Time-Dependent Hydrothermal Synthesis of TiO₂ in the Presence of Zn²⁺: Effects on Photoconductivity

Abstract

Titanium dioxide nanoparticles were synthesized via hydrothermal treatment of tetrabutyl titanate in sulfuric acid, with controlled reaction times (10 h and 24 h) and zinc sulfate as a modifier. XRD confirmed exclusive formation of the anatase phase, with longer reaction times promoting crystallite growth. SEM and BET analyses showed that introducing Zn during synthesis suppressed agglomeration, decreased the particle size, and modified porosity while maintaining the mesoporous nature of all samples. UV–Vis diffuse reflectance spectroscopy showed a band gap near 3.2 eV, which was unaffected by Zn content or morphology. Photoconductivity studies showed a several-orders-of-magnitude increase in conductivity under vacuum conditions, especially in samples heat-treated for 24 h, due to the generation of oxygen vacancies and ${\mathrm{Ti}}^{3+}$ states that prolong the carrier lifetime. In particular, the TS24Z8 sample exhibited a photoconductivity enhancement of five orders of magnitude relative to its dark conductivity and nearly 30 times higher than that of the commercial P25 benchmark. In contrast, in air, photoconductivity remained low because of strong surface recombination with adsorbed oxygen. These results emphasize the critical influence of hydrothermal duration and zinc incorporation on the defect structure and electronic response of ${\mathrm{TiO}}_2$, offering insights for improved photocatalytic and optoelectronic applications.

1. Introduction

Titanium dioxide (${\mathrm{TiO}}_2$) is one of the most extensively studied semiconductor materials due to its excellent chemical stability, non-toxicity, and unique electronic and optical properties. These characteristics make it highly suitable for applications such as photocatalysis, solar energy conversion, and environmental remediation [1]. However, the practical performance of pristine ${\mathrm{TiO}}_2$ is often limited by its wide band gap, which restricts absorption to the ultraviolet region, and by the rapid recombination of photogenerated electron–hole pairs [2]. As a result, strategies to enhance its visible-light activity and charge separation efficiency have become a major focus in materials science research [3].

Beyond its fundamental scientific significance, ${\mathrm{TiO}}_2$ has established itself as one of the most versatile functional materials, with applications spanning both environmental and energy-related fields. In environmental remediation, ${\mathrm{TiO}}_2$-based photocatalysts are widely utilized for the degradation of organic contaminants in wastewater treatment [4,5,6], the removal of volatile pollutants in air purification [7,8,9,10], and as self-cleaning and anti-fogging coatings on glass [11,12,13], ceramics [14], and construction surfaces [15]. The strong oxidative capability of ${\mathrm{TiO}}_2$ under irradiation has also enabled its use in antimicrobial and antibacterial coatings [16,17,18], with growing interest in the healthcare and food packaging industries [19]. In the field of energy conversion [20] and storage, ${\mathrm{TiO}}_2$ has been incorporated as a photoanode material in dye-sensitized and perovskite solar cells [21,22,23], as an electrode material in lithium-ion and sodium-ion batteries [24,25,26], and as a photoelectrode in water-splitting systems for sustainable hydrogen production [27,28,29]. More recently, ${\mathrm{TiO}}_2$ nanostructures have been explored for applications in gas sensors [30], supercapacitors [31], and even biomedical devices such as drug delivery systems [32]. These diverse applications demonstrate the wide technological impact of ${\mathrm{TiO}}_2$ while at the same time highlighting the urgency of addressing its intrinsic drawbacks in order to optimize efficiency and extend its practical utility in real-world systems.

