Ba-doped TiO2 nanotube hybrid nanoarchitecture as highly efficient photocatalysts

Pristine TiO₂ was sourced from commercially available anatase TiO₂, purchased from Alfa Aesar. Barium acetate with a molecular weight of 255.43 g/mol, purchased from Merck, and sodium hydroxide (NaOH) with a molecular weight of 40 g/mol and HCl, both obtained from Sigma-Aldrich, were all utilized without any further purification. Purified water used for the analysis was obtained from the Millipore Milli-Q water system. Indigo carmine (IC), Azocarmine-G (AZO), and Benzopurpurin (BP) were purchased from Himedia (USA), Biognost (EU, Croatia), and Sigma-Aldrich (UK), respectively.

TiO₂ nanotubes were synthesized from the anatase powder, by using a simple hydrothermal method. About 1.2 g of pristine TiO₂ was dissolved in 20 ml 10 N NaOH solution. The mixture was transferred to the Teflon-lined container of an autoclave. The autoclave was placed inside the oven for 15 h at 130 °C. Then, the autoclave was left to cool down to room temperature, and the precipitate was washed multiple times with distilled water and three times with 0.1N HCl, followed by a final wash with distilled water. The filtration was performed with a suction pump, and the sample was dried at 80 °C for 3 h.

The synthesis of barium-doped TiO₂ nanotubes was also carried out using the hydrothermal method. Approximately 1.2 g of barium acetate was measured along with 1.2 g of the previously prepared TiO₂ nanotubes. Both were mixed thoroughly with 20 ml of 10 N NaOH. After mixing, the solution was transferred to an autoclave. The autoclave was placed in an oven at 130 °C for 15 h. After cooling, the solution in the autoclave was washed several times with distilled water and 0.1 N HCl. The neutrality was checked after an additional wash with distilled water. The filter paper was transferred to a petri dish for drying at 80 °C for 3 h in an oven.

The morphology of the samples was investigated using a TEM (Phillips CM 12) and SEM (NOVA Nano SEM450, FEI, Thermo Fisher, Netherlands). The crystal structure of the sample was determined using XRD. The XRD pattern was acquired using Cu-Kα radiation (1.54060 Å) on a Malven Panalytical Xpert X-ray diffractometer. The synthesized nanoparticles were characterized by FTIR spectra from 400 to 4000 cm⁻¹, recorded on a Perkin Elmer Frontier MIR‐FIR spectrophotometer. UV–visible (UV–Vis) spectrophotometric experiments were carried out using a Biochrom UV–Visible spectrophotometer. The samples were scanned at medium speed within the 200–900 nm range.

The photocatalytic activity of both TiO₂ and Ba-doped TiO₂ was evaluated through the degradation of IC, AZO, and BP dyes under sunlight irradiation. The measurements for all experiments were taken from 9 a.m. to 1 p.m., with an average temperature above 30 °C and a reported UV index of 9. All experiments were conducted at the same time of the day under comparable sunlight exposure. Solar intensity was measured using a lux meter and found to vary within ± 5%, confirming stable irradiation conditions. The dyes were chosen as models of organic pollutants due to their non-biodegradable nature and strong absorbance. The stock solution of each dye was prepared by dissolving 50 mg of dye in 1000 ml of double-distilled water. Different concentrations (10, 20, and 30 mg/L) were then prepared from the stock solution using distilled water. Both TiO₂ and Ba-doped TiO₂ photocatalysts were added to 50 ml of these dye solutions and magnetically stirred for 20 min in the dark to ensure uniform suspension and to eliminate dye surface adsorption by the photocatalysts. After exposure to sunlight, aliquots of the solution were extracted with a syringe at fixed intervals of 20 min and then centrifuged to settle the metal oxide at the bottom. The photocatalytic activity of the samples was measured by analyzing the concentration of residual dye in the supernatant. Spectrophotometric analysis was used to investigate degradation efficiency by monitoring dye absorbance. The maximum absorbance wavelengths of IC, AZO, and BP were 610, 540, and 495 nm, respectively. Changes in absorbance were used to calculate the degradation rate after exposure to visible light. The degradation efficiency of the dyes is determined using Eq. (1) as follows:where (\({A}_{0}\)) is the initial dye absorbance before irradiation, and (\({A}_{t}\)) is the absorbance after irradiation within time (t).

