Dual-functional ZnO/g-C₃N₄ nanocomposites: effective adsorption – photocatalysis for both direct blue 71 and methylene blue removal

Abstract

The ZnO/g-C₃N₄ nanocomposites (NCPs) were fabricated via a facile, one-step thermal decomposition of Zn(NO₃)₂·6H₂O and urea. The successful chemical integration and composition of the NCPs were confirmed through FTIR and XPS analyses, while XRD patterns and HRTEM images revealed that interfacial interactions between the two phases induce lattice distortion at the ZnO surface. The BET and UV-vis absorption analyses demonstrated that the ZnO/g-C₃N₄ NCPs exhibit enhanced visible-light absorption alongside a significant increase in pore size and volume compared to pure ZnO. These synergistic structural and optical properties led to a marked improvement in adsorption – photocatalytic efficiency for Methylene Blue (MB) and Direct Blue 71 (DB 71). Under optimal conditions (catalyst weight: 0.03 g per 50 mL; dye concentration of 1.0 × 10⁻⁵ mol L⁻¹ for MB and 25 ppm for DB 71), the ZnO/g-C₃N₄ NCPs achieved a superior removal efficiency of 98.2% for MB (at 55 °C and pH = 13) and 98.8% for DB 71 (at 25 °C and pH = 3) following 60 min of dark adsorption and 120–150 min visible-light irradiation. The NCPs achieved maximum adsorption capacities of 100.7 mg g⁻¹ for MB and 154.08 mg g⁻¹ for DB 71, performing optimally in alkaline and acidic media, respectively. Adsorption equilibrium data for both MB and DB 71 dyes correlated strongly with the Langmuir isotherm model, suggesting monolayer formation on a homogeneous adsorbent surface. The adsorption process is driven by a synergistic combination of molecular interactions – primarily physical adsorption – mediated by a network of electrostatic forces, π–π stacking, and hydrogen bonding. Furthermore, the photocatalytic degradation follows a Z-scheme mechanism, in which ˙OH radicals and h⁺ are identified as the primary active species for MB degradation, while ˙O₂⁻ radicals are the dominant species in the degradation of DB 71. These results highlight the potential of the ZnO/g-C₃N₄ nanocomposite for high-performance applications in integrated adsorption-photocatalytic wastewater treatment.

ZnO/g-C₃N₄ optimally adsorbs MB at high pH and DB 71 at low pH. Its Z-scheme photocatalysis is driven by ˙OH and h⁺ for MB degradation, whereas ˙O₂⁻ and e⁻ dominate DB 71 removal.

1. Introduction

The rapid pace of industrialization and economic growth in recent decades has brought unprecedented benefits to human society, yet it has also intensified environmental degradation to alarming levels. Among various forms of pollution, the contamination of water resources by persistent organic pollutants has emerged as a critical global challenge, posing significant threats to aquatic ecosystems and public health. Industrial effluents from sectors such as textiles, leather, cosmetics, paper, and pharmaceuticals contain large quantities of synthetic dyes. These dyes, comprising over 40 000 varieties with more than 7000 distinct chemical structures, are characterized by exceptional chemical stability, photostability, and resistance to natural degradation.¹ Such wastewater often exhibits high electrical conductivity, elevated chemical oxygen demand (COD) and biochemical oxygen demand (BOD), and intense coloration, leading to severe ecological damage.^2–5 In response, numerous physical, chemical, and biological treatment strategies have been explored; however, many are constrained by high operational costs, secondary pollution risks, or limited effectiveness against structurally diverse dye molecules.^6–8 Photocatalysis has gained attention as a sustainable and economically viable technology capable of degrading a broad spectrum of organic dyes into environmentally benign products under mild conditions.^9–12 Nevertheless, existing research has predominantly focused on single-class dyes, either cationic or anionic, without addressing the simultaneous removal of both types through combined adsorption – photocatalysis, despite the fact that real industrial effluents often contain complex mixtures of dyes with diverse ionic properties. Addressing this gap, the present study develops a ZnO/g-C₃N₄ heterojunction photocatalyst capable of rapid and high-efficiency degradation of both methylene blue (MB, cationic) and Direct Blue 71 (DB 71, anionic), providing an integrated approach toward broad-spectrum dye removal from wastewater.

Zinc oxide (ZnO) is an n-type bandgap semiconductor that possesses several important physical properties, including a direct band gap structure of approximately 3.37 eV at room temperature and a large free exciton binding energy (∼60 meV), which leads to a high probability of optical transitions. In addition, ZnO exhibits high thermal stability, a high melting point (1950 °C), non-toxicity, low cost, and environmental friendliness.^13,14 Graphitic carbon nitride (g-C₃N₄) is a metal-free photocatalyst composed solely of carbon and nitrogen, which are naturally abundant elements. Typically synthesized from nitrogen-rich precursors containing C–N frameworks, g-C₃N₄ combines environmental benignity with a moderate band gap of ∼2.7 eV, high chemical stability, and strong visible-light absorption up to ∼460 nm, making it an attractive candidate for adsorption-based and photocatalytic applications.¹⁵ Recent studies have highlighted the versatility of g-C₃N₄ in diverse applications, including photocatalysis, pollutant adsorption, electrode fabrication, antibacterial surfaces, and superhydrophilic coatings.^16–20 However, despite their individual merits, both ZnO and g-C₃N₄ suffer from intrinsic drawbacks. ZnO shows limited photocatalytic activity under visible-light irradiation and undergoes rapid recombination of photogenerated electron–hole pairs.²¹ Similarly, g-C₃N₄ exhibits a high recombination rate, low quantum efficiency, and a relatively small surface area-to-volume ratio. Coupling ZnO with g-C₃N₄ has been demonstrated to significantly enhance photocatalytic activity by leveraging the synergistic effects, including improved charge separation, broadened light absorption, and complementary adsorption capabilities. Various strategies have been explored for synthesizing g-C₃N₄, including hydrothermal methods, chemical vapor deposition (CVD), plasma techniques, and salt-assisted thermal decomposition.^22,23 Among these, salt-assisted thermal decomposition is particularly attractive for large-scale applications due to its operational simplicity, low cost, and ability to yield high-quality heterojunction structures.²⁴

The primary objective of this work is to design and fabricate a ZnO/g-C₃N₄ heterojunction photocatalyst with dual-function capabilities – simultaneous adsorption and photocatalytic degradation – toward both cationic and anionic dyes. While various ZnO/g-C₃N₄ composites have been investigated for individual dye degradation, this study specifically focuses on optimizing the synergistic interaction between these two phases to achieve the simultaneous, high-efficiency removal of co-existing cationic (MB) and anionic (DB 71) dyes. By systematically correlating physicochemical properties with adsorption – photocatalysis performance and proposing a mechanistic model for the observed synergistic effects, this work provides valuable insights for the rational design of next-generation, high-performance photocatalysts.

