A Visible-Light-Active TiO₂/Bi₂O₃/g-C₃N₄ Heterojunction: Synthesis, Photocatalytic Degradation of Rhodamine B, and Antibacterial Activity

Abstract

Ternary heterojunction photocatalysts enhance the separation and transport of photogenerated charge carriers, thereby boosting their redox activity for use in environmental and sustainable energy applications. This study focuses on the synthesis of a TiO₂/Bi₂O₃/g-C₃N₄ heterojunction composite via a ceramic method with TiO₂ loadings of 80%, 85%, and 90% (denoted 80T-BC, 85T-BC, and 90T-BC, respectively) to investigate structure–property–performance relationships in photocatalytic dye degradation. The structural, optical, and morphological properties of the synthesised materials were characterised using X-ray diffraction (XRD), scanning electron microscopy (SEM), and diffuse reflectance UV–Vis spectroscopy (DRS). The photocatalytic performance was evaluated by measuring the degradation of Rhodamine B under visible light irradiation. Under optimised conditions (pH 6, initial RhB concentration of 5 mg/L, and a reaction time of 120 min), a degradation rate of 99% was achieved. Furthermore, the semiconductor demonstrated significant antibacterial activity against both Gram-negative (Pseudomonas aeruginosa) and Gram-positive (Staphylococcus aureus) bacteria. This study presents a promising strategy for modifying TiO₂-based semiconductors by incorporating different metal oxides. The formation of the resulting heterojunction significantly enhances photocatalytic efficiency, demonstrating strong potential for practical environmental remediation.

1. Introduction

The complexity of environmental pollution in hospital settings results from the presence of effluents containing recalcitrant pollutants such as pharmaceuticals, dyes, and microorganisms [1,2]. In addition, contaminated work surfaces and personal protective equipment, such as medical gowns, can serve as favorable environments for nosocomial pathogens, contributing to healthcare-associated infections (HAIs) [3,4]. Moreover, biological aerosols released by bacteria in indoor spaces, and the inhalation of airborne pathogenic bacteria that release biological toxins or endotoxins [5,6,7,8], as well as exposure to airborne microorganisms often dispersed by ventilation, heating, and air conditioning (VHAC) systems, pose extremely dangerous health risks, causing infections and inflammatory or allergic diseases [9,10]. Their importance as major reservoirs and vectors of nosocomial infections is highlighted by the high microbial load detected in the air, on surfaces, and on medical devices, including pathogenic microorganisms such as Staphylococcus aureus, Escherichia coli, and Pseudomonas aeruginosa [11,12,13,14,15]. Eliminating these chemical and biological pollutants becomes challenging using conventional treatment methods, highlighting the need for advanced and multifunctional materials capable of implementing on-demand, environmentally friendly remediation strategies.

Heterogeneous photocatalysis, induced by semiconductor materials, has become the cornerstone of advanced oxidation processes (AOPs), used to remove recalcitrant pollutants, due to its ability to generate reactive oxygen species (ROS), such as hydroxyl radicals (•OH) and superoxide anions (•O₂⁻), under light radiation. These species can mineralise organic pollutants and inactivate bacteria through oxidative damage that affects DNA and protein synthesis, leading to membrane rupture and cell death [12,16,17,18]. Degradation and antibacterial activity are closely linked; however, they are generally studied using different approaches, and very little effort has been made to better understand this relationship by synthesizing a single material with a dual effect.

Among the semiconductor photocatalysts, Titanium dioxide (TiO₂) is the most widely studied thanks to its high activity, chemical stability, and non-toxicity. However, its wide band gap (~3.2 eV for anatase) restricts photoactivation to visible light, accounting for only around 4% of the solar spectrum. Additionally, the rapid recombination of photogenerated electron–hole pairs limits its overall quantum yield [19].

In order to exploit visible light, which constitutes the majority of both natural and artificial light, extensive research has been conducted into developing new photocatalysts, such as Bismuth oxide and graphitic carbon nitride. Bismuth oxide (Bi₂O₃) is a promising material for use under visible light due to its narrower band gap (~2.1–2.8 eV), and has shown potential for organic degradation and antibacterial applications [20]. However, pure Bi₂O₃ has a major drawback: poor charge carrier separation. Graphitic carbon nitride (g-C₃N₄), a metal-free polymer highly sensitive to visible light (with a band gap of ~2.7 eV), has attracted considerable interest due to its ease of synthesis, thermal stability, and band positions that are suitable for redox reactions [21]. In recent years, graphitic C₃N₄ (g-C₃N₄) has become an excellent photocatalyst for the degradation of organic matter due to thermochemical stability [22]. Nevertheless, the rapid recombination of photogenerated carriers in pure g-C₃N₄ hinders its practical application.