To obtain ${\mathrm{TiO}}_2$ materials with desirable properties, several synthesis techniques have been developed and widely studied [33,34]. Conventional methods such as sol–gel processing, precipitation, and chemical vapor deposition (CVD) allow for the preparation of ${\mathrm{TiO}}_2$ with controlled particle sizes and phase compositions, while approaches like hydrothermal and solvothermal synthesis provide additional advantages in adjusting the morphology, crystallinity, and surface area. More advanced strategies, including atomic layer deposition (ALD) and electrochemical methods, have been employed for fabricating ${\mathrm{TiO}}_2$ thin films and nanostructures with precise dimensional control. Each of these routes offers distinct benefits and limitations, and the choice of synthesis method often determines the resulting photocatalytic and electronic performance of ${\mathrm{TiO}}_2$-based materials. Rajkumar and Arunachalam [35] synthesized Lanthanum-doped ${\mathrm{TiO}}_2$ using the co-precipitation method and annealed it at 600 °C. Structural and morphological analyses confirm improved crystallinity and needle-like particle formation upon doping and annealing. Photoconductivity studies in air show that ${\mathrm{LaTiO}}_2$ exhibits a photosensitivity of 2.40, compared to 0.28 for pure ${\mathrm{TiO}}_2$, corresponding to an 8.6-fold enhancement, which is attributed to reduced recombination centers and prolonged carrier lifetimes due to effective photodesorption of surface oxygen molecules. In this work of ours [36], ${\mathrm{TiO}}_2$/ZnO nanocomposite thin films were synthesized via the sol–gel method, and their photoconductivity was studied at 300 K. The composites exhibited significantly reduced photoconductivity compared to the pure oxides, with ${\mathrm{Ti}}_{50}$${\mathrm{Zn}}_{50}$O showing ${\mathsf{\sigma }}_{\mathrm{p}}$ around 2.60 × ${10}^{-8} {\Omega }^{-1}{\mathrm{m}}^{-1}$ in vacuum and 3.8 × ${10}^{-10} {\Omega }^{-1}{\mathrm{m}}^{-1}$ in air, indicating enhanced electron–hole recombination. Fukuhara et al. [37] synthesized fine ${\mathrm{TiO}}_2$ powders via chemical vapor deposition (CVD) using titanium tetraisopropoxide and studied their photocatalytic activity. ${\mathrm{TiO}}_2$ produced at 200 °C was amorphous and inactive, while higher temperatures (up to 500 °C) yielded crystalline particles with improved activity despite the reduced surface area. Calcination at 550 °C further enhanced the performance of low-temperature powders by increasing crystallinity. Notably, although the 500 °C ${\mathrm{TiO}}_2$ had a smaller surface area than commercial JRC-TIO-4, its activity per unit surface area was higher, demonstrating that controlled CVD synthesis can yield more efficient photocatalysts. In this work [38], we studied Zn-modified ${\mathrm{TiO}}_2$/ZnO nanocomposites prepared via a one-step hydrothermal/solvothermal method. Zn incorporation promoted anatase crystallization, reduced aggregation, and slightly modified the band gap. The 8% Zn sample (TZ8) showed the highest photoconductivity (${\mathsf{\sigma }}_{\mathrm{p}}$ = 5.80 × ${10}^{-4} {\Omega }^{-1}{\mathrm{m}}^{-1}$) in vacuum, 46% higher than ${\mathrm{TiO}}_2$ and 65% higher than P25, corresponding to enhanced photocatalytic NOx degradation under UV light. Sasinska et al. [39] investigated the effect of hydrogen plasma treatment on ${\mathrm{TiO}}_2$ thin films grown by atomic layer deposition to enhance photoanode performance. Plasma-treated ${\mathrm{TiO}}_2$ showed improved visible-light absorption, likely due to surface reduction, with XPS confirming the formation of ${\mathrm{Ti}}^{3+}$ states. Photocurrent densities increased eightfold (1 ma/cm² at 0.8 V vs. NHE) compared to untreated ${\mathrm{TiO}}_2$ (0.12 mA/cm²); this is attributed to localized mid-gap states that improve separation and transport of photoexcited charge carriers under UV illumination. Kamble et al. [40] prepared amorphous ${\mathrm{TiO}}_2$ thin films on FTO glass via low-cost chemical bath deposition. FE-SEM revealed creamy, high-surface-area morphology, and UV–Vis/PL studies showed band gaps of 2.84–3.43 eV. Photoelectrochemical measurements indicated a 3.24% conversion efficiency with a 0.43 fill factor, demonstrating the films potential for solar cell applications.

In this work, ${\mathrm{TiO}}_2$ nanoparticles were synthesized via hydrothermal treatment of tetrabutyl titanate in sulfuric acid, with zinc sulfate as a modifier and reaction times of 10 h and 24 h. XRD confirmed pure anatase formation, with longer durations promoting crystallite growth. SEM and BET analyses showed that Zn incorporation suppressed agglomeration, reduced the particle size, and modified porosity while maintaining mesoporous characteristics. UV–Vis spectroscopy indicated a stable band gap of approximately 3.2 eV, independent of Zn content. Photoconductivity measurements revealed a strong enhancement under vacuum conditions, particularly for the samples heat-treated for 24 h, due to oxygen vacancies and ${\mathrm{Ti}}^{3+}$ states that prolong the carrier lifetime, while in air, recombination with adsorbed oxygen limited conductivity. This work demonstrates that the combined influence of ${\mathrm{Zn}}^{3+}$ presence during hydrothermal synthesis and reaction duration plays a key role in controlling the defect structure and electronic properties of ${\mathrm{TiO}}_2$. The transient photoconductivity results clearly show that the extended reaction time in the presence of Zn promotes the formation of oxygen vacancies and ${\mathrm{Ti}}^{3+}$ centers, which markedly enhance the photoconductive response compared to commercial ${\mathrm{TiO}}_2$ (P25). This Zn/time synergy provides a simple and controllable route to improve ${\mathrm{TiO}}_2$ performance through self-doping induced by oxygen vacancies, without introducing additional dopants. These findings offer useful insights for the rational design of defect-engineered ${\mathrm{TiO}}_2$ materials for photocatalytic and photoelectronic applications.

