Synergistic Interfacial Engineering of Green TiO₂@g‑C₃N₄ for Noble-Metal-Free Photocatalytic H₂ Generation under Natural Sunlight

Abstract

The plausible design for accurate, sunlight-responsive photocatalysts for feasible hydrogen generation remains a critical focus in solar energy conversion exploration. In this work, a green TiO₂@g-C₃N₄ heterojunction was produced and systematically assessed for enhanced photocatalytic H₂ evolution in an aqueous glycerol solution according to direct sunlight. Structural, morphological, and interfacial features were extensively determined utilizing TEM, XPS, and photoluminescence (PL) spectroscopy, verifying the accomplished generation of a robust heterojunction with strong Ti–N–C interfacial bonding. Gas chromatography confirmed the absence of H₂ and O₂ prior to irradiation, guaranteeing precise baseline assessments. Pristine g-C₃N₄ and green TiO₂ displayed hydrogen generation rates of 3.75 and 44.5 μmolh^–1, respectively, while green TiO₂ most notably surpassed traditional white TiO₂ (<40 μmolh^–1) owing to oxygen vacancies boosting visible-light absorption. Interestingly, the designed green TiO₂@g-C₃N₄ heterostructure attained a superior H₂ evolution rate of 110.25 μmol h^–1, a 29.4- and 2.4-fold improvement over g-C₃N₄ and green TiO₂, respectively. This elevated photocatalytic achievement is credited to escalated interfacial charge movement, prolonged charge carrier lifetimes, and upgraded surface redox kinetics. This work offers beneficial insights into the rational design of g-C₃N₄-based heterostructures for highly effective, noble-metal-free, solar-driven hydrogen production.

1. Introduction

Rising environmental concerns and the depletion of fossil fuels have made the development of sustainable energy solutions a global necessity. , Researchers have been putting a lot of effort into creating a renewable, affordable, sustainable, and environmentally benign energy source substitute for fossil fuels. , Among the various alternatives, hydrogen (H₂) has emerged as a promising clean fuel owing to its high energy density (120–142 MJ kg^–1) and carbon-free nature. It is projected to meet nearly 90% of the world’s energy demand by 2080. At present, over 44.5 million tons of hydrogen are produced annually, mainly via steam reforming and water electrolysis. − However, these methods are costly and demand large amounts of external energy. Molecular H₂ is speculated to be the most beneficial energy route for absorbing sunlight. Water splitting, an acknowledged chemical phenomenon that has sparked the creation of numerous innovations, including electrolyzers, photoelectrochemical (PEC) cells, and photocatalytic processors for particulate catalysts in an aqueous suspension, is a source of this substantial amount of energy. , Solar-driven photocatalytic water splitting offers a practical and eco-friendly strategy, using abundant sunlight and water to produce hydrogen without additional bias or corrosive electrolytes. Yet, the efficiency of this process is hindered by rapid charge recombination, suboptimal band edge alignment, and limited solar absorption. − Sunlight water splitting activities do not need a bias voltage or electrolytes, which are very alkaline/alkaline solutions, like PEC water splitting does. Thus, designing advanced semiconductor photocatalysts with improved light harvesting and charge separation is critical for green hydrogen production.

Since the first effort at water splitting using photocatalysis over Pt/TiO₂ was attempted in 1972, scientists have been extensively at work trying to figure out photocatalytic ways to produce hydrogen from water sources. Several semiconductors, including TiO₂, Cu₂O, SrTiO₃, ZnFe₂O₄, CdS, g-C₃N₄, WO₃, BiVO₄, MOF, and perovskite materials, have been noticed as photocatalysts for H₂ generation. In recent decades, semiconductor-based photocatalysts have gained a lot of attention due to the fact that photocatalysis successfully handles the expanding worldwide ecological and energy emergencies. ,

In photocatalysis, titanium dioxide has emerged as the standard oxide semiconductor. , The conduction and valence band positions align with redox potentials, enabling eco-friendly reactions, supported by TiO₂’s remarkable chemical stability, strong photocorrosion resistance, nontoxic character, and low cost, making it highly promising. Despite extensive efforts, TiO₂ performance is limited by its ∼3.2 eV bandgap, restricting absorption to ∼4% UV light, and by rapid electron–hole recombination, which significantly reduces its photocatalytic efficiency. Designing diverse TiO₂ nanostructures helps suppress charge recombination, while elemental doping effectively shifts its absorption edge from the ultraviolet to the visible region of sunlight, enhancing photocatalytic activity. ,, Creating a heterojunction an interface among two distinct semiconductors that have distinct band gaps with better electron–hole pair segregation is another method for achieving a visible photon responsiveness. The interface that exists between the semiconductor sensitizer and the TiO₂ platform ought to be coordinated in an optimal heterojunction, and the locations of the corresponding VBs and CBs should facilitate it easier to separate the charge carriers generated by photosynthesis.

The unusual electronic composition and excellent thermal and chemical resilience of g-C₃N₄, a well-known 2D nonmetallic inorganic semiconductor exhibiting visible photon response, have garnered significant interest in the discipline of photocatalysis. , A polymeric semiconductor’s comparatively small bandgap (∼2.7 eV) allows for visible light absorption beyond 460 nm. Despite this, g-C₃N₄ has been shown to be capable of photocatalytic water splitting for the creation of H₂. , Nevertheless, restricted electron accessibility and rapid charge recombination hinder its solar conversion effectiveness. Researchers have been actively working to rationally construct materials inspired by multiphase heterostructures over the past ten years to get beyond the drawbacks of using specific semiconductors for the generation of H₂. An ideal photocatalytic heterojunction demands proper band alignment, sunlight utilization, stability, and efficient charge separation. ,