2. Experimental

2.1. Chemicals

Zinc nitrate hexahydrate (Zn(NO₃)₂·6H₂O, 99.9%), urea (CH₄N₂O, 99.5%), sodium hydroxide (NaOH, 96.0%), and hydrochloric acid (HCl, 38.0%) were purchased from Xilong Scientific Co., China. Methylene blue (MB, C₁₆H₁₈N₃SCl, 99.5%) and Direct Blue 71 (DB 71, C₄₀H₂₃N₇Na₄O₁₃S₄, 50%) were obtained from Xiamen Hisunny Chemical Co., China. All chemicals were used as received without further purification. Distilled water was employed in all photocatalytic activity evaluations.

2.2. Preparation of ZnO/g-C₃N₄ nanocomposite

Graphitic carbon nitride (g-C₃N₄) was synthesized via the thermal polymerization of 10 g of urea (CH₄N₂O) at 550 °C for 2 h in air. During the calcination process, urea molecules underwent condensation and polymerization reactions, yielding high-purity g-C₃N₄ with a layered structure. In the next step, a total of 10 g of urea was dissolved in an aqueous solution containing a specified amount of Zn(NO₃)₂·6H₂O, which was completely dissolved in 10 mL of distilled water. The Zn(NO₃)₂·6H₂O content was varied according to the compositions listed for the different samples. The resulting homogeneous mixture was dried at 80 °C for 2 h and subsequently calcined at 550 °C for 2 h to obtain the ZnO/g-C₃N₄ nanocomposites. A schematic representation of the fabrication procedure for the ZnO/g-C₃N₄ nanocomposite is shown in Fig. 1.

Fig. 1. Schematic illustration of the synthesis procedure for ZnO/g-C₃N₄ nanocomposites.

2.3. Characterization methods

The crystalline structure of the materials was analyzed by X-ray diffraction (XRD) using a Bruker D8 Advance diffractometer. Surface morphology was examined via scanning electron microscopy (SEM, Hitachi S-4800). The optical properties were investigated through UV-vis absorption spectroscopy (JASCO V-750). The adsorption and photocatalytic degradation of MB and DB 71 dyes were monitored based on UV-vis absorption spectra recorded with a Shimadzu UV-2450 spectrophotometer. The specific surface area and pore size distribution were determined by N₂ adsorption–desorption measurements using a NoVa 600 BET surface area analyzer.

2.4. Adsorption–photocatalytic degradation of MB and DB 71

The adsorption and photocatalytic performance of the prepared samples were evaluated through the degradation of methylene blue (MB) and Direct Blue 71 (DB 71) under dark and illuminated conditions. In a typical experiment, 0.03 g of ZnO/g-C₃N₄ was dispersed in 50 mL of MB (1.0 × 10⁻⁵ mol L⁻¹) or DB 71 solution (25 ppm). The reactor consisted of a 200 mL cylindrical glass beaker. The suspension was stirred at 200 rpm in the dark for 60 min to establish adsorption–desorption equilibrium and ensure uniform dispersion of the catalyst particles. The adsorption process was further investigated at different contact times (15–60 min, in 15 min intervals). Subsequently, the suspensions were irradiated with a 200 W incandescent lamp (Spectronics, USA) with an emission range of ≥350 nm. The experimental setup was intended to evaluate the performance under full-spectrum irradiation; therefore, no cut-off filters were employed. The distance between the light source and the liquid surface was fixed at 20 cm. The reaction vessel was stirred continuously for 150 min for MB and 120 min for DB 71. At predetermined time intervals, the catalyst powder was separated by centrifugation, and the absorbance of the supernatant was measured at the maximum absorption wavelength of each dye using a UV-vis spectrophotometer. The removal efficiency (H) of MB and DB 71 was calculated based on the change in dye concentration before and after the reaction, using the following equation:

1

where C₀ and C represent the initial dye concentration and the concentration at time t, respectively.

3. Results and discussion

3.1. XRD patterns and HR-TEM images analysis

XRD patterns of ZnO, g-C₃N₄, ZnO/x%g-C₃N₄ samples and changes in cell volume (V) and lattice strain (ε) according to the weight percentage of g-C₃N₄ were shown in Fig. 2. In Fig. 2a, XRD pattern of ZnO exhibits distinct diffraction peaks at 2θ = 31.85°, 34.32°, 36.22°, 47.51°, 56.59°, 62.84°, 66.36°, 68.01°, and 69.07° corresponding to the (100), (002), (101), (102), (110), (103), (200), (112) and (201) planes of the hexagonal wurtzite structure, respectively (JCPDS card no. 05-0664).²⁵ The XRD pattern in Fig. 2b reveals two diffraction peaks at 2θ = 12.91°, and 27.48° corresponding to (100) and (002) planes of the g-C₃N₄ materials (JCPDS card no. 066-0813).¹⁵ As observed in Fig. 2c, the XRD patterns of all ZnO/x%g-C₃N₄ NCPs display the characteristic diffraction peaks of bare ZnO without any impurity phase being detected. The absence of g-C₃N₄ peaks is due to its low concentration relative to ZnO, which may overshadow the signal, or the formation of an amorphous g-C₃N₄ layer on the ZnO surface.²⁶ However, as the g-C₃N₄ content increases, the ZnO diffraction peaks gradually weaken and shift in position, suggesting that ZnO and g-C₃N₄ interactions distort the ZnO lattice and diminish its crystallinity (see Fig. 2c). The similar findings were also reported in previous works.^27–31

Fig. 2. XRD patterns of (a) ZnO, (b) g-C₃N₄, (c and d) ZnO/x%g-C₃N₄ samples and (e) changes in cell volume (V) and lattice strain (ε) according to the weight pecent of g-C₃N₄.

To confirm the interaction between ZnO and g-C₃N₄, the structural parameters such as the lattice parameters (a and c), unit cell volume (V), distortion parameter (u), the bond length (L), and lattice strain (ε) of ZnO and ZnO/x%g-C₃N₄ samples are calculated. For (100) and (002), the values of a, b, V, u, L, and ε are calculated using the eqn (2)–(8) (ref. 32 and 33) and the results are presented in Table 1.