2. Materials and Methods

During the preparation of titanium dioxide, the raw materials and reagents listed in

Table 1. The chemical substances, with their abbreviations, sources, and properties, are listed here.

Reagent/Material	Abbreviation	Source	Purity
Titanium butoxide (${\mathrm{C}}_{16}$${\mathrm{H}}_{36}$${\mathrm{O}}_4$Ti)	TBT	Merck (Darmstadt, Germany)	97%
Sulfuric acid (${\mathrm{H}}_2$${\mathrm{SO}}_4$)	SA	Fluka (Buchs, Switzerland)	95%
Zinc sulfate heptahydrate (${\mathrm{ZnSO}}_4·$7${\mathrm{H}}_2$O)	ZS	Sigma-Aldrich (Darmstadt, Germany)	99%

were used. Titanium alkoxide, specifically tetrabutyl titanate (TBT), served as the precursor. Deionized water was employed for hydrolysis, while ${\mathrm{Zn}}^{2+}$ ions were introduced as modifiers. The acidic environment provided by sulfuric acid was employed to control the phase formation of ${\mathrm{TiO}}_2$. Figure 1 schematically illustrates the stages involved in the synthesis of the titanium dioxide powder.

For the synthesis step, 5.50 mL of ${\mathrm{H}}_2$${\mathrm{SO}}_4$ was added to 100 mL of deionized water to prepare a 1 M sulfuric acid solution. Then, 3.32 mL of TBT was added dropwise, keeping the solution transparent. In some syntheses, specified amounts of ${\mathrm{ZnSO}}_4·$7${\mathrm{H}}_2$O were also incorporated. The solution was stirred for approximately 30 min and then transferred into Teflon-lined autoclaves that were filled to 80% of their volume. After sealing the autoclaves in steel containers, they were placed in an oven at 180 °C for either 10 or 24 h. Following the reaction, the autoclaves were removed from the oven and allowed to cool to room temperature. The mixture was collected, and the precipitate was separated from the supernatant by centrifugation. The precipitate was washed several times with deionized water until the sulfuric acid was completely removed and the pH reached neutrality. Finally, the washed sample was placed in a glass dish and dried in an oven at 60 °C overnight to remove residual moisture.

Table 2. Compositions of the samples along with their respective reaction times.

Samples	ZS (g)	Abbreviation	Reaction Time (h)
${\mathrm{TiO}}_2$ 10 h	0	TS10	10
${\mathrm{TiO}}_2$ 10 h 4% ZS	0.120	TS10Z4	10
${\mathrm{TiO}}_2$ 10 h 8% ZS	0.183	TS10Z8	10
${\mathrm{TiO}}_2$ 24 h	0	TS24	24
${\mathrm{TiO}}_2$ 24 h 4% ZS	0.120	TS24Z4	24
${\mathrm{TiO}}_2$ 24 h 8% ZS	0.183	TS24Z8	24

summarizes the compositions of the samples along with their respective reaction times. Here we would like to add that the Zn:Ti molar ratio was selected based on established practices, where low modifier concentrations (<10%) effectively tune both the electronic and structural properties of metal oxides [41,42]. The use of ${\mathrm{ZnSO}}_4$ as a precursor is particularly advantageous, as sulfate ions (${\mathrm{SO}}_4^{2-}$) are known to favor anatase phase formation during ${\mathrm{TiO}}_2$ crystallization [43].

The crystalline phases of the prepared samples were analyzed using the Siemens D500 (Karlsruhe, Germany) X-ray diffractometer, with diffraction patterns recorded within the 2$\mathsf{\theta }$ range of 20–70°, employing a step size of 0.03° and a counting time of 4 s per step. The surface morphology and nanostructural features were investigated by scanning electron microscopy (SEM) using an FEI Inspect instrument (Eindhoven, The Netherlands), operating with a tungsten filament at 25 kV and utilizing a secondary electron detector for image acquisition. Measurements were taken manually for a statistically significant number of particles to ensure representative sampling. Textural properties, including porosity and the surface area, were determined through nitrogen adsorption–desorption analysis carried out at liquid-nitrogen temperature using a Quantachrome Autosorb-iQ system (Graz, Austria). The specific surface area was calculated by the Brunauer–Emmett–Teller (BET) method, while the pore size distribution was obtained using the Barrett–Joyner–Halenda (BJH) approach. Finally, the optical characteristics were assessed by UV–Vis diffuse reflectance spectroscopy with a Shimadzu UV-2100 (Kyoto, Japan) spectrophotometer equipped with an integrating sphere, employing ${\mathrm{BaSO}}_4$ as the non-absorbing reference material.