By employing these techniques, the collaborative effect involving TiO₂ and g-C₃N₄ can be leveraged to increase the photocatalytic efficiency. , In addition to TiO₂’s capacity to absorb UV light, g-C₃N₄ may also absorb the visible spectrum, increasing the effectiveness of photocatalytic processes. Additionally, g-C₃N₄ may establish a heterojunction with TiO₂ to minimize the recombination of the photogenerated electron–hole pairs. Water splitting under both ultraviolet and visible light illumination is improved by the resultant nanoheterojunctions’ ability to use a wider spectrum of sunshine. , As documented by Yan and Yang, the productivity of photocatalytic hydrogen synthesis can be increased by employing g-C₃N₄/TiO₂ photocatalysts made by ball-milling and then annealing g-C₃N₄ and TiO₂. Since the CB edge of the TiO₂ is lower than the position of the g-C₃N₄, they hypothesized that photogenerated electrons stimulated by visible photon illumination on the g-C₃N₄ CB would migrate to the location of TiO₂. Wang et al. demonstrated that tailoring TiO₂@g-C₃N₄ microstructures, including contact surfaces, defects, and nanosheets, enhanced photocatalytic efficiency. The optimized heterojunction achieved 436.3 μmol h^–1 g^–1 H₂ production, attributed to favorable band alignment, intimate interfacial contact, and efficient charge carrier separation. Biswal et al. developed a Ti₃C₂/N,S-TiO₂@g-C₃N₄ ternary heterojunction via annealing and ultrasonic impregnation. At 4 wt % Ti₃C₂, it achieved 495.06 μmol h^–1 H₂, enhanced by dual heterojunction formation, efficient carrier separation, and dual charge transfer pathways. Nevertheless, g-C₃N₄ and TiO₂ do not form an intact and intimate interface with these particular heterojunction substances, which results in slow carrier exchange at the heterojunction interfaces. Therefore, relatively minimal enhancement in photoactivity in TiO₂@g-C₃N₄ heterojunction materials. If g-C₃N₄ and TiO₂ form an intimate interface, the movement of the photogenerated carriers across the heterojunction interface will be enhanced.

Moreover, the green template design strategy adopted for TiO₂ preparation offers a sustainable and eco-friendly. Plant extracts act as natural reducing and stabilizing agents, eliminating the need for toxic reagents while simultaneously controlling particle size, morphology, and dispersion. Unlike conventional chemical methods, the green template strategy ensures an eco-friendly, cost-effective synthesis while maintaining precise control over TiO₂ morphology and dispersion. Coupled with the rational design of the TiO₂@g-C₃N₄ interface, this work introduces a sustainable pathway to produce a high-performance photocatalyst. The combination not only reduces the environmental footprint of catalyst preparation but also enhances hydrogen generation efficiency, thereby offering a practical model for merging environmental safety with energy conversion effectiveness.

In this study, two-dimensional g-C₃N₄ nanosheets were fabricated through a thermal polymerization strategy and subsequently integrated with TiO₂ nanoparticles synthesized via a green template-assisted solvothermal approach, forming a TiO₂@g-C₃N₄ heterojunction nanoheterojunction. The hybrid material was systematically explored for its photocatalytic efficiency in H₂ generation under sunlight. The synergistic interaction between g-C₃N₄ and TiO₂ was found to enhance charge carrier separation significantly, boosting the photocatalytic performance. Furthermore, a plausible photocatalytic mechanism was proposed to elucidate the charge transfer dynamics and active species generation within the TiO₂@g-C₃N₄ heterostructure during H₂ production. This study offers valuable insights into the design of efficient, environmentally benign photocatalysts for sustainable energy applications, emphasizing the potential of heterojunction engineering to optimize photocatalytic hydrogen evolution under ambient sunlight conditions.

2. Experimental Details

2.1. Materials

Melamine (C₃H₆N₆, CAS No. 108–78–1, ≥99%) and titanium­(IV) butoxide (Ti­(OC₄H₉)₄, CAS No. 5593–70–4, 97%) were used as precursors for the production of TiO₂ and g-C₃N₄, accordingly. The solvents were Milli-Q water from a water purification system and absolute ethanol (C₂H₆O, CAS No. 64–17–5, 99.5%). Every chemical was acquired from Sigma-Aldrich and used without additional purification. The green template was prepared from a plant extract that utilized leaves from the Moringa plant (Moringa oleifera) from Wadi Kabir, Muscat, Sultanate of Oman.

2.2. Synthesis of Photocatalyst Materials

2.2.1. Synthesis of g-C₃N₄

Graphitic carbon nitride (g-C₃N₄) was obtained through a one-step thermal polymerization technique employing melamine as the starting material. A total of 1 g of melamine was transferred into a lidded crucible and subjected to thermal treatment in a muffle furnace. The system was heated from room temperature to 550 °C at a consistent rate of 10 °C per minute and held at the target temperature for 3 h to complete the polymerization process.

2.2.2. Synthesis of Green Template-Assisted TiO₂

A green synthesis approach was employed for the preparation of titanium dioxide nanoparticles using Moringa leaves as a natural biotemplate. Initially, 10 g of cleaned, dried Moringa leaves were boiled in 150 mL of deionized water at 80 °C for 2 h. The extract was allowed to cool, filtered, and stored at 4 °C for later use. Separately, 2 mL of Titanium­(IV) butoxide was mixed with 20 mL of ethanol and vortexed to obtain a uniform solution. This mixture was then added dropwise to 50 mL of the prepared plant extract under constant stirring for 2 h. The resulting solution was transferred to a Teflon-lined autoclave and heated at 160 °C for 12 h. The formed precipitates were repeatedly washed with water and ethanol over 2 days, then dried at 80 °C for 12 h. Finally, the product was calcined at 450 °C for 3 h at a heating rate of 3 °C/min.

2.2.3. Synthesis of 1:1 Mass Ratio TiO₂@g-C₃N₄ Nanoheterojunctions

To fabricate the nanoheterojunctions, synthesized TiO₂ and graphitic carbon nitride (g-C₃N₄) were combined in equal mass ratios. Both materials were placed together into a conical flask (capacity 250 mL), followed by the addition of 50 mL of deionized water and 20 mL of ethanol. The suspension was stirred continuously for 2 h to ensure uniform dispersion of the components. After thorough mixing, the solution was transferred into a Teflon-lined autoclave (capacity: 150 mL) and subjected to solvothermal treatment at 160 °C for 12 h. Upon completion, the resulting mixture was carefully washed several times with deionized water and ethanol to eliminate residual impurities. Finally, the nanoheterojunction was dried in an oven at 80 °C for 12 h, yielding well-integrated TiO₂/g-C₃N₄ nanostructures ready for further characterization and photocatalytic application.