2

3

4

V = 0.866 × a² × c 5

6

7

8

9

where, θ and λ correspond to the diffraction angle and the X-ray wavelength (λ = 1.54 Å), (a and c) are the lattice constants, and (hkl) are the Miller indices. D is the average crystallite size of the sample. β is the FWHM of a diffraction peak, and K is the Scherrer constant (K = 0.9). The average crystallite size was derived using the Debye–Scherrer eqn (9) from the full width at half maximum (FWHM) of the (101) peak. The calculated D value of the samples is also presented in Table 1. It can be observed that the crystallite size of the ZnO/x%g-C₃N₄ samples exhibits a gradual upward trend and is larger than that of pure ZnO. With an increase in g-C₃N₄ concentration, the expanded spatial accommodation and higher density of anchoring sites enable the ZnO nuclei to easily develop and orient along these g-C₃N₄ layers. This lowers the thermodynamic energy barrier for crystallization, promoting the ordered growth of ZnO crystals and leading to a larger crystallite size than that observed in free crystallization. Table 1 shows that the c/a ratio is approximately 1.6, which is characteristic of the hexagonal ZnO structure. As the g-C₃N₄ concentration increases, the lattice parameters a and c tend to rise, leading to an expansion in the unit cell volume V. Furthermore, the increase in these lattice parameters results in higher lattice distortion, bond length, and microstrain. Specifically, the volume V of ZnO increases significantly at low g-C₃N₄ concentrations before gradually decreasing at higher loading levels. In contrast, the lattice strain exhibits a continuous upward trend with increasing g-C₃N₄ content (Fig. 2e). This suggests that ZnO/g-C₃N₄ interactions distort on the surface of the ZnO lattice.

Table 1. The structural parameters, the average crystallite size, and lattice strain of ZnO/x%g-C₃N₄ samples.

ZnO/x%g-C3N4 (x = 0–9)	Lattice constants (Å)		c/a ratio	Volume (Å3)	Distortion parameter (u)	Bond length L (Å)	Crystallite size D (nm)	Lattice strain ε (×10−3 lines−2/m−4)
	a	c
x = 0	3.247	5.210	1.60	47.573	0.379	1.977	19.05	0.0117
x = 1.0	3.272	5.201	1.59	48.244	0.381	1.986	22.58	0.1583
x = 1.5	3.282	5.210	1.59	48.610	0.382	1.991	25.79	0.1624
x = 3.0	3.265	5.213	1.59	48.129	0.380	1.984	26.78	0.1628
x = 6.0	3.262	5.210	1.60	48.014	0.380	1.983	34.39	0.1684
x = 10	3.255	5.210	1.60	47.807	0.380	1.980	31.06	0.1704

To further confirm the formation of the ZnO and g-C₃N₄ phases in the sample, a high-resolution TEM (HRTEM) image is analyzed. The TEM image in Fig. 3a reveals a polycrystalline microstructure. The corresponding SAED pattern in Fig. 3b presents distinct diffraction rings, which can be indexed to the structure of ZnO. Further examination of the dark-highlighted regions in Fig. 3c demonstrates well-defined polycrystalline domains. The Fast Fourier Transform (FFT) pattern obtained from the dark-boxed region, along with the corresponding applied mask image shown in the inset of Fig. 3c, exhibits symmetrical diffraction spots. Moreover, the inverse FFT (IFFT) image in Fig. 3d illustrates a highly ordered atomic lattice structure. The measured interplanar spacing of 0.281 nm corresponds precisely to the (100) plane of ZnO.^34,35 The XRD patterns only exhibited peaks corresponding to the ZnO phase, likely due to the low concentration of g-C₃N₄ in the nanocomposite. Consequently, the HRTEM findings are in excellent agreement with the XRD data, confirming the phase purity and distribution.^36,37

Fig. 3. (a) TEM image, (b) SAED pattern, (c) high-resolution TEM (HR-TEM) and FFT image and Apply mask are inserted. (d) The IFFT image corresponds to lattice plane fringes of ZnO.

3.2. FESEM images and EDS spectra

Fig. 4a presents the FESEM images of g-C₃N₄, ZnO samples, ZnO/x%g-C₃N₄ (x = 1.0–10.0) NCPs. As observed in Fig. 4a, the g-C₃N₄ sample exhibited a granular morphology, with particle sizes ranging from several tens to several hundreds of nanometers. Smaller particles tended to agglomerate, forming larger clusters. In contrast, the ZnO sample displayed significantly larger particle sizes, with some particles reaching dimensions of up to 1.0 µm (Fig. 4b). The FESEM images of the ZnO/x%g-C₃N₄ (x = 1.0–10.0) NCPs revealed the coexistence of two distinct particle size populations: smaller particles dispersed and anchored onto the surfaces of larger particles (Fig. 4c–f). This heterogeneous distribution is indicative of the successful integration of ZnO and g-C₃N₄ phases, where the smaller g-C₃N₄ particles serve to increase the available surface area and facilitate intimate interfacial contact with ZnO.^38,39 The intimate contact between the nanoscale g-C₃N₄ particles and the microscale ZnO grains is expected to facilitate the formation of effective heterojunction interfaces, thereby promoting charge carrier separation, enhancing light absorption, and improving both adsorption capacity and photocatalytic activity under visible-light irradiation.

Fig. 4. FESEM images of (a) g-C₃N₄, (b) ZnO, and (c–f) ZnO/x%g-C₃N₄ (x = 1.0–10.0) samples. (g and h) EDS spectra of ZnO and ZnO/10.0% g-C₃N₄ samples, respectively.

Fig. 4g and h presents the EDS spectra and the corresponding measurement locations for the ZnO and ZnO/g-C₃N₄ samples, respectively. For ZnO, the EDS spectrum revealed the presence of Zn, O, and C, with atomic percentages of 45.10 at%, 37.10 at%, and 17.80 at%, respectively (Fig. 4g). It is noted that, in EDS measurements, the detection of carbon is frequently attributed to the surface-adsorbed CO₂ and is therefore typically disregarded as an intrinsic component of the material. In contrast, the ZnO/g-C₃N₄ NCPs exhibited, in addition to Zn, O, and C, a clear signal for nitrogen, indicating the successful incorporation of g-C₃N₄ (Fig. 4h). Notably, the atomic percentage of carbon in the composite (42.56 at%) was significantly higher than that in pure ZnO (17.80 at%). This marked increase in C content can be attributed to the substantial contribution of carbon from the g-C₃N₄ framework.^40–42 The detection of nitrogen, together with the pronounced increase in carbon content, confirms the successful integration of g-C₃N₄ in the ZnO lattice.

3.3. FTIR and XPS spectra investigation

Fig. 5 presents the FTIR spectra of ZnO, g-C₃N₄ samples, and ZnO/10.0%g-C₃N₄ NCPs. ZnO shows a broad absorption band at 3200–3400 cm⁻¹, attributed to the –OH stretching vibration of adsorbed water molecules⁴³ and absorption features at 430 and 900 cm⁻¹ related to Zn–O stretching vibrations.⁴⁴ The g-C₃N₄ exhibits peaks at 1232, 1395, 1535, and 1637 cm⁻¹, attributed to the C–N and C N stretching vibrations,⁴³ while a sharp band at 805 cm⁻¹ corresponds to the s-triazine ring breathing mode. A broad absorption band at 3000–3400 cm⁻¹ originates from adsorbed water and surface N–H groups. In the ZnO/10.0%g-C₃N₄ NCPs, a new band appears at 715 cm⁻¹, attributed to a shifted Zn–O vibration resulting from ZnO and g-C₃N₄ interactions. The retention of the 3000–3400 cm⁻¹ band confirms that the g-C₃N₄ framework remains intact.^29,43 Furthermore, the characteristic g-C₃N₄ band displays a red shift and decreased intensity in the NCPs, indicating the formation of a conjugated interface rather than simple physical adsorption.