Photoconductivity measurements were performed in both vacuum and air at room temperature (300 K). For electrical characterization, ${\mathrm{TiO}}_2$ powders were compacted into pellets under a pressure of 30 bar, yielding samples with thicknesses of 0.5–0.9 mm and a diameter of 10 mm. Silver electrodes were then deposited by thermal evaporation through a mask to ensure uniform contact geometry, and copper wires were attached using silver paste. The linear current–voltage (I-V) characteristics confirmed the ohmic behavior of the contacts. Each pellet was mounted in an individual sample holder to prevent cross-contamination and maintain stable electrical connections. During testing, a bias voltage of 5 V was applied, as the I-V response remained ohmic across the 1–50 V range. Measurements were performed inside a vacuum cryostat, where photocurrent was monitored under both illumination and dark conditions using a Keithley 6517A electrometer (Cleveland, Ohio, USA), with data collected at 10 s intervals. Illumination was provided by a 100 W xenon lamp equipped with a heat filter (KG3) to minimize thermal effects. The light intensity at the sample surface was adjusted to 240 W/m² and measured using an Ophir PD300-BB radiometer head (Andover, Massachusetts, USA). Each specimen was exposed to two light cycles of 1200 s that interspersed with two dark intervals of equal duration. Prior to testing, the pellets were annealed at 170 °C for 90 min to remove any residual photogenerated effects, then stored overnight in darkness to stabilize at ambient temperature.

3. Results and Discussion

3.1. Crystalline Structures

Figure 2 shows the X-ray diffraction (XRD) patterns of the sample series after 10 and 24 h of reaction times. All samples crystallized exclusively into the anatase phase, in agreement with the reference pattern [JCPDS No. 73-1764], indicating that the acidic environment, particularly the presence of sulfate ions, favors the selective formation of anatase [43]. No peaks corresponding to zinc were detected, likely due to its removal during washing. Using the Scherrer equation, the crystallite sizes were calculated from the three main anatase reflections, and the results are summarized in

Table 3. Crystallite size of the main anatase peaks for the sample series.

Samples	d(101) nm	d(004) nm	d(200) nm
TS10	15.8 ± 0.3	17.9 ± 0.4	15.0 ± 0.5
TS10Z4	16.1 ± 0.3	15.3 ± 0.5	16.1 ± 0.6
TS10Z8	16.6 ± 0.3	18.1 ± 0.4	16.5 ± 0.6
TS24	17.5 ± 0.4	21.7 ± 0.5	17.8 ± 0.3
TS24Z4	19.7 ± 0.5	18.6 ± 0.4	19.3 ± 0.8
TS24Z8	22.2 ± 0.3	24.3 ± 0.7	22.6 ± 0.6

.

For the TS10 series, crystallite sizes range from 15.8 to 18.1 nm, suggesting that the gradual addition of Zn during the synthesis stage results in only a slight increase in size, while in the TS24 series, the progressive addition of ZS during the synthesis stage produces a significant increase, with values between 17.5 and 24.3 nm. These findings indicate that longer reaction times (i.e., extended hydrothermal treatment in the autoclave) promote further crystallite growth [44,45,46]. Moreover, the XRD pattern of the TS10Z4 sample exhibits anisotropic crystallite growth along different crystallographic directions: the broad (004) diffraction corresponds to a reduced crystallite size along the [001] axis, while the intensity of the (101) reflection is related to the [101] direction. This anisotropy is consistent with our observations, where the TS10Z4 sample shows broadening of (004) alongside a narrower (200) peak, suggesting preferential growth along the a-axis and enhanced exposure of 001 facets [47,48,49].

3.2. Morphology and Band Gap Analysis

The morphology of the samples was examined using scanning electron microscopy (SEM). Figure 3 presents SEM images for all samples, illustrating both particle size and surface morphology.

The TS10 sample exhibits an average particle size of approximately 160 nm and shows evidence of agglomeration. With the addition of Zn during the synthesis process (4%Zn and 8%Zn), the average particle size decreases to 90 nm and 77 nm, respectively, accompanied by a slight reduction in agglomerates. The TS24 sample displays a smaller average particle size of about 85 nm, which further decreases to 78 nm and 68 nm upon the addition of 4%Zn and 8%Zn, respectively. This indicates that Zn addition during synthesis reduces the particle size and agglomeration in both series. A comparison between the two series reveals that the TS10 samples generally exhibit larger particle sizes than those of the TS24 series. Moreover, the 24 h samples show fewer agglomerates, suggesting that prolonged hydrothermal treatment in the autoclave (i.e., increased reaction time) promotes more pronounced crystal growth [46].

In order to determine the reflection and absorption regions of the samples in the UV-Vis range, they were analyzed using UV-Vis spectroscopy. The data obtained from the spectra were then used, along with the Kubelka–Munk function [50], to calculate the band gap energy (E_g) of the samples, with the results summarized in

Table 4. Band gap energies of the samples.