2.3. Detailed Materials Characterization

The structural, optical, and compositional characteristics of TiO₂, g-C₃N₄, and the green-synthesized TiO₂@g-C₃N₄ nanoheterojunction were thoroughly investigated using a range of analytical techniques. The optical behavior of the materials was assessed by UV–Visible diffuse reflectance spectroscopy (UV–Vis DRS) using a SHIMADZU UV-2600i spectrophotometer. This instrument, capable of extending measurements into the near-infrared region up to 1400 nm, utilized BaSO₄ as a baseline reference. To examine the surface morphology and elemental distribution, Field Emission Scanning Electron Microscopy (FESEM) was conducted with a JEOL JSM-7001F microscope, coupled with Energy Dispersive X-ray Spectroscopy (EDS) for elemental analysis. The chemical bonding and functional groups present in the samples were identified using Fourier Transform Infrared (FTIR) spectroscopy over a scanning range of 400–4000 cm^–1, performed on an FTIR α (ECOART) spectrometer. The Particle structures of the synthesized materials were determined through X-ray diffraction (XRD) using an X’Pert Pro diffractometer, equipped with a Cu Kα radiation source (λ = 1.54056 Å). Diffraction patterns were recorded within a 2θ range from 10° to 90° to identify distinct crystal phases. Surface elemental states and chemical compositions were analyzed by X-ray Photoelectron Spectroscopy (XPS) using a multiprobe system from Omicron Nanotechnology (Germany), with data interpretation conducted through Casa XPS software. Additionally, the photoluminescence (PL) characteristics of the nanomaterials were examined using a Shimadzu RF-5301PC spectrophotometer, employing an excitation wavelength of 330 and 310 nm to evaluate their emission behavior, g-C₃N₄, TiO₂@g-C₃N₄, and TiO₂, respectively.

2.4. Photoelectrochemical Properties

Photoelectrochemical (PEC) measurements were performed using a conventional three-electrode system connected to a CHI 760D electrochemical workstation (CH Instruments). The working electrode was prepared by coating fluorine-doped tin oxide (FTO) glass (2 cm × 2 cm) with a suspension of the photocatalyst (TiO₂, g-C₃N₄, or TiO₂@g-C₃N₄), while a platinum wire and an Ag/AgCl electrode served as the counter and reference electrodes, respectively. To fabricate the working electrode, 200 μL of a 0.83 g/L photocatalyst suspension was drop-cast onto the FTO substrate and allowed to dry at room temperature for 12 h. The photocatalyst suspension was prepared by dispersing 3 mg of the material in 3 mL of ethanol and 500 μL of Nafion solution, followed by sonication for 30 min to achieve a uniform dispersion. A 500 W xenon lamp equipped with an AM 1.5 filter was used as the simulated solar light source. The transient photocurrent responses of the different photocatalysts were evaluated in a 0.5 M Na₂SO₄ electrolyte at an applied potential of 0.4 V versus Ag/AgCl under intermittent light irradiation. Additionally, electrochemical impedance spectroscopy (EIS) measurements were carried out under the same conditions to assess the electrodes’ charge transfer resistance and interfacial properties. The EIS spectra were recorded at 0.4 V versus Ag/AgCl over a frequency range of 100 kHz to 0.1 Hz under illumination, providing insight into the photoelectrochemical performance of the synthesized materials.

2.5. Photocatalytic Experimental Setup for H₂ Production

The hydrogen evolution performance of pristine TiO₂, g-C₃N₄, and their TiO₂@g-C₃N₄ nanoheterojunctions was systematically evaluated under natural sunlight, as demonstrated in Scheme . For each experiment, 10 mg of the photocatalyst powder was uniformly dispersed in 50 mL of an aqueous solution containing 10 vol.% glycerol, serving as a sacrificial electron donor. The suspension was prepared within a Kjeldahl-type quartz photoreactor (180 mL capacity), which was externally wrapped with aluminum foil to maintain dark conditions. The mixture was magnetically stirred at room temperature for 15 min to promote adsorption equilibrium between the photocatalyst surface and reactants. After sealing the reactor with a Suba-Seal rubber septum, it was evacuated for 15 min to remove trapped gases and subsequently purged with high-purity nitrogen gas to establish an oxygen-free environment. A Shimadzu GC-2014 gas chromatograph equipped with a molecular sieve (5 Å) column and a thermal conductivity detector (TCD) was used for hydrogen detection, with ultrahigh purity nitrogen as the carrier gas. A baseline (zero-hour) chromatogram was obtained prior to irradiation to confirm the absence of hydrogen and oxygen. After removing the aluminum covering, the photocatalytic hydrogen evolution experiments were conducted by exposing the reactor to direct sunlight on clear days, between 10 a.m. and 2 p.m. The system was irradiated continuously for 4 h, with gas sampling every 2 h using a gastight Hamilton micro syringe (250 μL). The solar irradiance was monitored hourly using a lux meter, recording an average intensity of 112,600 ± 4000 lx. This procedure was consistently applied to all photocatalysts for comparative evaluation.

1. Schematic Illustration of the Synthesis of g-C₃N₄, Green TiO₂, and TiO₂@g-C₃N₄ Nanoheterojunction Photocatalyst .

^a [The Scheme is drawn using BioRender].