Previous studies have explored various strategies for reducing the bandgap energy of titanium dioxide (TiO₂), including n-type and p-type doping [23]. Another widely studied approach is to form heterojunctions, i.e., interfaces between two or more n-type or p-type semiconductors [24,25]. The recent literature has also highlighted the incorporation of semiconductor materials into TiO₂ [26]. These composite materials exhibit enhanced optical properties compared to pure TiO₂, due to vacant d orbitals that facilitate electron mobility [27,28]. The improvement in the photocatalytic activity of these systems is primarily attributed to alterations in the electronic configuration of the dopant metal, arising from charge transfer between the d-electrons of the transition metals and the valence band (VB) of TiO₂ [29,30]. A powerful strategy for overcoming the inherent limitations of single semiconductors is to construct heterojunction composites. Coupling materials with staggered band alignments creates an interfacial electric field that forces the spatial separation of photogenerated electrons and holes. This dramatically prolongs the lifetime of charge carriers and enhances photocatalytic performance [31,32]. Binary TiO₂/g-C₃N₄ and Bi₂O₃/g-C₃N₄ heterojunction systems have been widely studied for their ability to degrade pollutants [33,34]. TiO₂/g-C₃N₄ systems demonstrate enhanced visible-light activity, though they exhibit reduced redox potential in type-II configurations and complex fabrication in Z-scheme designs. Bi₂O₃/g-C₃N₄ composites offer better visible absorption, but they suffer from rapid charge recombination and insufficient active sites. The limitations of incomplete charge separation, narrow light absorption and compromised redox power necessitate the use of ternary heterojunctions, which provide multiple charge transfer pathways and enhanced surface reactivity [35].

The TiO₂/Bi₂O₃/g-C₃N₄ system is expected to promote efficient charge separation by facilitating either a step (S)-scheme or a Z-scheme charge transfer mechanism. In this process, less useful charge carriers with weaker redox power recombine at the interface, while the most potent electrons and holes remain in their respective components [36]. This achieves exceptional charge separation without compromising high redox potentials [37,38], making it an excellent design concept for maximizing photocatalytic efficiency for both organic pollutants and biological contaminants degradation and deactivation.

This work focuses on the synthesis and application-oriented testing of a new ternary heterojunction nanocomposite: TiO₂/Bi₂O₃/g-C₃N₄. We hypothesize that optimally integrating these three semiconductors will create an efficient charge transfer pathway, resulting in a material with significantly improved photocatalytic activity in the visible spectrum compared to binary or single-component materials. The efficiency of the synthetised materials is evaluated using two models: (1) degradation of the toxic, recalcitrant dye Rhodamine B (RhB), which is used as a representative model compound for dye pollutants from various industrial sources, including but not limited to hospital-related discharges; and (2) antibacterial activity against the nosocomial pathogens Escherichia coli (E. coli) and Staphylococcus aureus (S. aureus), which represent Gram-negative and Gram-positive bacteria, respectively. Converging materials engineering, photocatalysis, and antimicrobial design, this study provides a comprehensive blueprint for multifunctional nanocomposites that tackle coupled chemical and biological contaminants. Our work paves the way for smarter, safer, and more sustainable infection control and environmental safety protocols in healthcare.

2. Results and Discussion

2.1. Characterization of Synthesized Catalyst

2.1.1. X-Ray Diffraction (XRD) Analysis

Figure 1 displays the X-ray diffraction (XRD) patterns of the synthesized heterojunction nanocomposites.

The diffraction peaks confirm the successful formation of multiphase crystalline structures composed of anatase TiO₂, monoclinic α-Bi₂O₃, and graphitic carbon nitride (g-C₃N₄). For the TiO₂/Bi₂O₃ (90/10) sample calcined at 600 °C, the diffractogram clearly reveals the coexistence of three crystalline phases: anatase TiO₂ (tetragonal) as the major phase, rutile TiO₂ (tetragonal) as a minor phase, and monoclinic α-Bi₂O₃. The characteristic peaks of anatase TiO₂ (JCPDS Card No. 21-1272) are observed at 2θ = 25.3°, 37.8°, 48.1°, 54.0°, and 62.7°, corresponding to the (101), (004), (200), (105), and (204) planes, respectively. The presence of α-Bi₂O₃ (JCPDS Card No. 41-1449) is evidenced by diffraction peaks at 2θ = 27.4°, 33.2°, and 46.3°, attributed to the (120), (200), and (041) planes. For the ternary TiO₂/Bi₂O₃/g-C₃N₄ composites with mass ratios of 8:1:1 and 8.5:0.5:1, the diffractograms exhibit similar diffraction patterns, including the anatase, rutile, and α-Bi₂O₃ phases. A peak at 2θ = 27.53° may be attributed to both g-C₃N₄ and α-Bi₂O₃, suggesting possible peak overlap. However, this reflection is most likely dominated by α-Bi₂O₃, as g-C₃N₄-a non-metallic polymeric semiconductor, typically exhibits poor crystallinity after calcination, and its characteristic (002) interlayer stacking peak near 27.5° may be overshadowed by the more intense diffraction from the crystalline oxide phases. Consequently, while XRD alone cannot definitively confirm the presence of g-C₃N₄, the samples are consistent with ternary TiO₂/α-Bi₂O₃/g-C₃N₄ composites in which anatase remains the predominant phase. The formation of such multiple heterojunctions is particularly advantageous for photocatalytic applications, as it extends light absorption into the visible range through the incorporation of narrower-band-gap semiconductors (Bi₂O₃ and g-C₃N₄) and promotes efficient separation of photogenerated charge carriers at the interfaces between phases. This synergistic effect is expected to reduce electron–hole recombination and enhance the photocatalytic degradation efficiency of organic pollutants in water.

2.1.2. Scanning Electron Microscopy (SEM)

The surface morphology of the synthesized pure TiO₂, Bi₂O₃, 80T-BC, 85T-BC and 90T-B catalysts was examined using Scanning Electron Microscopy (SEM), with the corresponding micrographs presented in Figure 2.