The morphology of the TiO₂ nanotubes and Ba-doped TiO₂ is shown in Fig. 1a and b, respectively. TiO₂ has a tubular shape with a thickness of approximately 8.84 ± 1.7 nm and an average length of 95.34 ± 40 nm, which is consistent with values reported in the literature [16]. As depicted in the SEM images in Fig. 2a and b, a dramatic morphological change is observed in TiO₂ nanotubes upon incorporation of Ba. The Ba-doped TiO₂ nanotubes show a trend toward shorter morphology when compared to pure TiO₂ nanotubes. This shortening can be ascribed to the incorporation of Ba²⁺ ions into the TiO₂ lattice, inducing strain in the structure as well as influencing the kinetics of growth during synthesis. The mismatch in the ionic radius between Ba²⁺ (1.61 Å) and Ti⁴⁺ (0.68 Å) induces extensive lattice distortion, thus suppressing the elongation of TiO₂ nanotubes [17].

Full size image

Full size image

These figures also reveal the diameter change of TiO₂ nanotubes upon Ba doping, showing that Ba doping induces changes in both the inner and outer diameters of the nanotubes. The Ba-doped samples show a tendency toward a more uniform diameter distribution, though with a slight reduction in average diameter compared to undoped samples. This phenomenon is consistent with previous studies demonstrating that alkaline earth metal doping affects the electrochemical anodization process, leading to modified pore formation dynamics [15]. The most dramatic dimensional change is in wall thickness. The wall thickness is larger in the Ba-doped TiO₂ systems compared to the undoped systems, as observed from their denser-looking SEM micrographs. This increase in thickness can be ascribed to the addition of larger Ba²⁺ ions, which occupy interstitial sites, leading to lattice expansion. The increase in wall thickness is directly proportional to the Ba doping concentration, with higher concentrations giving enhanced thickening effects [18]. Meanwhile, the Ba-doped TiO₂ nanotubes displayed a hybrid architecture [24], as confirmed by the TEM micrograph.

The XRD patterns shown in Fig. 1c illustrate the peaks of the TiO₂ nanotubes, located at angles 25.32°, 37.81°, 48.03°, 53.89°, 55.06°, 62.61°, 68.76°, 70.21°, and 75.13°, corresponding to the (101), (004), (200), (105), (211), (204), (220), (116), and (215) planes, respectively. This observed diffraction is equivalent to the tetragonal anatase structure of TiO₂ and precisely matches the XRD standard of TiO₂ (JCPDS: 020–2242) [19]. The XRD pattern indicates the pure-phase nature of the TiO₂ nanotube crystal, which is generally dominated by the thermodynamically stable (101) facets [20, 21].The XRD pattern of Ba-doped TiO₂ nanotubes showed additional peaks at angles 22.01°, 38.77°, 45.01°, 50.67°, and 65.75°, corresponding to the (100), (110), (111), (200), (211), and (220) planes, which match the barium-doped TiO₂ crystal structure (JCPDS: 31–0174) [21‐23]. An enhancement of the crystalline nature was observed after doping with Ba, which has a higher ionic radius [21]. From the XRD line broadening, the synthesized materials were estimated to be in the nanometer size range. It has also been reported that mixed rutile and anatase phases of TiO₂ nanoparticles exhibit higher photoactivity than either pure phase when excited by UV light [24]. Figure 1d shows the FTIR spectra of both TiO₂ and Ba-doped TiO₂ nanotubes, highlighting the changes in chemical bond states. The appearance of several absorption bands corresponds to organic groups and alkanes. The absorption band between 615 and 695 cm⁻¹ was attributed to the Ti–O–Ti bond, which confirmed the crystal nature of TiO₂ and formation of the anatase phase. The same absorption range indicates the characterization of the oscillation of M–O bonds (M = Ti and Ba) [21, 25]. The visible bands at 750 cm⁻¹ and 872 cm⁻¹ stopped changing after the doping and indicated Ti–O vibrational bands [22].