3. Results and Discussion

3.1. Characterization of Nanoheterojunction Photocatalyst

To comprehensively uncover the microstructure, optical behavior, and chemical makeup of TiO₂, g-C₃N₄, and their TiO₂@g-C₃N₄ heterojunctions, a suite of advanced techniques including XRD, FT-IR, XPS, DRS, PL spectra, SEM, TEM, EDX, and elemental mapping were employed, offering deep insights into their structural and functional characteristics. The crystallographic properties and phase purity of TiO₂, g-C₃N₄, and their TiO₂@g-C₃N₄ nanoheterojunctions were thoroughly characterized by powder X-ray diffraction (XRD). The diffraction patterns, shown in Figure a, reveal that the prominent peaks of TiO₂ correspond to the anatase phase, with diffraction peaks at 2θ values of 25.14°, 37.67°, 47.91°, 54.53°, 62.61°, 69.80°, and 79.90°, which are attributed to the (101), (004), (200), (211), (204), (116), and (215) planes, respectively. These peaks are in excellent agreement with the standard JCPDS card #21–1272, ,, confirming the Particle nature of TiO₂. For g-C₃N₄, characteristic diffraction peaks at 2θ = 12.08° and 27.71° were observed, tris-s-triazine phase, corresponding to the (100) and (002) planes, respectively, in line with JCPDS card #87–1526. ,, The absence of additional diffraction peaks suggests that the materials are highly pure, with no detectable impurities or secondary phases. The diffraction patterns of the TiO₂@g-C₃N₄ nanoheterojunctions show an influence of TiO₂ on the peak intensities, with TiO₂ dominating due to its sharper and more defined peaks. Notably, the (101) peak of TiO₂ at 25.14° exhibits broadening and a shift to higher 2θ values (Figure b). The observed higher-angle shift in the XRD pattern of TiO₂@g-C₃N₄ is attributed to lattice strain and slight contraction caused by strong interfacial interaction between TiO₂ and g-C₃N₄. The formation of the heterojunction induces structural distortion and modifies the local electronic environment, leading to a decrease in interplanar spacing (d-spacing). Such shifts have been commonly reported in TiO₂-based heterostructures and are consistent with the formation of a stable TiO₂-g-C₃N₄ interface. The reduction in crystallite size, common in nanoheterojunctions, leads to broader XRD peaks due to the inverse relationship between peak width and particle size. Additionally, lattice strain may arise from the interaction between TiO₂ and g-C₃N₄, as the materials may have mismatched lattice parameters or different thermal expansion properties. This strain disturbs the regular atomic arrangement in the crystal, causing both peak broadening and shifts to higher 2θ angles. In the case of g-C₃N₄, the sharp (100) peak originates from the in-plane arrangement of tri-s-triazine units, which are relatively well ordered, whereas the broad (002) reflection corresponds to interlayer stacking. The broadness of the (002) peak reflects poor crystallinity along the stacking direction because the layers are held together only by weak van der Waals forces, leading to structural disorder in that dimension. For the TiO₂@g-C₃N₄ composite, there is the weak additional reflections around 2θ ≈ 19° and 30° which do not appear in either pure TiO₂ or g-C₃N₄. Since they are absent in the green-synthesized TiO₂, their origin is most likely associated with structural rearrangements that occur during the formation of the heterojunction. The intimate interfacial contact between TiO₂ nanoparticles and g-C₃N₄ nanosheets can introduce local lattice strain, defect sites, or partial ordering of atoms at the interface, which can give rise to new diffraction features. As these peaks are very weak, they represent only minor structural modifications rather than significant impurities. Importantly, such interfacial distortions do not hinder the photocatalytic activity; in fact, they may provide additional defect states that promote charge separation and transport, thereby contributing positively to the overall performance.

1.

(a, b) Powder XRD patterns, (c) FT-IR analysis, (d) XPS survey spectra of pristine and nanoheterojunction photocatalyst, and XPS analysis of (e) Ti 2p, (f) O 1s, (g) C 1s, and (h) N 1s elements.

The crystallite sizes (D) were calculated using the Debye–Scherrer eq , providing insights into the average particle size of the synthesized materials.

$$D_{\mathit{hkl}}=K\lambda /\beta \cos \theta$$ 1

In which K represents the shape factor (0.94), λ indicates the X-ray wavelength, β denotes the full width at half-maximum (fwhm) of the diffraction peak, and θ appears the Bragg angle, was adopted to derive the average dimensions of the grains of pure green TiO₂, pristine g-C₃N₄, and the TiO₂@g-C₃N₄ nanoheterojunction. According to Figure S2 in the Supporting Information, the predicted crystallitee sizes are 11.08 nm for TiO₂, 19.44 nm for g-C₃N₄, and 18.83 nm for the TiO₂@g-C₃N₄ nanoheterojunction. When TiO₂ and g-C₃N₄ interact during formation, it may promote the proliferation of TiO₂ crystallites and prevent their full dispersion, leading to greater particle sizes in the combined substance. This could explain the increase in particle size seen in the TiO₂@g-C₃N₄ heterojunction when contrasted with pure TiO₂.

Fourier Transform Infrared (FTIR) spectroscopy was employed to investigate the functional groups present in the synthesized TiO₂ nanoparticles, g-C₃N₄, and their TiO₂@g-C₃N₄ nanoheterojunctions. The recorded spectra, obtained in the wavenumber range of 400–4000 cm^–1, are displayed in Figure c. A prominent absorption band observed around 436 cm^–1 is attributed to the characteristic Ti–O–Ti stretching vibrations, which are indicative of the anatase phase of TiO₂, typically appearing within the fingerprint region of 420–650 cm^–1. ,, Additionally, a distinct peak at 803 cm^–1 corresponds to the out-of-plane bending vibrations of C–H groups and the triazine ring units in the g-C₃N₄ framework. The absorption band near 1593 cm^–1 is associated with N–H bending vibrations and contributions from –CC– stretching in aromatic structures. Moreover, a crest at 1618 cm^–1 can be assigned to the stretching vibrations of CO groups and CN bonds within the tri-s-triazine rings of g-C₃N₄. A broad absorption band around 3120 cm^–1 is attributed to N–H stretching vibrations, indicative of the amine functional group present in g-C₃N₄. , The FTIR spectra of the TiO₂@g-C₃N₄ nanoheterojunctions exhibited all the major characteristic bands of both TiO₂ and g-C₃N₄, though with reduced intensity as the TiO₂ content increased. This attenuation suggests the formation of heterojunction interfaces between the two components. Additionally, a noticeable shift in the absorption bands toward higher wavenumbers was observed in the heterojunctions compared to pure g-C₃N₄, attributed to strong interfacial interactions and electronic coupling between TiO₂ and g-C₃N₄. ,