Fig. 5. FTIR spectra of ZnO, g-C₃N₄ and ZnO/10%g-C₃N₄ samples.

The chemical composition and surface electronic states of ZnO sample and ZnO/10%g-C₃N₄ NCPs were investigated using XPS spectra as presented in Fig. 6. The survey spectrum (Fig. 6a) reveals the coexistence of C, N, Zn and O elements (for ZnO), and C, N, Zn and O elements (for ZnO/10%g-C₃N₄ NCPs), confirming the high purity of the investigated samples. As shown in Fig. 6b, the high-resolution Zn 2p spectrum of the ZnO/g-C₃N₄ sample displays two peaks located at the binding energies of 1021.1 and 1044.4 eV, which are attributed to Zn 2p_3/2 and Zn 2p_1/2, respectively.⁴⁵ The 23.3 eV binding energy separation between these two peaks indicated +2 states of Zn ions in the NCPs.⁴⁰ The Zn 2p spectrum of ZnO sample shows two binding energies of 1021.3 and 1044.6 eV. The peak positions of the Zn 2p shift to the direction of high binding energy, which can be explained by photogenerated electrons moving from g-C₃N₄ to ZnO.⁴⁶ The XPS spectrum of O 1s (Fig. 6c) is deconvoluted into two peaks at 528.1 eV and 530.8 eV, which are assigned to the chemisorbed H₂O and surface hydroxyl group (–OH) of the NCPs, respectively.^40,45,47,48 The N 1s spectrum (Fig. 6d) is deconvoluted into two peaks at 395.2 eV and 397.6 eV. The low-energy peak originates from the C–N C group, which demonstrates the presence of a graphite-like sp²-bonded structure in g-C₃N₄.⁴⁷ The other one can be ascribed to the tertiarynitrogen (N–C₃).⁴⁸ The high-resolution C 1s spectrum (Fig. 6e) is deconvoluted into two distinct peaks. The peak at 284.5 eV is assigned to adventitious carbon (C–C), while the peak at 288.3 eV is attributed to the sp²-bonded carbon in the triazine rings (N–C N).⁴⁸ These results confirmed that the ZnO/g-C₃N₄ NCPs has been successfully synthesized.

Fig. 6. XPS spectra of the ZnO sample and ZnO/10.0%g-C₃N₄ NCPs: (a) survey, (b) Zn 2p. XPS spectra of the ZnO/10.0%g-C₃N₄ NCPs: (c) O 1s, (d) N 1s and (e) C 1s.

3.4. UV-vis absorption spectra and BET analysis

Fig. 7a presents the UV-vis absorption spectra of the ZnO, g-C₃N_4, and ZnO/x%g-C₃N₄ (x = 1.0–10) NCPs. All ZnO/x%g-C₃N₄ NCPs spectra exhibit a distinct shift in the absorption edge relative to pristine ZnO and g-C₃N₄. Notably, the absorption edges of the composites are positioned between those of the two individual components. This red shift is attributed to the interfacial interactions between ZnO and g-C₃N₄, which effectively narrow the optical band gap of ZnO and enhance its absorption capacity within the visible-light region.⁴⁹ The optical band gap energy values for the synthesized samples were determined using Tauc's eqn (10):

(αhν)ⁿ = A(hν − E_g) 10

where hv and E_g are the photon energy and bandgap value, α is the absorption coefficient, and A is the parameter. The exponent n characterizes the nature of the electronic transition, with n = 2 for direct transitions and n = 1/2 for indirect transitions. For these materials, the relationship between [F(R)·hν]² with hν was plotted according to the Kubelka–Munk function, as shown in Fig. 7b.³⁸ The estimated band gap energies were 3.11 eV for ZnO, 2.78 eV for g-C₃N₄, while ZnO/x%g-C₃N₄ NCP ranged from 3.03 eV to 2.78 eV. Therefore, the presence of g-C₃N₄ with ZnO results in intermediate values, confirming strong electronic coupling between ZnO and g-C₃N₄. Such coupling facilitates broader spectral absorption and enhances photocatalytic efficiency under visible-light irradiation.

Fig. 7. (a) UV-vis absorption spectra and (b) plot of (αhν)^1/2versus photon energy for ZnO, g-C₃N₄, and ZnO/x%g-C₃N₄ samples. (c) N₂ adsorption–desorption isotherms of hysteresis loop for ZnO, g-C₃N₄ and ZnO/10%g-C₃N₄ samples.

The Brunauer–Emmett–Teller (BET) specific surface area of ZnO, g-C₃N₄ and the ZnO/10%g-C₃N₄ NCPs were determined from N₂ adsorption–desorption isotherms. As shown in Fig. 7c, the N₂ adsorption–desorption isotherms for ZnO, g-C₃N₄ and the ZnO/10%g-C₃N₄ NCPs exhibit Type IV characteristics with H₃ hysteresis loop.^38,50 According to IUPAC classification, this confirms the mesoporous nature of the samples.^38,51,52 The BET surface areas for ZnO, g-C₃N₄ and the ZnO/10%g-C₃N₄ NCPs are found to be 3.67, 52.36, and 64.28 m² g⁻¹, respectively. The significant increase in the specific surface area of the ZnO NPs is attributed to the incorporation of g-C₃N₄.²⁶ Additionally, pore size and pore volume of ZnO, g-C₃N₄, and the ZnO/10.0%g-C₃N₄ NCPs are analysed and summarized in Table 2. The results demonstrate a significant increase in both pore size and pore volume for the ZnO/10.0%g-C₃N₄ NCPs relative to pure ZnO. This expansion of surface area and porosity is expected to provide a higher density of active sites for adsorption and photocatalytic reactions, thereby bolstering the synergistic effect between these two processes.^53,54

Table 2. BET surface areas, pore size, and pore volume of ZnO, g-C₃N_4, and the ZnO/10.0%g-C₃N₄ NCPs.