The image of pure TiO₂ (Figure 2a) reveals particles with an approximately spherical morphology. These particles exhibit a non-uniform size distribution and a propensity for dense agglomeration. This agglomeration is typical of metal oxide nanoparticles synthesised using high-temperature methods. SEM image of Bi₂O₃ shows the presence of agglomerated structures. Following the incorporation of the various components, the morphological characteristics of the composites did not change significantly. The 80T-BC, 85T-BC, and 90T-B composites (Figure 2b,d,e) have a compact structure. The SEM image of the sample 90T-B appears to be more compact than the other two samples 80T-BC and 85T-BC, and exhibits locally denser and brighter areas, showing partial segregation relative to the distribution of Bi₂O₃ particles on the TiO₂. The presence of g-C₃N₄ in the 80T-BC and 85T-BC simples provides better homogeneity and dispersion of phases. This morphology suggests good integration of the components, where dense packing could potentially offer greater surface accessibility, which is beneficial for catalytic applications. Fundamentally, this specific structural arrangement creates an extensive network of interparticle connections. Such intimate contact and interconnection are highly advantageous for photocatalytic applications as they facilitate the efficient transfer of charge carriers between particles. This can potentially reduce electron–hole recombination and improve overall photocatalytic performance.

2.1.3. Diffuse Reflectance Spectroscopy (DRS)

The optical properties and band gap energies of the synthesised TiO₂/Bi₂O₃/g-C₃N₄ ternary heterojunction material were investigated using UV-Vis diffuse reflectance spectroscopy (DRS). Figure 3 shows the DRS spectra of the heterojunction and its pristine components (TiO₂, Bi₂O₃, and g-C₃N₄) across a wavelength range of 200–800 nm. Pristine TiO₂ exhibits a sharp absorption edge at around 380 nm, which is attributed to its intrinsic band-to-band transition. This UV absorption is characteristic of anatase TiO₂ as a wide-band-gap semiconductor (3.15 eV), with negligible absorption in the visible region (>400 nm). In contrast, Bi₂O₃ shows an absorption edge extending to around 420 nm, confirming its responsiveness to visible light, with an estimated band gap of approximately 2.65 eV. The TiO₂/Bi₂O₃/g-C₃N₄ ternary heterojunction exhibits a composite absorption profile integrating features of its three constituents. Notably, it exhibits significantly greater absorption in the visible spectrum (400–600 nm) than TiO₂, while maintaining strong UV absorption. This enhanced visible-light harvesting capability is attributed to the incorporation of narrow-band-gap components (Bi₂O₃ and g-C₃N₄), which can absorb lower-energy photons. The gradual rather than abrupt absorption edge indicates effective electronic coupling between the components, which is essential for efficient heterojunction performance. The optical band gap energies were determined using the Tauc plot method based on the Kubelka–Munk function (Equation (1)):

$$F\left(R\right)={\displaystyle \frac{{(1-R)}^2}{2R}}$$

(1)

where R is the reflectance. For indirect band gap semiconductors (TiO₂ and g-C₃N₄), the Tauc relation is expressed as Equation (2):

$${\left[F\left(R\right)h\nu \right]}^{1/2}=A\ast (h\nu -E_g)$$

(2)

while for direct band gap semiconductors (Bi₂O₃), Equation (3) can be used:

$${\left[F\left(R\right)h\nu \right]}^2=A\ast (h\nu -E_g)$$

(3)

The effective optical band gap of the TiO₂/Bi₂O₃/g-C₃N₄ ternary heterojunction has been estimated to be approximately 2.94 eV. This value lies between the band gaps of its constituents, Bi₂O₃ (2.65 eV) and TiO₂ (3.15 eV), strongly suggesting that each component that is active in visible light contributes synergistically to the composite’s extended absorption profile. The reduction in the band gap observed compared to pure TiO₂, and its positioning between the band gaps of Bi₂O₃ and g-C₃N₄ (

Table 1. Optical characteristics of synthesized materials.

Materials	Valence Band (VB)	Conduction Band (CB)	Band Gap (eV)	Wavelength (nm)
TiO2	2.7	−0.45	3.15	394
Bi2O3	3.13	0.40	2.73	468
g-C3N4	1.4	−1.3	2.7	459
90T-B	-	-	3.00	413
85T-BC	-	-	2.94	421
80T-BC	-	-	2.86	433

) implies that an integrated electronic structure has formed rather than a simple physical mixture. Previous studies on similar ternary systems have reported comparable band gap values ranging from 2.2 to 2.9 eV depending on the composition and synthesis method used [39,40]. The staggered (type II) or staircase alignment of the bands, which usually occurs between these three components, should create an optimal energy landscape at the heterojunction interfaces. Various studies have proposed Z-type or double Z-type charge migration mechanisms for TiO₂/Bi₂O₃/g-C₃N₄ heterojunctions. These mechanisms facilitate the efficient spatial separation of photogenerated charge carriers by directing electrons and holes to migrate in opposite directions across the interfaces [40].

2.2. Photocatalytic Activity Study

The photocatalytic performance of the TiO₂, Bi₂O₃, 80T-BC, 85T-BC, and 90T-B heterojunction was evaluated through Rhodamine B (RhB) degradation under visible light xenon lamps BEETRO^® (75 W, wavelength = 400 nm). Figure 4 presents the time-dependent degradation profiles, kinetic analysis, and comparative performance with control samples.

Figure 4 shows the photocatalytic degradation efficiency of rhodamine B (RhB) on the ternary heterojunction TiO₂/Bi₂O₃/g-C₃N₄ compared to TiO₂ and Bi₂O₃. Upon visible radiation irradiation, the ternary heterojunction exhibited significantly enhanced photocatalytic activity compared to all control samples. Complete RhB degradation (>99%) was achieved within 120 min, whereas TiO₂ reached only 78% in the same time. The degradation kinetics were analyzed using the pseudo-first-order kinetic model (Equation (4)) [41] and reported in

Table 2. Values of the rate constant (k) for the degradation of the RhB dye 85TB-C, pH = 6.