Moreover, the appearance of additional shoulders at 782 cm⁻¹ or 1001 cm⁻¹ can be attributed to Ba–O interactions and lattice distortions arising from Ba incorporation, the vibrational bands located at 1253 cm⁻¹ and 1503 cm⁻¹ are assigned to C–O and C=C stretching modes, respectively, which likely originate from residual organic moieties or carbonate species retained on the nanotube surface. These residues may derive from synthesis precursors, solvent remnants, or adsorbed carbonaceous species [22, 26], showed a change in intensity after Ba doping. The absorption at 1639 cm⁻¹ is attributed to the bending vibration (δHOH) of physisorbed water, which is enhanced upon Ba doping due to increased surface hydroxylation [12, 15]. Furthermore, the band observed at 1750 cm⁻¹ is attributed to C=O stretching vibrations, which may arise from surface-adsorbed aldehydes, ketones, or esters, consistent with organic residues remaining after processing or environmental adsorption. The broad band of hydroxyl groups around 3400 cm⁻¹ and 3910 cm⁻¹ is associated with the O–H stretching vibration of hydroxyl groups, attributed to water absorption on the surface of the sample [21].

The wavelength range of 300–400 nm in the TiO₂ absorption spectrum corresponds to Ti species in tetrahedral and octahedral coordination [27, 28]. The absorption edges of anatase TiO₂ nanorods and Ba-doped TiO₂ appear at approximately 343 and 353 nm, respectively. This observation indicates a red shift of ~ 10 nm compared to TiO₂ nanorods. The optical bandgap was identified using Tauc’s plot, as shown in Fig. 2d. The following Eq. (2) [29] is used to estimate the bandgap for a material with a direct bandgap:where A is a constant, E_g is the energy gap, ν is the frequency of the incident radiation, and h is Planck’s constant. Here, n depends on the nature of the transitions, which can have different values; in this case, for an allowed direct transition, n = ½.

Ba doping significantly affects the electronic structure of TiO₂, as evidenced by the optical absorption spectra and bandgap analysis. The UV–Vis absorption data, as shown in Fig. 2c, demonstrate a shift in the absorption edge, indicating modification of the electronic density of states. Ba²⁺ incorporation introduces additional energy levels within the TiO₂ bandgap, as shown in Fig. 2d, effectively increasing the concentration of charge carriers. From Tauc plot analysis, it reveals bandgap narrowing from 3.4 eV (pristine TiO₂) to 3.13 eV (Ba-doped TiO₂), suggesting enhanced charge carrier generation under UV illumination [30, 31].While the introduction of Ba²⁺ ions with an ionic radius of 1.35 Å creates a complex effect on charge carrier mobility, the increased wall thickness and changes in crystal structure might initially suggest reduced mobility due to increased scattering. The bandgap of TiO₂ nanorods has previously been reported as 3.4–3.3 eV [32], 3.2–3.1 eV [33], and 3.0 eV [34]. Likewise, Ba-doped TiO₂ nanoparticles have been synthesized with reported bandgap energies of 3.26 eV [35] and 3.23 eV [21]. Additionally, another investigation showed a reduction in bandgap energy compared to TiO₂ nanoparticles as the weight ratio of Ba increased. The resulting bandgap energies of TiO₂ nanoparticles and 1%, 2%, and 3% Ba-doped TiO₂ were 3.18, 3.11, 3.10, and 3.22 eV, respectively [12]. One report studied the effect of 5% Ba-doped TiO₂ nanorods deposited on Bi₂O₃ film, which slightly enhanced light absorption and narrowed the bandgap (~ 2.6 eV) [36]. Safeen et al. reported a reduction in the bandgap of TiO₂ to 3.11 eV and to 2.5 eV for Co-doped TiO₂, making it an effective absorber in the visible region [19, 37].