Samples	BET surface areas (m2 g−1)	Pore size		Pore volume
		BJH adsorption average pore (Å)	BJH desorption average pore (Å)	BJH adsorption cumulative volume of pores (cm3 g−1)	BJH desorption cumulative volume of pores (cm3 g−1)
ZnO	3.6707	129.506	174.768	0.013082	0.015662
g-C3N4	52.3666	1457.270	1329.016	2.178743	2.178574
ZnO/10.0%g-C3N4	64.2857	340.968	561.105	0.588292	1.059811

3.5. Adsorption–photocatalytic degradation of MB and DB 71

Fig. 8a–c shows UV-vis absorption spectra of ZnO, g-C₃N_4, and the ZnO/10.0%g-C₃N₄ NCPs after 60 minutes in the dark and 150 minutes under visible-light irradiation. The ZnO/10.0%g-C₃N₄ NCPs exhibit a significantly higher degradation rate for MB compared to pure ZnO and g-C₃N₄. This accelerated removal efficiency is reflected in the UV-vis absorption spectra shown in Fig. 8d, where the characteristic peak intensity at 664 nm decreases more rapidly over time for the ZnO/10.0%g-C₃N₄ NCPs. Under identical experimental conditions (catalyst weight of 0.03 g per 50 mL MB solution at 1.0 × 10⁻⁵ mol L⁻¹, 25 °C, pH = 13, 60 min dark adsorption), the adsorption efficiency of the ZnO/10.0%g-C₃N₄ sample (92.9%) is several times greater than that of pure ZnO (4.8%) and g-C₃N₄ (23.3%). This superior performance is attributed to the enhanced porosity and larger specific surface area of the ZnO/10.0%g-C₃N₄ NCPs, which facilitate the capture of a higher density of MB molecules on the catalyst surface, as shown in the BET analysis.³⁶ Upon reaching adsorption–desorption equilibrium and subsequent exposure to visible light (incandescent lamps, after 150 minutes of illumination), the MB degradation efficiency of the ZnO and g-C₃N₄ samples further increased to 47.7% and 54.6%, respectively. Notably, the MB degradation efficiency ZnO/10.0%g-C₃N₄ sample substantially increased to 95.3%. These findings clearly indicate that the ZnO/g-C₃N₄ NCPs exhibited markedly superior performance compared to the individual components (ZnO and g-C₃N₄), owing to the synergistic interactions between the two constituent phases.⁴⁰ Thus, this combination not only optimized the adsorption capacity but also enhanced the photocatalytic activity, thereby improving the overall treatment efficiency for practical applications. Fig. 8e provides visual evidence of MB decolorization across various photocatalysts after a 340 minute reaction period.

Fig. 8. (a–c) UV-vis absorption spectra of ZnO, g-C₃N₄ and the ZnO/10%g-C₃N₄ NCPs after 60 minutes in dark and 240 minutes under visible light. (d) The relationship between the MB removal efficiency of ZnO, g-C₃N₄ and the ZnO/10%g-C₃N₄ NCPs samples and time. (e) Visual evidence of MB decolorization in the presence of various photocatalysts after a 340 minute reaction period.

A similar trend was observed for DB 71, shown in Fig. 9a–c. Under identical experimental conditions (catalyst weight of 0.03 g per 50 mL DB 71 solution at 25 ppm, 25 °C, pH = 3, 60 min dark adsorption and 120 minutes of illumination), the ZnO/g-C₃N₄ NCPs achieved a superior removal efficiency of 98.8%. This performance significantly outperforms standalone ZnO (7.9%) and g-C₃N₄ (72.2%) (Fig. 9d). Fig. 9e presents the visual decolorization of DB 71 after a 180 minute reaction period using various photocatalysts. These findings suggest that the synergetic integration of ZnO and g-C₃N₄ provides a broad-spectrum, highly efficient platform for degrading both cationic and anionic dyes, highlighting its potential for industrial wastewater remediation.

Fig. 9. (a–c) UV-vis absorption spectra of ZnO, g-C₃N₄ and the ZnO/6.0%g-C₃N₄ NCPs after 60 minutes in the dark and 120 minutes under visible light. (d) The relationship between the DB 71 removal efficiency of ZnO, g-C₃N₄ and the ZnO/6.0%g-C₃N₄ NCPs samples and time. (e) Visual evidence of DB 71 decolorization in the presence of various photocatalysts after a 180 minute reaction period.

3.6. Effect of the g-C₃N₄ concentration, catalyst weight, pH, and temperature on the removal efficiency of MB and DB 71

Under experimental conditions including catalyst weight of 0.03 g per 50 mL MB solution at 1.0 × 10⁻⁵ mol L⁻¹, pH = 13, 60 min dark adsorption and 150 minutes of illumination, the effect of the g-C₃N₄ concentration on the removal efficiency of MB is presented in Fig. 10a. In the ZnO/g-C₃N₄ NCPs system, increasing the g-C₃N₄ content from 1.0% to 10.0% results in a significant enhancement in MB removal efficiency, surging from 21.1% to an optimal 95.3%. This improvement is primarily attributed to the dominant role of both physical and chemical adsorption mechanisms on the material's surface.⁴⁰ At a low g-C₃N₄ loading (1.0%), the ZnO surface remains insufficiently coated, leading to a restricted contact area and poor dye retention. Conversely, as the g-C₃N₄ concentration increases, its characteristic layered architecture – boasting a high specific surface area – functions as a structural ‘scaffold’. This creates a dense capillary network that facilitates the deep infiltration and capture of MB molecules. Further evaluation of catalyst dosages ranging from 0.01 g to 0.05 g in Fig. 10b revealed that the optimal mass of ZnO/g-C₃N₄ for MB removal was 0.03 g per 50 mL solution. The observed increase in removal efficiency from 0.01 g to 0.03 g is due to the availability of more active sites on the ZnO/g-C₃N₄ surface.³⁶ However, a further increase in catalyst mass led to a decline in efficiency, which can be attributed to a decrease in dye adsorption per unit mass of catalyst, possibly due to particle aggregation or light scattering effects at higher loadings.³³ The effects of pH and temperature on MB removal efficiency were evaluated by adjusting the pH of MB solutions from 3 to 13 and varying the temperature from 25 °C to 55 °C. As shown in Fig. 10c, the variation in adsorption capacity with pH can be interpreted in terms of the surface charge properties of ZnO/g-C₃N₄. As shown in Fig. 10c, the MB removal efficiency increased progressively with pH. At pH = 13, the removal efficiency reached 94.6% after only 15 min of adsorption and approximately 95.3% after 210 min of treatment. It is explained that at higher pH, the surface of ZnO/g-C₃N₄ becomes negatively charged, enabling strong electrostatic attraction between MB cations and the functional groups on the composite surface, thereby enhancing adsorption capacity.^29,55 As the temperature increases from 25 to 55 °C, the MB removal efficiency rises from 95.3% to 98.2% (Fig. 10d). This enhancement is attributed to increased molecular collision efficiency, promoted redox reactions, and improved degradation performance.⁵⁶ Similarly, recent studies reported that the degradation efficiency of 20 ppm bromophenol blue using a Fe₃O₄/r-GO composite improved from 91.87% to 98.27% as the temperature rose from 20 °C to 35 °C, owing to an accelerated reaction rate.^57,58 A higher reaction rate facilitates the generation of ˙OH radicals, thereby accelerating dye degradation. However, beyond the optimum temperature, the efficiency declines due to exacerbated desorption at elevated temperatures.