Catalyst	Rate Constant k (min−1)	R2
Bi2O3	0.0048	0.74
TiO2	0.0165	0.72
80T-BC	0.0113	0.97
85T-BC	0.0338	0.95
90T-B	0.010	0.96

:

$$\left({\displaystyle \frac{Cads}{Ct}}\right) =k\ast t$$

(4)

where k is the apparent rate constant, C₀ is the initial concentration after dark adsorption, and C is the concentration at irradiation time t. The linear plots of $-$ln(C_ads/C_t) versus irradiation time confirm that RhB photodegradation follows pseudo-first-order kinetics under the experimental conditions.

Table 2 shows the calculated rate constants (k) and correlation coefficients (R²) for RhB degradation using the synthesised catalysts. The ternary heterojunction 85T-BC showed the highest rate constant among all materials, at k = 0.0338 1/min, which is around 2.0 times higher than pure TiO₂ (0.0165 1/min) and approximately 7.0 times higher than pure Bi₂O₃ (0.0048 1/min). This enhancement of reaction kinetics highlights the synergistic effect of combining TiO₂, Bi₂O₃, and g-C₃N₄ in a single heterojunction architecture. The superior kinetic efficiency of 85T-BC can be attributed to its structural and optical properties. As demonstrated by the DRS analysis, the ternary heterojunction shows an increased absorption range extending into the visible spectrum (up to 421 nm) and an optimised band gap of 2.94 eV, allowing for the effective collection of visible light. Furthermore, the Z-scheme charge transfer mechanism depicted in Figure 5 ensures that photogenerated electrons and holes are effectively separated by the interfaces of the three components, thereby suppressing the detrimental electron–hole recombination that limits the efficiency of pure TiO₂ and Bi₂O₃.

Figure 5a shows the effect of active radicals on the photocatalytic degradation performance of the 85T-BC sample, as determined using selective scavengers. In the absence of any scavenger, the degradation efficiency reached 99%. Adding benzoquinone (BQ), a superoxide radical scavenger, reduced the efficiency to 68.24%, indicating that superoxide radicals play a moderate role in the process.

A more significant decrease was achieved with the addition of isopropanol (IPA), a hydroxyl radical (•OH) scavenger, which lowered the efficiency to 29.31%. The most notable inhibition was caused by EDTA-2Na, a hole scavenger, which lowered the degradation efficiency to just 20.43%. These results indicate that, while all three reactive species participate in the photocatalytic process, holes (h⁺) and hydroxyl radicals (•OH) are the most significant active species that contribute to the high degradation efficiency of the TiO₂/Bi₂O₃/g-C₃N₄ sample, whereas superoxide radicals (•O₂⁻) play a comparatively minor role.

The reusability and photostability of 85T-BC were investigated through consecutive photocatalytic cycles. After each cycle, the catalyst was recovered by centrifugation and rinsed thoroughly with deionised water. It was then dried at 50 °C in ambient air for two hours. As shown in Figure 5b, the photocatalytic activity of 85T-BC towards RhB dye degradation under solar irradiation was virtually unaffected during the first two cycles and decreased by only 10% after five cycles.

2.3. Photocatalytic Mechanisms

The TiO₂/Bi₂O₃/g-C₃N₄ composite synthesised in this study exhibits a direct Z-scheme charge transfer mechanism, which enables the efficient spatial separation of photogenerated carriers. When the heterojunction is irradiated with light of equal or greater energy, the three semiconductors simultaneously generate electron–hole pairs (Figure 6). The favourable alignment of their band structures means that the electrons generated in the conduction band (CB) of TiO₂ (approximately −0.62 eV vs. ENH) and Bi₂O₃ (approximately +0.33 eV vs. ENH) migrate to the interface, where they recombine with holes generated in the valence band (VB) of g-C₃N₄ (approximately +1.44 eV vs. ENH) [40,42]. This interfacial recombination process is characteristic of the Z-scheme (Figure 5). This is due to the accumulation of electrons in the band gap (BG) of g-C₃N₄, which has a highly negative potential of approximately −1.22 eV vs. ENH and provides strong reducing power. This is ideal for reactions such as hydrogen production or CO₂ reduction. Conversely, holes accumulate in the valence bands of TiO₂ (approximately +2.65 eV vs. ENH) and Bi₂O₃ (approximately +3.13 eV vs. ENH), whose positive potentials provide exceptional oxidizing power for mineralizing organic pollutants or generating reactive oxygen species [42]. This orchestration of charge transfer minimizes the harmful recombination of carriers within the same material synergistically, thereby increasing their lifetime and giving the ternary heterojunction photocatalytic superior activity compared to its individual components.