X-ray photoelectron spectroscopy (XPS) was comprehensively employed to investigate the surface chemical composition, oxidation states, and interfacial interactions within the as-prepared TiO₂, g-C₃N₄, and TiO₂@g-C₃N₄ nanoheterojunctions. The wide-scan survey spectra obtained in the binding energy range of 0–1200 eV (Figure d) confirmed the presence of Ti, O, C, and N elements, consistent with the expected composition of the nanoheterojunction. The detection of these elements without any additional impurity signals suggests high purity and successful integration of TiO₂ and g-C₃N₄ components within the heterojunction nanostructure. The high-resolution Ti 2p spectrum (Figure e) for both pure TiO₂ and the nanoheterojunction revealed two characteristic peaks positioned at binding energies of approximately 458.8 and 464.1 eV, corresponding to Ti 2p_3/2 and Ti 2p_1/2, respectively. ,, The observed spin–orbit splitting of about 5.3 eV is in good agreement with the tetravalent Ti⁴⁺ state in the anatase phase, indicating the preservation of TiO₂’s oxidation state upon heterojunction formation. The O 1s core-level spectrum (Figure f) exhibited a dominant peak near 529.7 eV, which can be assigned to lattice oxygen (Ti–O–Ti bonds) in the TiO₂ framework. In the TiO₂@g-C₃N₄ nanoheterojunction, a slight shift and broadening of this peak were observed, implying possible interaction or chemical bonding between TiO₂ and the g-C₃N₄ matrix, likely through surface hydroxyl groups or interfacial oxygen linkages. ,, The C 1s spectrum for g-C₃N₄ (Figure g) presented three notable peaks at around 284.6, 286.9, and 288.1 eV. These peaks are attributed to graphitic carbon, sp²-bonded CN within the heptazine rings, and –N–CN aromatic structures. In the heterojunction, slight shifts in these peaks suggest electronic interactions and charge redistribution at the TiO₂/g-C₃N₄ interface. Similarly, the N 1s spectrum in Figure h displayed peaks at 396.3, 398.9, and 401.7 eV, assigned to sp²-hybridized N in C–NC, tertiary nitrogen (N–(C)₃), and amino-functionalized C–N–H groups, respectively. ,, In the XPS spectra, the C 1s and N 1s peaks of pristine g-C₃N₄ appear broader because of the coexistence of multiple bonding states (C–NC, C–(N)₃, N–CN, and graphitic N) and the presence of structural defects, which lead to overlapping signals and peak broadening. Upon formation of the TiO₂@g-C₃N₄ composite, these peaks become relatively narrower, reflecting a more uniform chemical environment. This change is due to interfacial interactions between TiO₂ and g-C₃N₄, which reduce surface disorder and stabilize the bonding configuration of carbon and nitrogen atoms. The narrowing of the peaks thus indicates improved structural uniformity and stronger electronic coupling at the heterojunction interface, which is beneficial for charge separation and photocatalytic activity. In the heterojunction, changes in peak positions and intensities indicate chemical bonding and strong interfacial coupling between TiO₂ nanoparticles and the g-C₃N₄ matrix, supporting the formation of a heterojunction structure that enhances charge transfer efficiency across the interface.

The optical properties of the synthesized nanomaterials were systematically investigated using UV–Vis diffuse reflectance spectroscopy (UV–Vis DRS) to evaluate their light absorption characteristics and band gap energies. The diffuse reflectance spectra of TiO₂, g-C₃N₄, and the TiO₂@g-C₃N₄ nanoheterojunction are presented in Figure a. Pure TiO₂ exhibited strong absorption in the ultraviolet region, with an absorption edge at approximately 390 nm, which is characteristic of its wide band gap semiconductor nature. In comparison, g-C₃N₄ demonstrated a broader absorption profile extending well into the visible region due to its relatively narrower band gap. Interestingly, the introduction of g-C₃N₄ into TiO₂ significantly modified the optical absorption behavior of the resulting nanoheterojunction. This coupling leads to the formation of a heterojunction interface, facilitating better electronic interactions between the two semiconductors. The optical band gap energies (Eg) of the materials were calculated from the diffuse reflectance data using the Kubelka–Munk function and Tauc plot method.

$$\alpha h\nu =A{(h\nu -E_{\mathrm{g}})}^n$$ 2

where α is the absorption coefficient, h is Planck’s constant, ν is the frequency of light, A is a proportionality constant, and n depends on the type of electronic transition (n = 2 for direct transitions and n = 1/2 for indirect transitions).

2.

(a) UV–vis DRS absorption spectra, (b) Bandgap energies by Tauc plot, and (c) Photoluminescence spectra of pristine g-C₃N₄, green TiO₂, and TiO₂@g-C₃N₄ heterojunction photocatalysts.

The predicted band gap energies varied from 2.69 eV for g-C₃N₄ to 3.18 eV for TiO₂, as seen in Figure b. This result confirms the successful formation of a heterojunction, enhancing visible light absorption and promoting efficient charge carrier separation, thereby improving photocatalytic performance. Moreover, an energy level diagram (ELD) was constructed to evaluate the relative positions of the conduction and valence band edges of the synthesized catalysts, based on the experimental data obtained from Mott–Schottky measurements. Furthermore, the potential barrier positions for both TiO₂ and g-C₃N₄ nanomaterials were estimated using the following eqs and , enabling an approximate determination of the conduction band (E _CB) and valence band (E _VB) potentials ,,

$$E_{\mathrm{VB}}=\chi -E_0+(1/2E_{\mathrm{g}})$$ 3

$$E_{\mathrm{CB}}=E_{\mathrm{VB}}-E_{\mathrm{g}}$$ 4

In these equations, χ represents the absolute electronegativity of the semiconductor, calculated as the geometric mean of the constituent atomic electronegativities (Ti: 3.51 eV, O: 7.54 eV, C: 6.27 eV, and N: 7.30 eV), sourced from Pearson’s Absolute Electronegativity values. E ₀ corresponds to the energy of free electrons on the hydrogen scale (4.5 eV), while E _g is the experimentally determined band gap energy of the material. The calculated E_CB and E_VB values provide insights into the charge transfer dynamics and photocatalytic activity of the materials.