Samples	Eg (eV)
TS10	3.13
TS10Z4	3.20
TS10Z8	3.19
TS24	3.20
TS24Z4	3.15
TS24Z8	3.21

. Figure 4a,c show the diffuse reflectance spectra of this series of titania samples, while Figure 4b,d present the plot of [F(R)·E]^1/2 versus photon energy (eV), illustrating the method used to determine E_g. The calculated band gap values indicate that all samples absorb around 3.2 eV, consistent with the anatase band gap. Furthermore, the band gap appears to be independent of the crystallographic structure and morphology of the samples.

3.3. Specific Surface Area

Figure 5 presents the nitrogen adsorption–desorption isotherms and the pore size distribution histograms of the samples, while

Table 5. Summarized surface analysis results.

Samples	S (m2/g)	Vp (cm3/g)	Dpor (nm)	ϕ (%)
TS10	45.33	0.099	7.28	27.26
TS10Z4	62.48	0.376	19.64	58.82
TS10Z8	54.85	0.173	10.65	39.64
TS24	47.63	0.173	13.41	39.68
TS24Z4	46.20	0.362	24.54	57.92
TS24Z8	36.15	0.161	16.17	37.91

summarizes the results obtained from the Brunauer–Emmett–Teller (BET) surface area analysis, including pore diameter, pore volume, and porosity percentage.

The adsorption–desorption isotherms are of type IV [51], which is characteristic of mesoporous materials, with an H3 hysteresis loop. Within the TS10 series, the TS10Z4 sample exhibits the highest specific surface area, consistent with the XRD results, indicating exposure of the 001 facets; this agrees with the literature reporting that anatase crystals with exposed 001 facets possess the highest specific surface area [52,53,54]. In the TS24 series, the highest surface area is observed for the undoped TS24 sample, with only minor variations among the samples. Pore size distribution histograms show that the pore diameters of both the TS10 and TS24 series fall within the range of 7–25 nm, which is typical of mesoporous materials according to the IUPAC classification [52,53,54]. Finally, it is observed that the addition of ZS during synthesis changes the pore diameter across all samples, as it acts as an additional structural modifier.

3.4. Photoconductivity

3.4.1. Vacuum

Figure 6 shows photoconductivity (${\mathsf{\sigma }}_{\mathrm{p}}$) as a function of time at 300 K under vacuum for all samples, along with commercial titania P25, which is used as a reference material for comparison. During the first illumination cycle, the photoconductivity values of this series increase by 2–3 orders of magnitude relative to dark conductivity.

Specifically, at the initial stages of illumination, a rapid rise in photoconductivity is observed, followed by a slower increase over time, with the rate varying among the samples and exhibiting sublinear behavior [55]. In the TS10Z4 and TS10Z8 samples, the recombination rate is more pronounced, leading to lower photoconductivity values. After the first illumination, an initial sharp decrease occurs due to strong carrier recombination, followed by a linear decrease, indicating that the rates of thermal excitation and recombination change over time in the same manner. During the second illumination cycle, photoconductivity reaches higher values than in the first, suggesting that some traps remain filled at the end of the first dark period and contribute during subsequent illumination. Furthermore, the photoconductivity of the TS10 sample exceeds that of P25, while the values for the 4%Zn and 8%Zn samples are slightly lower than those of P25.

Following the transient photoconductivity plots for the TS24 sample series at 300 K under vacuum, as in the previous case, the photoconductivity values exhibit a sharp increase during the initial stages of illumination, followed by a slower rise as the carrier recombination rate becomes significant. In this series, photoconductivity increases by 4–5 orders of magnitude. Under acidic conditions, photoconductivity reaches higher values compared to basic conditions. This behavior is attributed to the formation of anatase in acidic media, whereas in basic media, both the rutile and anatase phases are present [38]. Regarding reaction time, it is observed that a longer duration (24 h compared to 10 h) leads to higher photoconductivity values, as longer treatment favors improved particle formation, which is also supported by the SEM images. Another crucial factor in enhancing photoconductivity, as noted in the introduction, is the presence of oxygen vacancies. These vacancies, or self-doping with ${\mathrm{Ti}}^{3+}$, can create intermediate energy states within the band gap. The mechanism of ${\mathrm{Ti}}^{3+}$ self-doping is based on the trapping of photogenerated electrons, which subsequently reduce ${\mathrm{Ti}}^{4+}$ ions to ${\mathrm{Ti}}^{3+}$ [56], while the photogenerated holes oxidize ${\mathrm{O}}^{2-}$ anions to ${\mathrm{O}}^{-}$ or even molecular ${\mathrm{O}}_2$. The detailed steps of oxygen vacancy formation are presented in Figure 7, which illustrates the ${\mathrm{Ti}}^{3+}$ states within the ${\mathrm{TiO}}_2$ band gap.