Fig. 10. Removal efficiency of ZnO/x%g-C₃N₄ NCPs for MB as a function of (a) g-C₃N₄ concentration, (b) catalyst weight, (c) pH and (d) temperature.

The impact of g-C₃N₄ loading on the removal efficiency of DB 71 was evaluated by varying the concentration from 1.0% to 10.0%. Experimental conditions were maintained at a catalyst dosage of 0.03 g per 50 mL of 25 ppm DB 71 solution at pH = 3, with 60 min of dark adsorption followed by 120 min of visible-light irradiation. As shown in Fig. 11a, the removal efficiency increased as the g-C₃N₄ concentration rose from 1.0% to 6.0%, reaching a maximum of 98.8%. The relationship between DB 71 degradation efficiency and the mass of the ZnO/6.0%g-C₃N₄ NCPs is illustrated in Fig. 11b. The results reveal that removal efficiency increased as the catalyst dosage rose from 0.01 to 0.03 g, reaching a maximum of 98.8% at 0.03 g. However, efficiency declined slightly at higher dosages up to 0.5 g. The initial improvement is attributed to the increased availability of active sites.^36,59 Conversely, the reduction in efficiency beyond 0.03 g stems from particle aggregation, which limits the effective surface area, and increased solution turbidity, which enhances light scattering and reduces photon penetration. However, a slight decline in performance was observed at higher concentrations up to 0.5 g. As shown in Fig. 11c, the highest removal efficiency of DB 71 was reached at pH = 3 and then gradually decreased with increasing pH solutions from 3 to 13. At pH = 5, the amino groups (–NH₂) of g-C₃N₄ become protonated to –NH₃⁺, rendering the surface positively charged and thereby favoring the adsorption of anionic dyes such as DB 71, which contains negatively charged sulfonate groups (–SO₃⁻), resulting in the maximum removal efficiency of 98.8%. In contrast, at higher pH, the surface of ZnO/g-C₃N₄ NCPs becomes negatively charged, resulting in decreased electrostatic attraction between the DB 71 anion and the functional groups on the composite surface, thereby reducing the removal efficiency of DB 71.^29,55 In contrast to the trend observed for MB, increasing the temperature from 25 to 55 °C leads to a decrease in DB 71 removal efficiency from 98.8% to 89.0% (Fig. 11d). Although elevated temperatures enhance the degradation kinetics, they simultaneously inhibit the exothermic adsorption of dye molecules. In this instance, the rate of desorption outweighs the degradation rate, resulting in the observed decline in overall efficiency at higher temperatures. Table 3 presents the removal efficiency of MB and DB 71 using various photocatalysts, indicating the effective removal efficiency of the synthesized ZnO/g-C₃N₄ NCPs.

Fig. 11. Removal efficiency of ZnO/x%g-C₃N₄ NCPs for DB 71 as a function of (a) g-C₃N₄ concentration, (b) catalyst weight, (c) pH, and (d) temperature.

Table 3. Recent studies on the removal efficiency of MB and DB 71 using various photocatalysts.

Materials	Dyes	Reaction conditions (°C, dosage, pH, T)	Ads. time (min)	Ads. eff. (%); qmax (mg g−1)	Photo. time (min)	Removal. eff. (%)	Light source	Ref.
Cr-doped ZnO	MB	20 ppm, 1.0 g L−1, pH 12, 25 °C	180	40	150	98%	UV lamp (100 W, spectronics, USA)	60
ZnO-zeolite	MB	100 ppm, 0.1 g L−1, pH 8, 25 °C	180	88.52	210	96.81% (UV)	UV (λ = 360 nm) and solar light	61
				41.32 mg g−1		96.7% (sunlight)
REC/Fe3O4/ZnO	MB	5–30 mg L−1, 0.9 g L−1, 25 °C pH 5 (dark opt.)	60	20–45	180	∼100%	Incandescent lamp (≥350 nm, 1900 µW cm−2)	62
		pH 6 (photo opt.)		35.1 mg g−1
TiO2/g-C3N4	MB	20 mg L−1, 200 mg L−1, pH 12, 25 °C	30	—	160	∼92%	Xe lamp (200 W, with 400–600 nm filter)	63
g-C3N4/MoO3	MB	10 mg L−1, 1.0 g L−1	30	46.5	180	93%	Xe lamp (300 W, λ ≥ 400 nm)	64
TiO2-PAA-GO	DB71	25 ppm, 1.0 g L−1, pH 6	60	32.42	90	98.26%	Compact lamp (36 W)	65
ZnO:Ce3+	DB71	50 ppm, 1.0 g L−1, pH 5	75	—	150	98.4%	Incandescent lamp	32
Co3O4/TiO2:(N, S)	DB71	25 ppm, 5.0 g L−1, pH 4	60	—	30	36 W compact bulb (180 min): 99%	Compact lamp (36 W, rang dong) UV-vis lamp (450 W)	66
						36 w compact: 46%; 450 W compact: 79%; sunlight: 97% (30 min)	Natural sunlight
ZnO/CoFe2O4	DB71	0.4 mM, 1.6 g L−1, pH 6	45	—	300	∼100%	Tungsten halogen (150 W, λ ≥ 400 nm, 80 600 lx, extech, USA)	67
CuO-ZnO	DB71	20 mg L−1, 1.5 g L−1, pH 7	30	—	180	97%	Natural sunlight	68
ZnO (H2O2)	DB71	100 ppm, 1.0 g L−1, pH 2.5	—	—	20	∼100%	Combined photo-ultrasonic (US/UV/ZnO)	69
ZnO/g-C3N4	MB	10−5 mol L−1 (∼3.2 mg L−1), 0.6 g L−1, pH 13	60	150	92.9%	98.2%	Incandescent lamp (200 W, ≥350, spectronics, USA)	This study
	DB71	25 ppm, 0.6 g L−1, pH 3	60	120	95.0%	98.8%	Incandescent lamp (200 W, ≥350 spectronics, USA)

3.7. Adsorption isotherm analysis

The adsorption capacity (q_e) is estimated following eqn (11).⁷⁰

11

where V (L) and m (g) are the volume of the solution and mass of the adsorbent. C and C_e represent the initial and equilibrium dye concentrations, respectively. The adsorption isotherms of MB and DB 71 were evaluated by fitting the experimental data to Langmuir and the Freundlich models. For MB, adsorption studies were performed at concentrations ranging from 0.5 × 10⁻⁵ to 13.3 × 10⁻⁵ mol L⁻¹ (0.03 g ZnO/10.0%g-C₃N₄ in 50 mL solution, 25 °C, pH = 13, 60 min contact time). For DB 71, the concentration range was 10–400 ppm under similar conditions (0.03 g ZnO/6.0%g-C₃N₄, 25 °C, pH = 3, 60 min contact time).