Accordingly, the valence and conduction band positions for pure TiO₂ were estimated to be +2.92 eV and −0.26 eV, respectively. In comparison, the band edge potentials for pristine g-C₃N₄ were predicted to be +1.57 eV for the valence band and −1.12 eV for the conduction band (detailed in Supporting Information Table S1). ,, The results offer valuable theoretical support for understanding the enhanced photocatalytic efficiency of the heterojunction material, correlating with its band structure alignment and charge separation behavior.

To further examine the separation efficiency of photogenerated charge carriers, photoluminescence (PL) spectra of all the produced nanoheterojunctions were acquired by employing an excitation wavelength of 330 and 310 nm to evaluate their emission behavior, g-C₃N₄, TiO₂@g-C₃N₄, and TiO₂, respectively. (Figure c). The PL emission of TiO₂ in the broad band around 350–500 nm region. , This broad emission was found to be caused by band-to-band transitions and excitonic emission associated with interfacial oxygen vacancies and structural imperfections. , A significant rate of photoinduced electron–hole pair recombination was indicated by the broad and strong emission peak of g-C₃N₄, which was centered at roughly 460 nm. It is interesting to note that the TiO₂@g-C₃N₄ heterojunction displayed a PL profile that was considerably less intense than that of pristine g-C₃N₄ and TiO₂. This quenching of PL intensity demonstrates that the formation of the heterojunction effectively prevents charge carrier recombination. Figure c PL observations provided additional evidence for agreement.

The morphological features, particle size distribution, elemental composition, and elemental spatial distribution of the synthesized green TiO₂, g-C₃N₄, and their nanoheterojunctions were comprehensively characterized using FE-SEM coupled with energy-dispersive X-ray spectroscopy (EDS). As depicted in Figure a, high-magnification SEM images (ranging from 1 μm to 100 nm) revealed that the green-synthesized TiO₂ nanoparticles exhibited predominantly spherical morphologies, with some variation in particle sizes indicative of polydispersity. In contrast, the SEM micrographs of g-C₃N₄, presented in Figure b, confirmed the presence of stacked, flat, two-dimensional sheet-like structures, consistent with its layered graphitic-like morphology. The microstructural examination of the TiO₂@g-C₃N₄ nanoheterojunction, shown in Figure c, demonstrated a hybrid architecture wherein spherical TiO₂ nanoparticles were uniformly anchored onto the surface of g-C₃N₄ nanosheets. The morphological features seen in the SEM examination were further validated by HR-TEM pictures, as demonstrated in Figure d–f. The TiO₂ nanoparticles that were green-synthesized showed up as distinct, sphere-shaped particles with a modest size dispersion (Figure d). On the other hand, g-C₃N₄ showed smooth surfaces and thin, translucent, sheet-like nanostructures (Figure e). A well-dispersed hybrid framework with close contact among the nanoparticles and the 2D nanosheet matrix was confirmed by the TEM analysis of the TiO₂@g-C₃N₄ nanoheterojunction, which showed TiO₂ nanoparticles effectively attached onto the surface of the g-C₃N₄ nanosheets (Figure f). This configuration suggests the formation of intimate interfacial contact between the two phases, facilitating efficient heterojunction formation. The adoption of plant extract in the green synthesis route offered significant advantages in terms of environmental safety and process feasibility. Acting as a natural reducing and stabilizing agent, the plant extract enabled the formation of well-dispersed TiO₂ nanoparticles without altering the intrinsic morphology of the g-C₃N₄ sheets in the resulting nanoheterojunction.

3.

(a–c) SEM, and (d-f) TEM images of green TiO₂, g-C₃N₄ nanosheets, TiO₂@g-C₃N₄ nanoheterojunction, (g) EDS analysis, and (h) Bright-field mapping images of as-synthesized TiO₂@g-C₃N₄ nanoheterojunction, (i) Ti, (j) O, (k) C, and (l) N elements.

EDS elemental mapping and analysis, as shown in Figure g, confirmed the presence of Ti, O, C, and N elements within the nanostructures, consistent with the expected composition of the heterojunction materials. Additionally, traces of Mg, Cl, K, and Al detected in TiO₂ samples were attributed to residual components from the plant-based green template. The homogeneous distribution of the principal elements throughout the TiO₂@g-C₃N₄ heterojunction, verified by bright and well-defined color mapping images in Figure h–l, provided further confirmation of the successful synthesis and effective integration of the two phases.

The microstructural characterization of green-synthesized TiO₂, g-C₃N₄, and their nanoheterojunction (Figure ) provides crucial insights into their structure–activity relationship. The spherical morphology and nanoscale dimensions of TiO₂ offer a high surface-to-volume ratio, enabling efficient photon absorption and rapid surface reaction kinetics. Meanwhile, the stacked, layered morphology of g-C₃N₄ facilitates extended light harvesting in the visible region and provides a large platform for anchoring TiO₂ nanoparticles. In the TiO₂@g-C₃N₄ heterojunction, the uniform dispersion and intimate interfacial contact between TiO₂ nanoparticles and g-C₃N₄ sheets promote effective charge separation and shorten the transport pathways of photoinduced carriers, thereby suppressing electron–hole recombination. The homogeneous elemental distribution confirmed by EDS mapping further supports this synergistic integration, which is essential for maintaining consistent charge flow across the heterointerface. Collectively, the structural features of nanoscale particle size, sheet-like morphology, and robust heterojunction interface directly contribute to enhanced charge transport and improved light absorption, ultimately accounting for the superior photocatalytic hydrogen generation observed under sunlight irradiation.