It has also been reported that the hydrothermal method [57,58], the solvothermal method [59,60], and the metal reduction method [61,62] are among the most suitable techniques for generating oxygen vacancies or ${\mathrm{Ti}}^{3+}$ self-doping. Wang et al. [57] demonstrated that by adjusting the hydrothermal conditions, the concentration of oxygen vacancies could be controlled. Metallic zinc (formed from ${\mathrm{Zn}}^{2+}$ under hydrothermal conditions) acts as a reductant, directly reducing ${\mathrm{Ti}}^{4+}$ to ${\mathrm{Ti}}^{3+}$ and creating oxygen vacancies. Simultaneously ${\mathrm{Zn}}^2$ incorporation into the ${\mathrm{TiO}}_2$ structure weakens the surrounding Ti-O bonds, significantly lowering the formation energy of oxygen vacancies and facilitating their creation. Both pathways synergistically enhance oxygen vacancy formation during synthesis, while the soluble ${\mathrm{Zn}}^2$ byproducts are subsequently removed during washing, leaving behind the stable self-doped ${\mathrm{TiO}}_2$ [42]. The presence of oxygen vacancies or ${\mathrm{Ti}}^{3+}$ states, either on the surface or within the bulk, can suppress electron–hole recombination. It is well established that the Fermi level (EF) lies very close to the conduction band when the concentration of oxygen vacancies is high. These vacancies represent defect states located near the conduction band (Figure 7), facilitating easier transfer of photogenerated carriers between bands. Since oxygen vacancies carry a positive charge, ${\mathrm{Ti}}^{3+}$ centers are formed near the conduction band. Moreover, ${\mathrm{Ti}}^{3+}$ ions act as hole traps, hindering carrier recombination and thereby extending the lifetime of the charge carriers.

The rise-and-decay behavior of the photoconductivity response is strongly influenced by defect states acting either as trapping sites or recombination centers. Hence, analysis of the photoconductivity transients provides insights into their nature and distribution [63]. The carrier lifetime can be extracted by fitting (Figure 6a inset) a double-exponential function to either the rising or decaying part of the photoconductivity curve, as illustrated in Figure 6. The decay process is described by Equation (1):

$${\mathsf{\sigma }}_{\mathrm{p}}={\mathsf{\sigma }}_{\mathrm{o}}+A_{\mathrm{r}}exp\left({\displaystyle \frac{t}{{\mathsf{\tau }}_{\mathrm{r}}}}\right)+A_{\mathrm{t}}exp(-{\displaystyle \frac{t}{{\mathsf{\tau }}_{\mathrm{t}}}})$$

(1)

where ${\mathsf{\sigma }}_{\mathrm{o}}$ represents the initial photoconductivity (at the instant when illumination is switched off), ${\mathsf{\tau }}_{\mathrm{r}}$ and ${\mathsf{\tau }}_{\mathrm{t}}$ are time constants corresponding to the lifetimes of photogenerated carriers, and $A_{\mathrm{r}}$ and $A_{\mathrm{t}}$ are pre-exponential factors. The constants ${\mathsf{\tau }}_{\mathrm{r}}$ and ${\mathsf{\tau }}_{\mathrm{t}}$ correspond to two different recombination pathways. In the first case, recombination through deep traps is characterized by a short ${\mathsf{\tau }}_{\mathrm{r}}$ (fast process), while in the second case, carriers captured in shallow traps exhibit a longer ${\mathsf{\tau }}_{\mathrm{t}}$ (slow process) [64,65]. The trap depth can be estimated using Equation (2):

$$E=k_{\mathrm{B}}T(lnS-ln({\displaystyle \frac{ln(\frac{{\mathsf{\sigma }}_{\mathrm{o}}}{{\mathsf{\sigma }}_{\mathrm{p}}})}{t}}))$$

(2)

where $k_{\mathrm{B}}$ is the Boltzmann constant and S is the “attempt-to-escape” frequency, which is typically on the order of 10⁹ at room temperature [63,66].

Based on the above discussion, it becomes evident that zinc promotes the formation of oxygen vacancies, as confirmed by the higher photoconductivity values observed for the TS24Z8 sample in the TS24 series. In contrast, such behavior is not observed in the TS10 series, where the addition of Zn during synthesis does not lead to increased photoconductivity. This suggests that preparation conditions, such as longer reaction times, play a crucial role in the generation of oxygen vacancies.

Table 6. Values of dark conductivity (${\mathsf{\sigma }}_{\mathrm{d}}$) before the first illumination and photoconductivity (${\mathsf{\sigma }}_{\mathrm{p}}$) at the end of the first and second illumination cycles, along with the trap energies and time constants under vacuum.