The photogenerated electrons and holes migrate to the catalyst surface after excitation. They can engage in redox reactions there with species like O₂ (Equation (5)) and H₂O (Equation (6)) that have been adsorbed from the surrounding media. Highly reactive species, such as hydroxyl radicals (OH•) and superoxide radical anions (O₂•⁻), are created as a result of this interfacial charge transfer [31,43]. According to Equations (7) and (8), these radicals start a series of oxidative processes that eventually mineralize the dye pollutant into safe for the environment by producing CO₂ and H₂O [44,45].

$$e^{-}+O_2\to O_2^{.-}$$

(5)

$$\begin{gathered} h^{+}+H_{2}O\to {OH}^{.}+H^{+} \\ {h}^{+}+{OH}^{-}\to {OH}^{.} \end{gathered}$$

(6)

$$O_2^{.-}+{OH}^{.}\to {HOO}^{.}$$

(7)

$${HOO}^{.}+e^{-}\to {HO}_2^{-}+h^{+}\to H_2{O}_2+hƲ \to {2OH}^{.}$$

(8)

$$\mathbf{Photocatalysis} \mathbf{decomposition}:$$

$$\left\{\begin{array}{c}{O}_2^{.-}+RhB\to {CO}_2+H_{2}O \\ {OH}^{.}+RhB\to {CO}_2+H_{2}O \end{array}\right.$$

(9)

Table 3. Comparative catalytic performance for the degradation of Rhodamine B (RhB).

Catalyst	Mechanism/Activation	Initial RhB (mg⋅L−1)	Operating Conditions (pH, Temp., Cat. Dose, Oxidant)	Time (min)	Degradation (%)	Source Document (Ref.)
g−C3N4/WO3/WS2	Photocatalysis (Visible Light)	25	300 W Xe lamp (>420 nm)	20	96.2	[46]
Flower-like Bi12TiO20/g−C3N4	Photocatalysis (Visible Light)	20	150 mW⋅cm−2 Xe lamp (>420 nm)	30	100	[46]
CdS/CQDs/g−C3N4	Photocatalysis (Visible Light)	10	300 W Xe lamp (>420 nm)	20	100	[46]
Yb oxide-based GO/g−C3N4/Fe2O3	Photocatalysis (Visible Light)	10	Light bulb irradiation	45	67.11	[46]
Ce-based GO/g−C3N4/Fe2O3	Photocatalysis (Visible Light)	10	Light bulb irradiation	45	63.08	[46]
Cu/SiO2	Fenton-like (H2O2)	10	pH 7, 60 °C, 5 g/L cat., 3 mM H2O2	10	95.0	[47]
Cu+−g−C3N4	Fenton-like (H2O2)	50	pH 7, 25 °C, 0.8 g/L cat., 40 mM H2O2	60	99.2	[47]
Cu/c−PCB	Fenton-like (H2O2)	10	pH 6.7, 30 °C, 1 g/L cat., 50 mM H2O2	360	95.8	[47]
Cu@CuO nanowires	Fenton-like (H2O2)	12	pH 7.3, 25 °C, 812 mM H2O2	16	100	[47]
Cu−Fe3O4 MNPs	Fenton-like (H2O2)	Not reported	pH 6.9, 25 °C	120	95.2	[47]
Mesh-type monolithic Cu/Fex/γ−Al2O3/Al	Fenton-like (H2O2)	10	Natural pH, 50 °C, 1200 ppm H2O2	60	99.5	[47]
Cu- and Fe-doped Al-MCM-41	Fenton-like (H2O2)	400	pH 7, 60 °C, 1 g/L cat., 40 mM H2O2	50	96.0	[47]
FeCu@BC600−2	Fenton-like (H2O2)	10	pH 3, 30 °C, 0.2 g/L cat., 1 mM H2O2	60	100	[47]
CdNiZnO NPs	Photocatalysis (UV Visible light)	30	15 W lamp (254 nm/365 nm)	50	98%	[48]
85T-BC	Photocatalysis (Visible Light)	5	pH 6	110	100	This study

reports a comparison of the catalytic performance of the designed heterojunctions with literature.

2.4. Antibacterial Activity

• Against Pseudomonas aeruginosa ATCC 9027

The antibacterial efficacy of the synthesized heterojunction was investigated against the Gram-negative model strain Pseudomonas aeruginosa ATCC 9027 using a direct confrontation assay. Bacterial viability was quantified by colony-forming unit (CFU) enumeration, and the corresponding inactivation efficiency (%) was calculated according to Equation (4). The results are summarized in

Table 4. Effect of the synthesized nanoparticles on S. aureus ATCC 6538 and P. aeruginosa ATCC 9027 viability. CFU Count and inactivation percentages (%) Inact from cultures obtained after 60 min contact between the bacterium and 85T-BC.

	Description	P. aeruginosa ATCC 9027		S. aureus ATCC 6538
		CFU	% Inact (Mean ± SD)	CFU	% Inact (Mean ± SD)
T1	Control (visible light)	37.5 ± 2.12	0.00 e	28,5 ± 4.94	0.00 e *
T2	Control (darkness)	253.5 ± 47.37	0.00 e	165 ± 35.35	0.00 e
T3	500 µg/mL (visible light)	9.5 ± 0.70	74.68 ± 0.45 d	9 ± 1.41	68.38 ± 0.53 d
T4	500 µg/mL (darkness)	179 ± 1.41	28.18 ± 12.86 d	40.5 ± 2.12	75.02 ± 4.07 c
T5	1000 µg/mL (visible light)	5 ± 2.82	86.86 ± 6.79 c	4 ± 1.41	86.19 ± 2.56 b
T6	1000 µg/mL (darkness)	17.5 ± 0.70	93.00 ± 1.03 b	15.5 ± 1.41	90.72 ± 1.01 b
T7	2000 µg/mL (visible light)	0.00	100.00 ± 0.00 a	0.00	100.00 ± 0.00 a
T8	2000 µg/mL (darkness)	2.5 ± 0.70	99.02 ± 0.10 a	0.5 ± 0.70	99.74 ± 0.37 a
T9	5000 µg/mL (visible light)	0.00	100.00 ± 0.00 a	0.00	100.00 ± 0.00 a
T10	5000 µg/mL (darkness)	0.00	100.00 ± 0.00 a	0.00	100.00 ± 0.00 a

* Different lowercase letters within the same column indicate statistically significant differences among treatments according to Tukey’s honestly significant difference (HSD) test (p < 0.05) following one-way analysis of variance (ANOVA).