3.2. Photocatalytic Ability of TiO₂@g-C₃N₄ in H₂ Generation

The optical performance of TiO₂, g-C₃N₄, and their TiO₂/g-C₃N₄ nanoheterojunction was explored via diffuse reflectance spectroscopy (DRS). Pure TiO₂ showed strong UV absorption around 390 nm, while g-C₃N₄ captured light well into the visible region. Remarkably, the TiO₂/g-C₃N₄ heterojunction exhibited even stronger visible-light absorption. This unexpected enhancement, diverging slightly from earlier findings, is likely driven by the formation of unique (Ti)–N–C bonds at the interface. These bonds not only strengthen the interaction between components but also improve charge separation and extend light absorption, making the heterojunction a promising candidate for visible-light-driven applications. ,

The photocatalytic hydrogen production efficiency of green TiO₂, g-C₃N₄, and their green TiO₂/g-C₃N₄ nanoheterojunction was systematically investigated in an aqueous solution containing 10 vol.% glycerol under sunlight irradiation. Gas analysis conducted before illumination confirmed the absence of hydrogen and oxygen, ensuring accurate baseline measurements. Upon exposure to sunlight, pristine g-C₃N₄ and green TiO₂ exhibited hydrogen evolution rates of 3.75 and 44.5 μmol.h^–1, respectively. Notably, the H₂ production rate of green TiO₂ significantly surpasses that of conventional white TiO₂, which typically produces less than 40 μmol h^–1 under similar conditions, as reported in earlier studies. ,,, This enhancement can be attributed to the presence of oxygen vacancies, which narrow its effective bandgap, thereby extending light absorption into the visible range and improving charge carrier separation. ,

Despite these improvements, both pristine materials exhibited limited activity due to residual recombination of photogenerated carriers and the absence of a cocatalyst such as platinum. In contrast, the green TiO₂/g-C₃N₄ nanoheterojunction displayed a remarkable photocatalytic enhancement, achieving a hydrogen production rate of 110.25 μmol.h^–1 under optimal green TiO₂ loading. This represents an approximate 29.4- and 2.4-fold increase compared to pristine g-C₃N₄ and green TiO₂, respectively, as depicted in Figure a,b. The superior activity stems from the formation of a heterojunction between green TiO₂ and g-C₃N₄, promoting efficient separation and transfer of photogenerated electrons and holes. Favorable band alignment allows electrons to migrate from the conduction band of g-C₃N₄ to green TiO₂, while holes remain in the valence band of g-C₃N₄, effectively suppressing recombination. ,, Additionally, the defect-rich structure, enhanced sunlight absorption, and efficient charge migration in green TiO₂ further contribute to the boosted hydrogen evolution performance of the heterojunction system. , The photocatalytic H₂ generation performance of the as-synthesized TiO₂@g-C₃N₄ nanocomposite was compared with previously reported photocatalyst systems under various sources of irradiations in Supporting Information Table S2. The results indicate that our green-synthesized heterojunction demonstrates competitive or superior activity, attributed to its uniform morphology, strong interfacial interaction, and efficient charge separation. This comparison highlights the effectiveness of the biotemplated synthesis strategy in enhancing photocatalytic hydrogen evolution.

4.

Photocatalytic efficiency of pristine and nanoheterojunctions for H₂ production. (a) Accumulated H₂ production under natural sunlight irradiation (b) rate of hydrogen production (c), recyclability test, (d) transient photocurrent properties, and (e) electrochemical impedance spectroscopy (EIS) plots (inset is the equivalent circuit).

Furthermore, the TiO₂/g-C₃N₄ heterojunction exhibits relatively stable behavior during photocatalytic generation of hydrogen during sunlight irradiation, in addition to its good photocatalytic performance, as shown in Figure c. Nonetheless, the third cycle shows a visible drop in H₂ generation. The visible drop-in activity is explained by a decrease in the concentration of sacrificial hole scavenger glycerol with irradiation, and not due to deactivation of the catalyst. Lakshmana Reddy et al. observed a gradual decrease in the total volume of H₂ production from one cycle to another with the TiO₂-based heterojunction photocatalyst. They explained that during prolonged recyclability studies, regain in photocatalytic activity was achieved by adding a small amount of glycerol to the reaction mixture. ,

Moreover, photogenerated charge carrier separation and transfer, specific surface area, surface imperfections, and light absorption capacity are all intimately related to the photocatalytic performance of semiconductor materials. Photocurrent response and electrochemical impedance spectroscopy (EIS) Nyquist plots were employed to assess the separation and transportation of charge carriers across the heterojunction interface, given the critical role that charge carriers play in photocatalysis. Photocurrent sensitivity evaluations were performed for the produced photocatalysts in order to examine the behavior of photogenerated charge carriers. Under solar-light irradiation, the TiO₂/g-C₃N₄ heterojunction exhibits a significantly increased photocurrent density, roughly twice that of pure g-C₃N₄ and almost 10 times greater than TiO₂, displayed in Figure d. This notable improvement suggests that TiO₂ and g-C₃N₄ heterojunction creation efficiently encourages the separation of photoinduced electron–hole pairs and speeds up charge carrier transmission. The heterojunction’s high interfacial contact significantly reduces recombination, which enhances charge transfer and photocatalytic performance in general. , Although TiO₂ shows higher charge transfer resistance (calculate and provide the values 2 or 3 folds) and lower photocurrent response than g-C₃N₄, its superior photocatalytic activity can be explained by its favorable band structure, stability, and stronger oxidative potential. TiO₂ possesses a wider band gap, which generates highly reactive photogenerated holes with strong oxidizing power, enabling more efficient surface redox reactions despite slower charge transport. In contrast, g-C₃N₄, while showing higher photocurrent and lower resistance, suffers from rapid charge recombination and weaker oxidation ability of its photogenerated holes. Thus, the intrinsic surface reactivity of TiO₂ dominates over its charge transport limitations, allowing it to achieve higher overall photocatalytic activity compared to g-C₃N₄ under similar conditions. On the other hand, lower photocurrent generation at the TiO₂ surface is ascribed to the lower light intensity available from sunlight than the C₃N₄ material, which absorbs the visible spectrum of sunlight.