Samples	${\mathsf{\sigma }}_{\mathrm{d}} \Omega {(}^{-1}{\mathrm{m}}^{-1})$	${\mathsf{\sigma }}_{\mathrm{p}1} \Omega {(}^{-1}{\mathrm{m}}^{-1})$	${\mathsf{\sigma }}_{\mathrm{p}2} \Omega {(}^{-1}{\mathrm{m}}^{-1})$	E1 (eV)	E2 (eV)	${\tau }_r$ (s)	${\tau }_t$ (s)
P25	1.53 × ${10}^{-7}$	2.24 × ${10}^{-4}$	3.52 × ${10}^{-4}$	0.671	0.671	43	1646
TS10	1.46 × ${10}^{-7}$	3.55 × ${10}^{-4}$	5.96 × ${10}^{-4}$	0.682	0.614	64	1043
TS10Z4	4.66 × ${10}^{-7}$	4.66 × ${10}^{-5}$	8.43 × ${10}^{-5}$	0.689	0.620	83	1336
TS10Z8	6.22 × ${10}^{-8}$	1.44 × ${10}^{-4}$	2.20 × ${10}^{-4}$	0.688	0.602	41	1285
TS24	1.44 × ${10}^{-8}$	7.78 × ${10}^{-4}$	1.20 × ${10}^{-3}$	0.690	0.604	44	1377
TS24Z4	6.04 × ${10}^{-7}$	3.14 × ${10}^{-3}$	4.71 × ${10}^{-3}$	0.688	0.602	41	1324
TS24Z8	9.98 × ${10}^{-8}$	6.70 × ${10}^{-3}$	9.00 × ${10}^{-3}$	0.692	0.613	41	1012

summarizes the dark conductivity and photoconductivity during the first and second illumination cycles under vacuum, as well as the corresponding trap energies.

3.4.2. Air

Figure 8 presents the transient photoconductivity as a function of time for all samples in air at 300 K, and

Table 7. Values of dark conductivity (${\mathsf{\sigma }}_{\mathrm{d}}$) before the first illumination and photoconductivity (${\mathsf{\sigma }}_{\mathrm{p}}$) at the end of the first illumination.

Samples	${\mathsf{\sigma }}_{\mathrm{d}} \Omega {(}^{-1}{\mathrm{m}}^{-1})$	${\mathsf{\sigma }}_{\mathrm{p}1} \Omega {(}^{-1}{\mathrm{m}}^{-1})$
TS10	1.46 × ${10}^{-6}$	2.65 × ${10}^{-7}$
TS10Z4	2.78 × ${10}^{-7}$	3.40 × ${10}^{-8}$
TS10Z8	2.26 × ${10}^{-7}$	2.55 × ${10}^{-8}$
TS24	3.50 × ${10}^{-8}$	4.46 × ${10}^{-8}$
TS24Z4	3.49 × ${10}^{-7}$	2.34 × ${10}^{-8}$
TS24Z8	2.76 × ${10}^{-7}$	4.89 × ${10}^{-7}$

summarizes the values of dark conductivity and photoconductivity at the end of the first illumination. It is evident that the photoconductivity values are very low compared to vacuum, which is attributed to strong carrier recombination during illumination.

When the sample is exposed to air, oxygen strongly influences its photoconductivity. Upon illumination, electron–hole pairs are generated, and adsorbed oxygen molecules on the material’s surface act as electron scavengers. By capturing the photogenerated electrons, these oxygen molecules reduce carrier lifetimes, keeping the photoconductivity low (${\mathrm{O}}_2$ + ${\mathrm{e}}^{-}\to$ ${\mathrm{O}}_2$⁻). The photogenerated holes migrate to the surface, where they neutralize the adsorbed ${\mathrm{O}}_2$⁻ ions, triggering the photodesorption of oxygen from the metal oxide surface (${\mathrm{O}}_2$⁻ + ${\mathrm{h}}^{+}\to$ ${\mathrm{O}}_2$). Thus, the balance between photoadsorption and photodesorption governs the amount of oxygen adsorbed on the surface [67]. Reports indicate that photoadsorption dominates, leading to reduced photoconductivity in air [68]. Additionally, the mobility of holes in ${\mathrm{TiO}}_2$ is much lower than that of electrons, making electrons the primary charge carriers [69].