M.A. Majeed et al. reported a progressive decrease in the bandgap of TiO₂ nanoparticles from 3.38 to 3.03 eV after doping with Nd and Yb ions [38]. Another study showed that TiO₂ nanotubes hybridized with Ag and graphene oxide exhibited a reduced bandgap and excellent photocatalytic degradation performance [39]. Due to the doping, the bandgap energies shifted as a result of exchange interactions between the s and p electrons of the host Ti atoms and the localized f-electron bands of Nd and Yb [40]. As a result of these interactions, which altered both the valence and conduction band edges, the nanocomposite’s bandgap became narrower. Consequently, the material is a better photocatalyst due to improved visible light absorption and a potential reduction in electron–hole pair recombination [41]. The observation of a red shift implies a reduction in the optical bandgap, which particularly affects the energy of the first excitation [42, 43] and is attributed to quantum confinement and crystal defects [44]. The reduction in bandgap can be explained by defect states and charge transfer interactions, such as exchange interactions between sp–d orbitals [19]. Bandgap narrowing occurs when the conduction band moves downward while the valence shifts upward due to extended defect levels formed at defect sites [45].

The degradation of IC, AZO, and BP dyes was studied to investigate the effect of Ba doping on TiO₂ nanorods. The dye concentration and photocatalyst loading were kept at 2 mg and 10 ppm, respectively. Figure 3 shows the increase in photodegradation ability after TiO₂ was doped with Ba. For IC dye, degradation with TiO₂ showed minimal changes of no more than 4%, whereas Ba–TiO₂ degraded ~ 99% of the IC dye. In the case of AZO, TiO₂ degraded around 20%, whereas Ba–TiO₂ degraded ~ 81%. For BP dye, TiO₂ and Ba–TiO₂ achieved degradation levels of ~ 43% and ~ 97%, respectively. The peak reduction was enhanced and attributed to the bandgap reduction of doped TiO₂, with the incorporation of Ba providing trap state, this phenomenon is explained as follows: Barium doping in TiO₂ nanotubes forms localized energy levels in the bandgap through defect mechanisms, i.e., both substitutional and interstitial defect mechanisms. The large ionic radii mismatch between Ba²⁺ (1.61 Å) and Ti⁴⁺ (0.68 Å) results in large lattice distortion when Ba²⁺ is in interstitial positions or when Ba²⁺ is replacing Ti⁴⁺. These lattice distortions result in a local electronic density of states, which results in forming shallow donor levels situated approximately 0.1–0.3 eV below the bottom of the conduction band [46].where \({\text{Ba}}_{{{\text{Ti}}}}^{\prime \prime }\) denotes Ba⁺² on the Ti site, forming two charges relative to Ti⁺⁴, V^∙∙_o denotes a doubly positively charged oxygen vacancy, and O^x_o is an oxygen atom occupying its regular lattice site. The charge compensation in this case occurs by generating oxygen vacancies, which are well-known to act as donor-type defects in TiO₂ [47]. This reaction shows that oxygen vacancy formation is associated with the emission of two electrons into the lattice. These excess electrons can either fill the shallower donor levels close to the conduction band edge or can localize on the Ti cations, thus lowering Ti⁴⁺ to the form of Ti³⁺ based on the following:where Ti^x_Ti denotes a titanium ion in its regular Ti⁴⁺ state, and 2Ti′_Ti corresponds to a Ti³⁺ ion with a single negative effective charge. This defect chemistry demonstrates that Ba incorporation not only introduces lattice strain and localized donor levels but also promotes oxygen vacancy formation and the reduction of Ti⁴⁺ to Ti³⁺. Together, these effects significantly modify the electronic structure of TiO₂ nanotubes, thereby influencing carrier dynamics and increasing recombination behavior. Table 1 illustrates the degradation percentages of IC, AZO, and BP, determined from the area under the curve for TiO₂ and Ba-TiO₂.

Full size image