The Langmuir isotherm is expressed as follows (12):⁷⁰

12

where q_e and q_m represent the equilibrium adsorption capacity and the maximum adsorption capacity (mg g⁻¹) of MB or DB 71 onto the adsorbent, respectively. k_L is the Langmuir constant (L mg⁻¹).

The Freundlich isotherm is represented by the following eqn (13):⁷⁰

13

where k_f is the Freundlich binding constant (L mg⁻¹), and 1/n is a constant related to the surface heterogeneity. As shown in Fig. 12, the Langmuir model provided a better fit to the experimental data than the Freundlich model for both MB and DB 71, consistent with previous reports in the literature.^26,71,72 MB and DB 71 adsorption onto ZnO/10.0%g-C₃N₄ follows the Langmuir isotherm (see Fig. 12a and c), suggesting that the adsorption process primarily occurs via monolayer coverage of the active sites on the material surface.²⁶ The maximum adsorption capacities (q_m) were calculated to be 100.7 mg g⁻¹ for MB, and 154.08 mg g⁻¹ for DB 71. The ZnO/g-C₃N₄ nanocomposite exhibited high adsorption capacities and short equilibrium times compared to several previously reported adsorbents, such as MgTiO₃/g-C₃N₄ (q_m = 232 mg g⁻¹ for Rhodamine B dye, 120 min of equilibrium time),⁵⁰ ZnO/g-C₃N₄ (q_m = 478.14 mg g⁻¹ for malachite green, 120 min of equilibrium time).²⁶ Although the adsorption capacity is not the highest reported, the significantly shorter equilibrium time highlights the potential of this material for efficient removal of organic dyes in practical wastewater treatment applications.

Fig. 12. Langmuir and Freundlich adsorption isotherm models for (a and b) MB and (c and d) DB 71.

3.8. Absorption and degradation mechanism of MB and DB 71

The ZnO/g-C₃N₄ composite exhibited superior dye removal efficiency compared to its individual components, attributed to the synergistic integration of surface adsorption and photocatalytic degradation mechanisms. The adsorption process plays a crucial role in enriching the pollutant concentration at the catalyst surface. The adsorption of MB and DB 71 dyes onto the ZnO/g-C₃N₄ NCPs surface is driven by a synergistic combination of molecular interactions. This mechanism primarily involves physical adsorption, where dye molecules are sequestered via electrostatic interaction forces, π–π stacking between aromatic rings, and hydrogen bonding.²⁶

Fig. 13a presents the absorption mechanism of MB for ZnO/g-C₃N₄ NCPs. The g-C₃N₄ component, with its nitrogen-rich layered structure, abundant lone-pair electrons, negatively charged surface, and extended π-conjugated system, enables strong interactions with dye molecules containing aromatic rings and azo groups through electrostatic attraction, hydrogen bonding, and π–π stacking interactions.^73,74 Furthermore, the incorporation of ZnO increases the specific surface area and facilitates a more uniform distribution of active sites (–OH).⁷⁵ In aqueous media, the ZnO/g-C₃N₄ surface undergoes hydroxylation, yielding active –OH groups. When the solution pH exceeds (high pH) the point of zero charge (pH > pHpzc), deprotonation occurs, imparting a net negative charge to the composite surface. This promotes strong electrostatic attraction toward cationic MB molecules. Furthermore, π–π stacking interactions between the conjugated heptazine rings of g-C₃N₄ and the aromatic phenothiazine framework of MB establish robust molecular adhesion. The absorption process is also complemented by hydrogen bonding, specifically between the methyl (–CH₃) protons of MB and the nitrogen sites of g–C₃N₄, as well as between the surface hydroxyls of ZnO and the heteroatoms within the MB aromatic system.

Fig. 13. (a and b) Schematic of the adsorption mechanism, (c and d) scavenger experiments for photocatalytic degradation, and (e) proposed degradation pathway of MB and DB 71 over ZnO/g-C₃N₄ nanocomposites.

The absorption mechanism of DB 71 for ZnO/g-C₃N₄ NCPs is shown in Fig. 13b. In acidic media (low pH), the ZnO/g-C₃N₄ surface undergoes protonation, acquiring a net positive charge. Conversely, the DB 71 molecule possesses multiple sulfonate groups (–SO₃), which impart a robust negative charge in aqueous solution. This disparity facilitates a strong electrostatic attraction, effectively sequestering the anionic dye molecules onto the catalyst's surface. Furthermore, the extensive aromatic system of DB 71 enables significant π–π stacking with the g-C₃N₄ layers; due to its larger molecular footprint compared to MB, the resulting contact area is substantially increased, enhancing structural stability. This adsorption is further stabilized by hydrogen bonding between the material's surface functional groups and the azo (–N N) or sulfonate (–SO₃) moieties of the DB 71 molecule.

To elucidate the photocatalytic degradation mechanisms of MB and DB 71 using the using ZnO/g-C₃N₄ composite, benzoquinone (BQ), isopropanol (IPA), and ethylenediaminetetraacetic acid (EDTA) were employed as scavengers for superoxide radicals (˙O₂⁻), hydroxyl radicals (˙OH), and photogenerated holes (h⁺), respectively.^76,77 For MB degradation, experimental conditions include catalyst weight of 0.03 g per 50 mL MB solution at 1.0 × 10⁻⁵ mol L⁻¹, pH = 7, 60 min of dark adsorption, and 150 minutes of illumination. The removal efficiency decreased significantly upon the addition of IPA and EDTA, indicating that the hydroxyl radicals (˙OH) and/and h⁺ play dominant roles in the process.^76,77 Conversely, the degradation performance remained unchanged upon the addition of BQ, confirming that·˙O²⁻ radicals have a negligible effect on MB degradation⁷⁸ (Fig. 13c). In contrast, for DB 71 degradation, experimental conditions include catalyst weight of 0.03 g per 50 mL DB 71 solution at 1.0 × 10⁻⁵ mol L⁻¹, 25 °C, pH = 7, 60 min of dark adsorption, and 120 minutes of illumination. The removal efficiency decreased substantially with the addition of BQ,^79,80 whereas it remained relatively unsignificantly affected with the addition of IPA and EDTA (Fig. 13d). This indicates that (˙O₂⁻) radicals are the primary active species involved in the photocatalytic degradation of DB 71.^79,80

The heterojunction formed between ZnO and g-C₃N₄ optimizes the separation of photogenerated electron–hole pairs under visible-light irradiation compared to the pristine counterparts.⁸¹ The positions of the valence band (VB, E_VB) and conduction band (CB, E_CB) were calculated using the empirical equations:⁸²