The interfacial transfer of charge resistance across the electrolyte–electrode interface for g-C₃N₄, TiO₂, and TiO₂/g-C₃N₄ photocatalysts was investigated using electrochemical impedance spectroscopy (EIS). The Nyquist plots, illustrated in Figure e, with the charge transfer resistance, are correlated with the arc diameter. ,, A smaller radius indicates more effective interfacial charge transmission. Out of all the samples evaluated, the TiO₂/g-C₃N₄ heterojunction showed the shortest semicircle, demonstrating its more effective charge separation and transfer capabilities at the heterojunction interface. These findings demonstrate the TiO₂/g-C₃N₄ structure’s increased potential for effective solar-driven water splitting and boosted hydrogen production capabilities.

Based on the previously reported empirical findings, we could recommend the photocatalytic H₂ production mechanism over the TiO₂@g-C₃N₄ heterojunction (Scheme ). In the TiO₂@g-C₃N₄ heterojunction, the photocatalytic H₂ generation follows an S-scheme charge transfer mechanism. Upon solar irradiation, both TiO₂ and g-C₃N₄ absorb photons, generating electron–hole pairs. In g-C₃N₄, electrons are excited to its conduction band (−1.12 eV), leaving holes in the valence band (+1.57 eV), while TiO₂ generates electrons in its conduction band (−0.26 eV) and holes in its valence band (+2.92 eV). At the interface, the less energetic electrons in the CB of TiO₂ recombine with the less oxidative holes in the VB of g-C₃N₄, a defining step of the S-scheme mechanism. , As a result, the highly reductive electrons in the CB of g-C₃N₄ and the strongly oxidative holes in the VB of TiO₂ are preserved. The electrons in g-C₃N₄ efficiently reduce protons to generate H₂, while the holes in TiO₂ oxidize sacrificial agents such as glycerol, sustaining proton supply. This selective recombination and retention of strong redox carriers suppresses electron–hole recombination, prolongs carrier lifetime, and ultimately enhances photocatalytic hydrogen evolution efficiency. , In accordance with FE-SEM and XPS studies of the produced heterojunction materials, g-C₃N₄ and TiO₂ form a strong interaction ((Ti)₂–N–C), which effectively enhances the splitting and transmission of the charge carriers formed by photons across the TiO₂@g-C₃N₄ hybrid interface. Consequently, as demonstrated in the previous discussion, g-C₃N₄ and TiO₂ can be closely coupled to confine an abundance of −OH groups across the TiO₂@g-C₃N₄ catalytic interface. This significantly improves the adsorbed H₂O’s interaction with g-C₃N₄ and speeds up with photogenerated holes in g-C₃N₄. Photogenerated holes are retained in the valence band (VB) of TiO₂, where they promote the oxidation of sacrificial agents, such as aqueous glycerol mixtures, leading to the formation of protons (H⁺) along with reactive oxidation intermediates. , Simultaneously, photogenerated electrons are predominantly accumulated in the conduction band (CB) of g-C₃N₄, where they efficiently participate in the reduction of protons (H⁺) to generate molecular hydrogen (H₂). Moreover, the TiO₂@g-C₃N₄ heterojunction’s advantageous band orientation allows for the spatial separation of photogenerated charge carriers, which successfully hinders electron and hole recombination, extending their lifetime and improving photocatalytic effectiveness. The cumulative hydrogen evolution capacity under solar irradiation is greatly enhanced by the directed movement of electrons and holes, which expedite the corresponding redox processes within the active sites of TiO₂ and g-C₃N₄. The heterojunction’s potential for long-term photocatalytic applications is highlighted by this mechanism and eqs 5–10.

10

2. Sunlight-Driven S-Scheme Photocatalytic H₂ Production Mechanism over the As-Prepared Green TiO₂@g-C₃N₄ Heterojunction .

^a [The Scheme is drawn using BioRender].

4. Conclusion

The present study explores a novel green TiO₂@g-C₃N₄ heterojunction as a highly capable, noble-metal-free photocatalyst for solar-driven hydrogen production. A comprehensive analysis of structural, morphological, and optical properties affirmed the formation of a sustainable interfacial interaction between green TiO₂ and g-C₃N₄, driven by Ti–N–C bonding. The defect-rich structure of green TiO₂, featuring abundant oxygen vacancies, plays a critical role in enhancing visible-light absorption and improving the charge separation efficiency, marking a significant advancement over conventional TiO₂ photocatalysts. Photocatalytic activity during sunlight exhibited a remarkable 110.25 μmol h^–1 hydrogen generation rate for the heterojunction, achieving 29.4- and 2.4-fold enhancements over pristine g-C₃N₄ and green TiO₂, respectively. The TiO₂@g-C₃N₄ interface’s effective S-scheme charge migration and strategic band orientation are responsible for the previously unheard-of increase in photocatalytic function. The prolonged carrier lifetimes and lower recombination losses due to the heterojunction architecture are major elements behind its improved effectiveness. These results highlight the significance of band structure stimulation, interface engineering, and defect engineering in the development of high-performance photocatalysts. This work lays a solid basis for future integration with environmentally friendly technologies by proposing an affordable, scalable approach for successful solar-driven hydrogen generation.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.5c07670.

• Photographic images of experimental set up for the sunlight driven H₂ production, change in crystallite size among the as-synthesized g-C₃N₄, green TiO₂, and TiO₂@g-C₃N₄ photocatalyst nanosamples, the potential edges of conduction and valence band of g-C₃N₄ and TiO₂, comparison study of photocatalytic H₂ generation over the as-synthesized TiO₂@g-C₃N₄ nanocomposite structures and the previously documented literature, and references (PDF)

⊥.

S.S.A and M.C.M contributed equally to this work.

The authors declare no competing financial interest.