.

The data demonstrate a pronounced antibacterial activity of the nanomaterial, strongly influenced by the applied concentration, thereby evidencing a clear dose–response relationship. Notably, this activity was significantly enhanced under visible light irradiation (Figure 7), highlighting the contribution of photocatalytic processes to the overall antibacterial performance. In contrast, control conditions exhibited substantial bacterial proliferation, particularly in the absence of light, confirming the negligible intrinsic inhibition under dark conditions. Exposure to visible light alone (i.e., in the absence of nanoparticles) induced only a modest reduction in bacterial growth, suggesting a limited photoinhibitory effect.

At a concentration of 500 µg/mL, the nanomaterial induced a marked reduction in bacterial viability under both illuminated and dark conditions, with a more pronounced effect observed under light irradiation. Increasing the concentration to 1000 µg/mL further enhanced antibacterial activity, indicating a progressive improvement in efficacy. At 2000 µg/mL, near-complete bacterial inactivation was achieved in the dark (99.02% ± 0.10), whereas total inhibition (100% ± 0.00) was observed under visible light exposure. At the highest tested concentration (5000 µg/mL), complete bacterial inactivation was achieved irrespective of illumination conditions.

Collectively, these findings (Table 4) demonstrate a synergistic interaction between photocatalytic activation and the intrinsic antimicrobial properties of the 85T-BC heterojunction, culminating in the complete suppression of P. aeruginosa at concentrations ≥2000 µg/mL under visible light irradiation.

• Against Staphylococcus aureus ATCC 6538

The antibacterial activity of the synthesized nanoparticles against the Gram-positive strain Staphylococcus aureus ATCC 6538 was evaluated by quantifying viable cells through colony-forming unit (CFU) enumeration. The corresponding inactivation efficiency (%) was calculated according to Equation (4), and the results are presented in Table 4.

The data reveal a pronounced antibacterial effect that is strongly dependent on nanoparticle concentration, indicating a clear dose–response relationship. In the control groups, a high bacterial load was observed, particularly under dark conditions, confirming the absence of significant intrinsic inhibition. Conversely, exposure to visible light alone resulted in a noticeable reduction in bacterial counts (Figure 8), suggesting a moderate photoinhibitory effect.

At a concentration of 500 µg/mL, nanoparticle treatment significantly reduced bacterial viability under both dark and illuminated conditions. This antibacterial effect increased progressively with concentration, with 1000 µg/mL and 2000 µg/mL leading to near-complete inactivation in the dark (99.74% ± 0.17) and complete inactivation under visible light irradiation (100% ± 0.00). At the highest tested concentration (5000 µg/mL), total bacterial inhibition was achieved irrespective of illumination conditions.

The antibacterial activity of the synthesized heterojunction against the Gram-positive Staphylococcus aureus ATCC 6538 and the Gram-negative Pseudomonas aeruginosa ATCC 9027 reveals distinct susceptibility patterns, particularly at lower nanoparticle concentrations. Overall, both strains exhibited a concentration-dependent reduction in viability; however, P. aeruginosa consistently demonstrated higher resistance compared to S. aureus under identical treatment conditions.

At 500 µg/mL under visible light irradiation, P. aeruginosa showed slightly higher inactivation efficiency (74.68%) compared to S. aureus (68.38%), suggesting an early contribution of photocatalytic activity. In contrast, under dark conditions, a marked difference was observed, with S. aureus exhibiting substantially higher susceptibility (75.02%) than P. aeruginosa (28.18%), indicating that the intrinsic antibacterial activity of the nanomaterial is less effective against Gram-negative bacteria in the absence of light.

This differential behavior can be attributed to structural differences in the bacterial cell envelope. Gram-negative bacteria possess an additional outer membrane enriched in lipopolysaccharides, which acts as a permeability barrier and limits nanoparticle penetration and interaction. Conversely, the thicker but more permeable peptidoglycan layer in Gram-positive bacteria facilitates greater interaction with the nanomaterial, enhancing its antibacterial effect under non-irradiated conditions [49,50].

At intermediate concentrations (1000 µg/mL), the antibacterial activity increased substantially for both strains, with reduced differences between them, suggesting that higher nanoparticle doses can partially overcome the protective barrier of Gram-negative bacteria. At concentrations ≥2000 µg/mL, both bacterial strains exhibited near-complete or complete inactivation regardless of illumination conditions, indicating that the combined effects of nanoparticle concentration and photocatalytic activation are sufficient to induce total bacterial eradication.

Under visible light irradiation, the differences between Gram-positive and Gram-negative bacteria became less pronounced across all concentrations, highlighting the dominant role of photocatalytically generated reactive oxygen species (ROS) in bacterial inactivation mediated by TiO₂, Bi₂O₃, and g-C₃N₄ under light activation. These ROS, including hydroxyl radicals and superoxide anions, are known to induce oxidative stress by damaging critical cellular components. Specifically, they disrupt bacterial cell envelopes and compromise membrane integrity, while also targeting intracellular biomolecules such as proteins and nucleic acids, ultimately leading to irreversible cell death [51,52,53,54].

