak intensity of 5% Mo-CN indicated suppressed charge recombination, highest charge separation, and transfer efficiency, thereby definitely preferred the photocatalytic water splitting process.

Photoelectrochemical (PEC) measurements. FTO photoanodes coated with pristine g-C 3 N 4 and
Mo-CN composite nanotubes are used for PEC measurements in a standard three-electrode system consisting of a reference (SCE) electrode, counter (Pt wire) electrode, and working (FTO) electrodes. 0.5 M Na 2 SO 4 at neutral pH (7) is used as the electrolyte solution. The cell is exposed to a solar irradiation source (1 SUN with AM filter 1.5G) at regular intervals to record photoresponse. Based on the initial cyclic voltammetry experi- www.nature.com/scientificreports/ ments, chronoamperometry is performed to study the photoresponse of pristine g-C 3 N 4 and Mo-CN composite nanotubes. Figure 7a shows the photocurrent density vs. time (Jp vs. t) profiles of these nanomaterials. At the same scale, the photoresponse of 5%Mo-CN/FTO electrode is the highest, i.e., ~ 3 µA at an applied potential of 0 V. Consequently, 5%Mo-CN nanotubes exhibit ≥ 5 times higher photocurrent compared to pristine g-C 3 N 4 and 15%Mo-CN nanotubes at 0 V. Furthermore, these nanostructures exhibit reversible photocurrent with excellent stability under on/off visible light illuminations. This, in turn, reflects the successful performance of 5%Mo-CN nanotubes composite material for effectual water splitting processes. The linear sweep voltammetry experiments further complement this study, where the photocurrent density enhances to ~ 5.5 mA at higher potential values recorded for 5%Mo-CN nanotubes. Figure 7b shows comparative linear sweep voltammetry profiles of pristine g-C 3 N 4 and Mo-CN composite nanotubes coated on FTO substrates. The measurements are performed in the potential range of 0-1.5 V against standard Ag/AgCl reference electrode under simulated visible light (1 SUN) illumination at a scan rate of 100 mV/s. Figure 7b shows that the current density of 5%Mo-CN nanotubes increases with applied potential and is estimated to be 5.5 mA/cm 2 . Compared to 15%Mo-CN and pristine g-C 3 N 4 nanotubes, and it is 2-3 times higher. It is believed that the higher the current density, the higher the electron/hole (e − /h + ) separability, which in turn improves the photocatalytic activity. Therefore, the transient photocurrent measurements and linear sweep voltammetry also substantiate the charge transfer dynamics at the g-C 3 N 4 nanotubes and Mo nanoparticles interface.
Consequently, the highest photoresponse and current density of 5%Mo-CN nanotubes at various potentials can be ascribed to the lower bandgap, high surface area, and superior interfacial charge transfer dynamics compared to 15%Mo-CN nanotubes. Moreover, considering the positive potential response, the oxygen evolution reaction (OER) is the preferred photoelectrochemical process observed during the linear sweep voltammetry measurements. Hence, 5%Mo-CN nanotubes demonstrate high photocurrent efficiency and photochemical oxygen evolution performance that is the most impressive achievement acclaimed in PEC studies.
The electrochemical impedance spectroscopy (EIS) was performed to investigate the interfacial charge transfer kinetics, such as the efficiency of blocking the recombination of photoinduced electron and hole pairs by the pristine g-C 3 N 4 nanotubes and Mo-impregnated g-C 3 N 4 nanotubes respectively. Figure 8a, b shows the EIS spectra of pristine g-C 3 N 4 and 5% Mo-g-C 3 N 4 and 15% Mo-g-C 3 N 4 samples at higher frequency Fig. 8a and low frequencies Fig. 8b. Several studies [67][68][69] illustrated that the semicircle diameter in a Nyquist plot is proportional to the charge-transfer resistance of the material under observation, providing valuable information on charge transfer processes. Hence, the smallest semicircle is observed for 5% Mo-g-C 3 N 4 in the Nyquist plot compared to 15% Mo-g-C 3 N 4 and g-C 3 N 4 , as shown in Fig. 8b. Thus, illustrating the low resistance for charge transport for 5% Mo-g-C 3 N 4 . On the other hand, the semi-circular Nyquist plots showed the largest diameter for g-C 3 N 4 and then for 15% Mo-g-C 3 N 4 . Therefore, we can conclude that 5% Mo-g-C 3 N 4 nanotubes possess the lowest charge transfer resistance (R ct ) among g-C 3 N 4 and 15% Mo-g-C 3 N 4 , which further confirms the fast electron transfer kinetics and thus efficient separation of photogenerated e − -h + pairs. Additionally, the small arc radius also demonstrates the fast interfacial charge transfer efficiency for 5% Mo-g-C 3 N 4 . Thus, the EIS observations are in line with the enhanced photocatalytic activity of the 5% Mo-g-C 3 N 4 Nanotubes compared to its counterparts (Fig. 7a, b). This is credited to photogenerated carriers in the Mo-dopped materials with higher separation efficiency than the pristine g-C 3 N 4 nanotubes and 15% MO-g-C 3 N 4 nanotubes. It is important to mention here, sometimes by increasing the amount of a doped metal from a certain level, the photocatalytic recital could not be improved. Guo et al. demonstrated this phenomenon; with higher doping amounts of metal, new recombination centers for carriers electrons and holes are created, limiting the photocatalytic performance of the photocatalyst 39 . Therefore, the lower performance of g-C 3 N 4 nanotubes is certainly due to the absence of Mo doped metal, but for 15% Mo-g-C 3 N 4 nanotubes, the decreased photoactivity could be due to higher Mo content.  (Fig. 6). Its bandgap is satisfactorily high to overcome the endothermic atmosphere of the water-splitting reaction. It is commonly recognized that theoretically estimated energy must be higher than the endothermic character of the water-splitting reaction to produce H 2 from water. This is further explained in Eqs. (1) and (2), as given below: From these equations, it is clear that a photocatalyst requires a minimal amount of energy equivalent to 1.23 V of redox potential to initiate the water-splitting reaction, i.e., H 2 production. Hence, to start solar-driven water splitting with a photoelectrocatalyst, the material's bandgap must be larger than this minimal energy. Sometimes, the intermediates formed during the 4-electron transfer reaction have higher energy; therefore, a certain inherent overpotential must be well-thought-out. This makes photons exceeding about 1.8 eV energy or higher that is practically suitable for water splitting 21,25,37,70 . Light absorption in photoelectrochemical reactions is associated with various instant molecular relaxations taking place between the material and the solvent thus, the bandgap of the photocatalyst should be in such a range that a photogenerated electron must have sufficient reduction power to reduce water to H 2 and the photogenerated hole has enough oxidation strength to oxidize water to O 2 21,37,70 . Figure 2S (Supplementary information) shows the proposed mechanism of electron/hole (e − /h + ) transfer in the 5%Mo-CN nanotubes. It is believed that the observed bandgap energy of 2.63 eV carries both the half-cell reactions independently. This type of mechanism is rarely observed in organic semiconducting material and indicates that 5%Mo-CN nanotubes are stable in water and can perform efficient visible-light-driven water splitting. This is because of the suitable microstructure of 5%Mo-CN nanotubes, smaller size and thereby, larger surface area (compared to 15%Mo-CN nanotubes), and better interfacial charge transfer dynamics. Also, the presence of a high concentration of sp 2 -hybridized nitrogen atoms, responsible for electron localization, ultimately shows high photocurrent efficiency. Furthermore, the impregnation of Mo nanoparticles on the g-C 3 N 4 surface also enhances the visible light-harvesting ability of g-C 3 N 4 nanotubes materials and optimized charge separation, which is also attributed to the reduction in bandgap energy (Eg) level 39 .
The crystalline nature of 5%Mo-CN nanotubes is also believed to improve the PEC properties by promoting the kinetics of charge diffusion in both the bulk and on the surface 11,71 . It is generally observed that g-C 3 N 4 is an effective photocatalyst for water splitting under visible light, but its efficiency can be further enhanced by introducing some sacrificial electron donor/acceptor or by doping it with transition metal or noble metal catalysts 11,71 . In this study, the impregnation of Mo nanoparticles on g-C 3 N 4 nanotubes also enhanced the photocurrent efficiency in a similar way for water reduction into H 2 or water oxidation into O 2 . It is observed that the oxygen evolution reaction (OER) is dominant in this case in the positive potential range due to kinetic effects 11,38,71,72 . Hence 5% Mo-CN nanotubes generated better photocurrent under 1 SUN visible light irradiation.

Conclusions
This article reports a facile strategy of synthesizing Mo impregnated g-C 3 N 4 nanotubes from melamine and illustrated its characterization by different spectroscopic, microscopic, and electrochemical techniques. As a result, 5%Mo-CN nanotubes exhibit low bandgap (2.63 eV), high nitrogen concentration, and excellent (1) 2H 2 O → O 2 + 4H + + 4e − + 0.82V (2) 2H + + 2e − → H 2 − 0.41 V Figure 8. EIS studies of as prepared g-C 3 N 4 Nanotubes and Mo-Impregnated g-C 3 N 4 Nanotubes Photocatalyst (a) Magnified view at the high frequency of 5% Mo-g-C 3 N 4 nanotubes presenting lowest charge transfer resistance (b) The semicircle of % Mo-g-C 3 N 4 is at a lower position in the Nyquist plot as compared to 15% Mog-C 3 N 4 and g-C 3 N 4 at low frequency. www.nature.com/scientificreports/ photoelectrochemical properties compared to pristine g-C 3 N 4 and 15%Mo-CN. Furthermore, the XRD pattern of 5%Mo-CN nanotubes revealed the crystalline structure of triazine-based g-C 3 N 4 that remained intact during the hydrothermal process. Simultaneously, XPS discloses the impregnation of Mo nanoparticles on g-C 3 N 4 nanotubes and the presence of metallic Mo, Mo(IV), C, and N in 5%Mo-CN nanotubes. These characteristics of 5%Mo-CN nanotubes ultimately helped in achieving better optical and photoelectrochemical properties. Furthermore, the 5%Mo-CN nanotubes exhibit a high current density of 5.5 mA/cm 2 and stable, repeatable photoresponse under visible-light illumination. The improved photocatalytic recital of 5% Mo-CN could be explained by its structural feature and carrier kinetic properties. When irradiated under visible light, increased light absorption can be achieved, and this is credited to the narrowed bandgap energy by Mo doping in g-C 3 N 4 nanotubes. Hence, more photogenerated carriers can easily jump in the conduction band due to decreased bandgap energy. Furthermore, enhanced charge carrier kinetics also improved the charge transfer and separation ability and made abundant electrons for redox reactions. The nanotube-like morphology of Mo impregnated g-C 3 N 4 enlarged the surface area, providing more surface active sites and channels for photocatalytic activities. Henceforth, 5%Mo dopped g-C 3 N 4 nanotubes composite is a potential photocatalyst for renewable energy applications.