Specifically, the TS10 and TS10Z8 samples exhibit maximum photoconductivity within the first seconds of illumination, followed by a decrease due to intense recombination persisting until the end of illumination. In contrast, the TS10Z4 sample shows a slower rate of decrease. After the first illumination, the TS10 and TS10Z8 samples display a sharp drop caused by fast recombination, followed by a significant thermal release that increases photoconductivity. For the TS10Z4 sample, the decrease in photoconductivity during the dark period is initially smaller and continues more gradually until the end of the dark interval. Concerning the TS24 series, A similar behavior is observed for all samples: an initial maximum at the beginning of illumination, followed by strong carrier recombination. Additionally, during the dark period, the TS24Z4 sample exhibits a slight increase in photoconductivity, which is attributed to thermal excitation. Finally, it should be noted that the extremely low values of photoconductivity in air, for both the TS10 and TS24 series, are due to strong electron trapping by oxygen molecules adsorbed on the ${\mathrm{TiO}}_2$ surface. The low photoconductivity of TS10Z4 is likely due to enhanced oxygen adsorption, which is facilitated by its high specific surface area resulting from the presence of ${\mathrm{Zn}}^{2+}$ ions during synthesis. Furthermore, the observed slower decay rate is consistent with these low photoconductivity values, as a lower initial carrier density leads to a slower recombination rate. The use of Zn as a reducing agent during synthesis introduced oxygen vacancies into the ${\mathrm{TiO}}_2$ structure. While these vacancies enhance photoconductivity in vacuum by providing charge carriers, they become preferential adsorption sites for atmospheric oxygen. The resulting surface reaction (${\mathrm{O}}_2$⁻ + ${\mathrm{h}}^{+}\to$ ${\mathrm{O}}_2$) converts these beneficial defects into recombination centers, explaining the suppressed photoconductivity observed in air measurements [57,70].

3.5. Comparison Results

Table 8. Comparison of ${\mathrm{TiO}}_2$-based materials synthesized by different methods, highlighting their targeted properties and corresponding performance changes.

Ref.	Synthesis Method	Material	Property	Change
[36]	Sol-gel	${\mathrm{TiO}}_2$/ZnO film	Photoconductivity	Performance reduced
[37]	Chemical vapor deposition (CVD)	${\mathrm{TiO}}_2$ powder	Surface area	Increased activity
[38]	Hydrothermal/solvothermal	${\mathrm{TiO}}_2$/ZnO powder	Photoconductivity	65% increase in vacuum
[39]	Atomic layer deposition (ALD)	${\mathrm{TiO}}_2$ film	Photocurrent density	×8 increase
[40]	Electrochemical	${\mathrm{TiO}}_2$ film	Conversion efficiency	Improved
CW	Hydrothermal	${\mathrm{TiO}}_2$ powder	Photoconductivity	×30 increase in vacuum

provides a comparison of ${\mathrm{TiO}}_2$-based materials prepared using various synthesis methods, emphasizing their intended properties and the resulting variations in performance. The optoelectronic and catalytic performance of ${\mathrm{TiO}}_2$-based materials is highly dependent on the synthesis route, as even similar compositions can exhibit drastically different behaviors. For instance, ${\mathrm{TiO}}_2$/ZnO films prepared by the sol–gel method showed suppressed photoconductivity [36], whereas the same system synthesized hydrothermally exhibited a 65% improvement under vacuum conditions [38], underscoring the decisive role of synthesis conditions in charge transport. Similarly, CVD-synthesized ${\mathrm{TiO}}_2$ powders achieved improved surface area and catalytic performance [37]. Thin films produced via ALD showed an eightfold increase in photocurrent density [39], and electrochemical methods further enhanced conversion efficiency [40]. Within this context, our hydrothermally synthesized ${\mathrm{TiO}}_2$ powders demonstrated a remarkable 30-fold increase in photoconductivity under vacuum compared to commercial P25, highlighting not only the critical influence of the synthesis strategy but also the potential of defect engineering to tailor ${\mathrm{TiO}}_2$ for advanced photocatalytic and optoelectronic applications.

4. Conclusions

This work demonstrates that the interplay between hydrothermal synthesis duration and zinc incorporation is critical for engineering the defect structure and electronic properties of ${\mathrm{TiO}}_2$. Structural and morphological analyses confirmed that extended hydrothermal treatment promotes crystallite growth, while Zn addition during synthesis suppresses agglomeration, reduces the particle size, and modifies porosity without compromising the mesoporous framework. Optical studies showed that the band gap remains stable near 3.2 eV, independent of reaction conditions.

Photoconductivity investigations revealed striking differences between vacuum and air environments. Under vacuum conditions, samples synthesized for 24 h displayed conductivity enhancements of up to five orders of magnitude relative to dark conditions. In particular, the TS24Z8 sample exhibited nearly 30-fold higher photoconductivity than commercial P25, a performance increase attributed to the synergistic effects of prolonged hydrothermal treatment and Zn-induced oxygen vacancies/${\mathrm{Ti}}^{3+}$ states that prolong carrier lifetimes. The enhanced photoconductivity of the final material provides direct evidence for the presence of these oxygen vacancies, which act as a form of self-doping in the ${\mathrm{TiO}}_2$ lattice. Conversely, in air, strong surface recombination with adsorbed oxygen suppressed photoconductivity across all samples.

These results highlight hydrothermal synthesis combined with controlled Zn incorporation as an effective strategy to tune charge carrier dynamics and defect chemistry in ${\mathrm{TiO}}_2$. The approach provides a pathway toward designing high-performance photocatalysts and optoelectronic devices, where precise defect engineering is essential for achieving superior activity and stability.