Recent studies have consistently demonstrated the remarkable antibacterial potential of TiO₂-based heterojunctions incorporating Bi₂O₃ and g-C₃N₄, highlighting their promise for advanced biomedical and environmental applications [55,56]. These composite systems exhibit enhanced photocatalytic and antibacterial performance under visible light irradiation, which is primarily attributed to improved charge carrier separation, increased specific surface area, and favorable heterojunction architecture.

The antibacterial efficiencies observed in the present study are consistent with previously reported TiO₂-based systems. For instance, Luthfiah et al. [57] reported inactivation rates of 98.69% for Staphylococcus aureus and 97.44% for Pseudomonas aeruginosa, as determined by optical density and dilution assays. Similarly, Li et al. [58] demonstrated that visible LED irradiation induces significant oxidative damage to the cell membrane of S. aureus, further supporting the role of ROS-mediated mechanisms in photocatalytic antibacterial activity.

3. Materials and Methods

3.1. Biological Material

In this study, two bacterial strains were used, including a Gram-positive bacterium, such as Staphylococcus aureus ATCC 6538, and a Gram-negative bacterium, Pseudomonas aeruginosa ATCC 9027 (ATCC: American Type Culture Collection).

3.2. Chemicals

The chemical reagents and solvents used in this study were analytical grade bismuth (III) oxide (Bi₂O₃, 99.9%, MW = 465.96 g mol⁻¹), titanium oxide (TiO₂, 99.5%, MW = 79.89 g mol⁻¹), Urea (CO(NH₂)₂), Rhodamine B (479.02 g mol⁻¹), NaOH (39.99 g mol⁻¹), HCl (36.47 g mol⁻¹), which were obtained from Sigma Aldrich^® (Darmstadt, Germany). All reagents were used without further purification, and all solutions were prepared with locally produced deionized water.

3.3. Synthesis of TiO₂/Bi₂O₃/g-C₃N₄ Catalyst

In this study, the heterojunctions TiO₂/Bi₂O₃/g-C₃N₄ were synthesized through a common solid-state reaction route [25]. Bismuth oxide (Bi₂O₃) powder was first homogenized by grinding in a ceramic mortar to ensure uniform particle size distribution. The ground Bi₂O₃ was subsequently calcined in a muffle furnace (Nabertherm^®, Lilienthal, Germany) at 500 °C for 2 h. Separately, urea was subjected to thermal polymerization by calcination at 550 °C for 2 h, leading to the formation of graphitic carbon nitride (g-C₃N₄). For the preparation of the composite material, (90-x)TiO₂/10Bi₂O₃/x g-C₃N₄, the appropriate amounts of the starting materials were mixed and thoroughly ground in an agate (or ceramic) mortar for 2 h [25]. The resulting mixture was then calcined at 600 °C for 6 h to form the final product. A series of samples was prepared with a fixed Bi₂O₃ content of 10 wt.% and varying TiO₂ concentrations. These composites are labeled as 80T-BC, 85T-BC, and 90T-B, where T stands for TiO₂, B for Bi₂O₃ and C for g-C₃N₄ while the number indicates the weight percentage of TiO₂ (i.e., y = 80%, 85%, and 90%, respectively).

3.4. Photodegradation of Rhodamine B Dye

The photocatalytic efficiency of the catalysts 80T-BC, 85T-BC, 90T-B, TiO_2, and Bi₂O₃ was evaluated through the degradation of Rhodamine B (RhB) dye in aqueous solution under visible light irradiation xenon lamps BEETRO^® (75 W, wavelength = 400 nm). A catalyst loading (0.5 g) was dispersed in a 100 mL solution containing RhB dye at a concentration of 5 mg/L and pH = 6. Before the light irradiation, the suspension of the catalyst in the RhB solution was first stirred for 30 min in darkness to achieve adsorption–desorption equilibrium. Then the solution was irradiated under the light source with constant stirring. Samples of 2 mL were taken every 15 min, and absorbances were measured at 665 nm. The degradation efficiency (R) was determined using Equation (10).

$$\mathrm{R} (\%)={\displaystyle \frac{c_0-c_t}{c_0}}\times 100={\displaystyle \frac{A_0-A_t}{A_0}}\times 100$$

(10)

where C₀ and A₀ represent the concentration and the absorbance of dye after the adsorption and desorption equilibrium, and C_t and A_t represent the concentration and the absorbance after a given time of degradation. Therefore, examining the degradation kinetics is essential, as it provides critical insights into the efficiency and rate of the process. To determine the rate constant, the Langmuir-Hinshelwood model (Equation (11)) was applied, which simplifies to a pseudo-first-order kinetic model (Equation (12)):

$$r={\displaystyle \frac{-d{C}_t}{dt}}={\displaystyle \frac{kK{C}_t}{1+K{C}_t}}$$

(11)

$$C_t=C_{ads}\ast e^{-k^{\prime }t}$$

(12)

where k is the photocatalytic rate constant [mg/(min·L)], K is the adsorption equilibrium constant [L/mg], and k′ is the pseudo-first-order rate constant [1/min].

The contributions of superoxide radicals (•O₂⁻), holes (h⁺) and hydroxyl radicals (•OH) to the photocatalytic mechanism were examined using targeted radical inhibitors. Benzoquinone (BQ) was used to quench superoxide radicals, sodium ethylenediaminetetraacetate (EDTA-2Na) to scavenge holes, and isopropanol (IPA) to inhibit hydroxyl radicals. The inhibitory experiments were performed under conditions similar to those of the photocatalytic tests, with the exception of the addition of the respective inhibitors to the solution.