E_VB = χ − E_e − 0.5E_g 14

E_CB = E_VB + E_g 15

where χ is the absolute electronegativity of the constituent atoms and E_e is the energy of free electrons on the hydrogen scale (∼4.5 eV).⁸³ For ZnO and g-C₃N₄, the value of χ is 5.78 eV and 4.67 eV, respectively. The E_VB and E_CB values are calculated at about −0.275 eV, 2.835 eV (for ZnO), and −1.22 eV, 1.56 eV (for g-C₃N₄), respectively. The redox potential for the reduction of adsorbed O₂ by electrons (e⁻) on a semiconductor surface to generate ˙O₂⁻ is −0.33 eV. The conduction band (CB) energy level of ZnO is −0.275 eV, which is slightly lower than the potential required for ˙O₂⁻ generation. In contrast, the CB level of g-C₃N₄ is −1.22 eV, which is sufficiently higher than the reduction potential, facilitating the formation of ˙O₂^−.84 Consequently, the photocatalytic generation of ˙O₂⁻ is thermodynamically hindered when electrons accumulate on the CB of ZnO, whereas it is significantly more efficient when electrons are concentrated on the CB of g-C₃N₄. Regarding the generation of ˙OH radicals, the redox potential for the oxidation of surface-adsorbed OH⁻ ions by holes (h⁺) is 2.28 eV.⁷⁶ The valence band (VB) level of ZnO is 2.835 eV, which is more positive than this potential, thereby favoring _˙OH production. Conversely, the VB level of g-C₃N₄ is 1.56 eV, which is insufficient to drive this reaction. These results indicate that during the photocatalytic process, the accumulation of h⁺ in the VB of ZnO promotes the efficient production of _˙OH, while the formation of _˙OH is negligible when h⁺ is confined to the VB of g-C₃N₄. Therefore, the degradation of MB and DB 71 by ZnO/g-C₃N₄ photocatalyst was identified as Z-scheme semiconductor photocatalyst, as presented in Fig. 13e.^31,76 Upon absorption of sufficient light energy, electrons in the valence bands (VB) of ZnO and g-C₃N₄ are excited to their respective conduction bands (CB), generating electron–hole (e⁻–h⁺) pairs, as described by eqn (16) and (17).^31,84 Because the ZnO/g-C₃N₄ composite forms a Z-scheme photocatalytic system, a significant portion of the electrons generated in the ZnO conduction band migrates to the g-C₃N₄ valence band. This charge transfer process effectively separates the e⁻/h⁺ pairs, thereby significantly enhancing the photocatalytic efficiency of the material. Subsequently, these electrons react with surface-adsorbed O₂ to generate ˙O₂⁻, while the holes react with H₂O or OH⁻ to produce ˙OH radicals, as detailed in eqn (18)–(20).⁸⁴ During the degradation of MB, dye molecules primarily react with the generated ˙OH radicals and/or directly with h⁺ to form intermediate species (eqn (21)).⁷⁶ In contrast, DB 71 dye molecules predominantly interact with the generated ˙O₂⁻ radicals and/or electrons (e⁻) as described by eqn (22).⁸⁴ These intermediates subsequently react with strong oxidizing species, undergoing further oxidation and decomposition into smaller molecules, or complete mineralization into CO₂ and H₂O.

ZnO + hν → e⁻ (ZnO) + h + (ZnO) 16

g-C₃N₄ + hv → e⁻ (g-C₃N₄) + h⁺ (g-C₃N₄) 17

e⁻ + O₂ → ˙O₂⁻ 18

h⁺ + H₂O → H⁺ + ˙OH 19

h⁺ + OH⁻ → ˙OH 20

˙OH + MB → H₂O + CO₂ + Degraded products 21

˙O₂⁻ + DB 71 → H₂O + CO₂ + Degraded products 22

To evaluate the structural stability after the treatment process, XRD patterns, FESEM images, and EDS spectra of the ZnO/10%g-C₃N₄ sample were measured before and after the removal of MB (at pH = 13) and DB 71 (at pH = 3). As shown in Fig. 14a, the XRD patterns indicate no change in the crystal structure of ZnO before and after the removal processes of MB and DB 71, corresponding to an unaltered structure. However, the diffraction peak intensities of the treated samples decreased compared to the initial sample, indicating a reduction in the crystalline quality of ZnO. Furthermore, the surface morphology of the post-treatment samples exhibits a slight alteration compared to the pristine sample (Fig. 14b–d), further confirming a decline in the material's stability. The post-degradation EDS spectra revealed the presence of S and Cl (Fig. 14e), and Na and S (Fig. 14f), indicating excess adsorbed MB and DB 71 molecules, respectively.^32,33 These results confirm that after the removal processes of MB (at pH = 13) and DB 71 (at pH = 3), the synthesized ZnO/g-C₃N₄ composite revealed acceptable structural stability after the reactions.

Fig. 14. (a) XRD patterns and (b–d) FESEM images of ZnO/10%g-C₃N₄ sample before and after removal processes of MB (at pH = 13) and DB71 (at pH = 3). (e and f) Corresponding EDS spectra after removal.

4. Conclusions

The ZnO/g-C₃N₄ nanocomposites were synthesized via a one-step thermal. Characterization via FTIR, XPS, XRD, and HRTEM confirmed successful integration, revealing interfacial interactions that induce lattice distortion at the ZnO surface. Compared to pure ZnO, the NCPs exhibited enhanced visible-light absorption and increased porosity (BET), significantly boosting adsorption-photocatalytic performance. Under optimized conditions, the ZnO/g-C₃N₄ nanocomposites achieved removal efficiency of 98.2% for MB (pH = 13, 55 °C) and 98.8% for DB 71 (pH = 3, 25 °C), with maximum adsorption capacities of 100.7 and 154.08 mg g⁻¹, respectively. The adsorption kinetics were well-described by the Langmuir isotherm model, driven by a synergistic interplay of electrostatic forces, π–π stacking, and hydrogen bonding. Furthermore, the photocatalytic degradation confirmed a Z-scheme mechanism, where ˙OH radicals and h⁺ were identified as the primary active species for MB degradation, whereas ˙O₂⁻ radicals dominated the degradation of DB 71. These findings showed that the synthesized ZnO/g-C₃N₄ nanocomposites could be used as a high-performance adsorbent – photocatalyst material for both MB and DB 71 removal applications in water pollutants.

Author contributions

Ngo Trung Hoc: investigation; writing – original draft; methodology; formal analysis. Le Dinh Tuan, Nguyen Van Quang, Quang Vinh Tran, Cao Thi Huyen Trang: investigation; formal analysis; methodology. Nguyen Thi Hanh, Nguyen Tu, Pham Thi Lan Huong: investigation; methodology; formal analysis; writing – review and editing.

Conflicts of interest

The authors declare no conflicts of interest.

Data availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.