3.5. The Characterization of Synthesized Catalysts

A series of characterization techniques was used to determine the structural, optical, and electronic properties of the synthesized 90-xTiO₂/xBi₂O₃/10 g-C₃N₄ heterojunction at different ratios and the raw materials Bi₂O₃, TiO₂.

The crystalline phases and crystallographic structure were determined by X-ray diffraction (XRD) using a PANALYTICAL EMPYREAN^® (Almelo, The Netherlands) diffractometer with CuKα radiation, scanning a 2θ range from 20° to 80°.

The surface morphology and particle shape were analyzed using a scanning electron microscope (SEM, JEOL Ltd., Tokyo, Japan) at different magnifications.

The optical properties, namely the estimation of the band gap, were studied by diffuse reflectance spectroscopy (DRS) between 200 and 800 nm using an Evolution 220 UV-Vis spectrometer (Thermo Fisher Scientific, Madison, WI, USA).

3.6. Antibacterial Activity

The antibacterial activity of the synthesized heterojunction nanoparticles was evaluated using a quantitative liquid-phase contact (confrontation) assay against representative Gram-negative and Gram-positive bacteria, namely Pseudomonas aeruginosa (ATCC 9027) and Staphylococcus aureus (ATCC 6538), respectively.

3.6.1. Preparation of Nanoparticle Suspensions

Stock suspensions of nanoparticles were prepared at concentrations of 500, 1000, 2000, and 5000 µg·mL⁻¹ by dispersing the appropriate mass of nanoparticle powder in sterile physiological saline (0.9% NaCl). Suspensions were magnetically stirred until complete homogenization and subsequently adjusted to pH 7.0 ± 0.1.

3.6.2. Preparation of Bacterial Inocula

Fresh bacterial cultures were grown overnight and adjusted to an optical density of OD₆₀₀ = 0.08–0.10, corresponding to approximately 10⁸ CFU·mL⁻¹. The suspensions were then serially diluted in sterile saline to obtain a working inoculum of ~10⁵ CFU·mL⁻¹.

3.6.3. Antibacterial Assay Conditions

For each experimental condition, 1 mL of the bacterial suspension (~10⁵ CFU·mL⁻¹) was mixed with 9 mL of nanoparticle suspension at the desired concentration. The mixtures were incubated for 60 min under both visible light irradiation and dark conditions. A total of 10 experimental conditions were evaluated for each bacterial strain:

• T1: control under visible light (bacteria + sterile saline solution);
• T2: control in dark (bacteria + sterile saline solution);
• T3: 500 µg/mL nanoparticles under visible light;
• T4: 500 µg/mL nanoparticles in dark;
• T5: 1000 µg/mL nanoparticles under visible light;
• T6: 1000 µg/mL nanoparticles in dark;
• T7: 2000 µg/mL nanoparticles under visible light;
• T8: 2000 µg/mL nanoparticles in dark;
• T9: 5000 µg/mL nanoparticles under visible light;
• T10: 5000 µg/mL nanoparticles in dark.

All experiments were performed in duplicate.

3.6.4. Enumeration of Viable Bacteria

Following incubation, aliquots (100 µL) from each treatment were spread onto nutrient agar plates and incubated at 37 °C for 18–24 h. Viable bacterial cells were quantified by counting colony-forming units (CFU).

3.6.5. Calculation of Antibacterial Activity

The antibacterial efficiency was expressed as the percentage of bacterial inactivation (%), calculated by comparing the number of viable cells after treatment (N) to that of the untreated control (N₀), according to Equation (13):

$$\mathrm{Bacterial} \mathrm{inactivation} (\%) = {\displaystyle \frac{N_0-N}{N_0}}\times 100$$

(13)

where N₀ is the CFU count of the control sample; N is the CFU count after treatment.

4. Statistical Analysis

Data from the microbiological analysis are presented as mean ± standard deviation (SD). Statistical differences among treatment groups were evaluated using one-way analysis of variance (ANOVA). When significant effects were detected, Tukey’s honestly significant difference (HSD) test was applied for post hoc multiple comparisons to identify statistically distinct groups. Differences were considered statistically significant at p < 0.05. All analyses were performed using the R statistical environment (R Foundation for Statistical Computing, Vienna, Austria).

5. Conclusions

This study successfully produced a novel ternary n-p heterojunction photocatalyst, TiO₂-Bi₂O₃-g-C₃N₄, using a ceramic technique. The prepared materials were thoroughly characterised and their photocatalytic effectiveness evaluated by study of the degradation of a cationic dye model pollutant (RhB). Structural analysis revealed that pure crystalline phases with a uniform distribution of spherical grains formed when Bi₂O₃ and g-C₃N₄ were added to the TiO₂ matrix.

Optical analysis using DRS showed that the ternary heterojunction has an extended absorption edge into the visible spectrum (up to ~421 nm) and an optimised band gap of 2.94 eV, allowing for the efficient harvesting of visible light. Near-complete degradation (>99%) was achieved within 110 min, with a rate constant (k = 0.0338 1/min) about 2.0 times higher than that of pure TiO₂ and around 7.0 times higher than that of pure Bi₂O₃.

Additionally, the composite exhibited notable antibacterial activity, particularly against S. aureus, achieving nearly complete eradication of viable cells at concentrations as low as 0.5 mg/mL. These results suggest that the synthesised composite could be an effective, multifunctional agent for cleaning contaminated water and disinfecting surfaces in hospital environments.