Enhancement of dielectric responses and conduction properties of Zn-doped TiO2 for energy storage and photosensitivity applications

This work deals with the physical investigations on Zn-doped TiO2 synthesized via a solvent-controlled non-aqueous sol–gel route. X-ray diffraction analysis indicated that TiO2 particles crystallized in both tetragonal anatase and rutile structures. FTIR analysis confirmed the insertion of Zn in TiO2 network and revealed the presence of surface defects in the prepared powders, as proved by their porous morphologies. UV–visible absorption was performed to provide an insight into the band gap variation in the Zn-doped TiO2 nanopowders as a function of Zn doping. On the other hand, the electrical properties were studied using complex impedance spectroscopy in the frequency range from 40 Hz to 1 MHz at temperature range 480–600 K. The impedance plots are well fitted to a (Rg//CPEg) − (Rgb//CPEgb) equivalent circuit. As well, the correlated barrier hopping model (CBH) was proposed to describe the conduction mechanism. Finally, Photocatalytic activities of the Zn-doped TiO2 nanoparticles were evaluated via the degradation of the rhodamine B (RhB) dye under UV light irradiation. The results showed enhanced performance by Zn doping, compared to the undoped TiO2 nanoparticles.


Introduction
Titanium oxide TiO 2 has been widely investigated thanks to its good thermal, physical, and electrical performances. It is a n-type semiconductor having a wide band gap energy of 3.2 eV at room temperature [1]. Also, it has a relatively high refractive index and dielectric permittivity [2], making it promising for various applications, such as positive electrode in batteries, transparent conducting electrodes in photovoltaic devices, optoelectronic devices, gas sensors, and photocatalytic degradation of organic dyes for water and air purification, self-cleaning surfaces, and antibacterial activity [3][4][5][6][7].
Currently, many complex architectures, especially two-dimensional (2D) nanostructures become seriously investigated. They usually exhibit a more excellent optoelectronic and photovoltaic performance due to their high charge mobility [8], which promote transport of electrons as the sheets have highly active surfaces. This, in turn, reduces the recombination rate and improves the separation capability, as mentioned in earlier reports [9,10] Several physical and chemical methods have been used to prepare TiO 2 nanoparticles and thin films, such as co-precipitation [11,12], polyol method [13], RF sputtering [14], chemical vapor deposition (CVD) [15], pulsed laser deposition [16], and spray pyrolysis [17]. Moreover, many efforts have been conducted to obtain nanopowders with desirable physical and/or chemical properties. Among the preparation techniques, sol-gel is cost-effective. It has many advantages such as homogeneity, lower crystallization temperature, a relative chemical purity of powders and self-limiting surface reactions. Commonly, TiO 2 contains different types of defects, due to oxygen vacancies [18], oxygen interstitial [19], Ti vacancies [20], Ti interstitial [21] and more complex defects [22]. Defects can also be monitored via appropriate substitution doping. Many works reported that small concentrations of doping metals such as Ta, Al, Yb, Nb, In, Sn, and Fe can significantly affect the optoelectrical and dielectric properties of TiO 2 [23][24][25][26][27]. This enhanced the charge storage mechanism by introducing more oxygen in TiO 2 network, which in turn creates additional energy levels in the band gap of TiO 2 . In addition, the surface morphology takes a leading role. Thus, porous, interconnected, conducting materials and materials with high surface to volume ratio such as nanocrystals can offer high specific capacitance value. To date, many researches have been focused on nanostructured TiO 2 (thin films, powders, nanocomposites, …) and it was reported that their morphology plays a crucial role in electric and dielectric properties due to their large surface area [28,29].
In this context, the exploration of the dielectric properties of the material is one of current research topics, as it can predict whether the material is a promising candidate for energy storage application. As well, many studies on photoactivity of TiO 2 oxide have been reported since the early twentieth century.
Indeed, it was reported that UV absorption may produce active oxygen species on TiO 2 surface causing the photo-bleaching of dyes. Recently, ultrafine nanoparticles of TiO 2 have been synthesized. These nanoparticles exhibit a noticeable photodegradation efficiency in visible light of methylene blue (MB) dye and complete photoreduction of Cr 4? ions [30]. In the same line, Mn-Zn co-doped TiO 2 , synthesized by a simple step sono-chemical route without annealing treatment process, showed a relatively high photocatalytic degradation performance of rhodamineB (RhB) dye under visible light [31].
Till now, the design of such nanoarchitecture monolith materials has been considered a longstanding challenge. So, by monitoring experimental conditions, we reach a simple process to synthesize dimensional nanomaterials with variable sizes, wide surface area, porosity, and tunable physicochemical characteristics such as light absorption, optoelectronic enhancement, and dielectric properties with the incorporation of Zn ions.
Although, to the best of our knowledge, those properties of dimensional monoliths have not so far been tackled, a growing interest is being devoted to such prospects, more particularly with complicated process by adding organic materials [32,33] or by one-or two-step hydrothermal method [34,35], coaxial electrospinning technique [36], and polymerization method [37].
Among these various nanostructures currently available, 2D nanosheets structures have attracted considerable attentions, because of their massive interior contact sites and large number of grain boundaries. Despite intriguing, some questions about the role of oxygen vacancies and microstructure on optoelectrical properties and photoactivity still remain unexplored. Herein, besides the synthetic protocol and the physical investigations of Zn-doped TiO 2 nanopowders, a specific emphasis is put on their photocatalytic properties against (RhB) dye under UV illumination.

Material preparation
Undoped and 1 mol%, 2 mol%, and 3 mol% Zndoped TiO 2 were synthesized via the sol-gel process. For that, given amounts of Ti(OPr) 4 and ethanol, in (2:3) ratio, were mixed in a glass container. After the complete mixing, glacial acetic acid was slowly added to the resultant homogeneous yellow solution in order to promote the acid hydrolysis at pH 3. In all obtained samples, a stoichiometric amount of commercial Zinc acetate (ZnC 4 H 6 O 4 , Aldrich) was introduced after 60 min of stirring to obtain Zn-doped TiO 2 . Thus, obtained sols were stirred, capped, and kept at ambient conditions for more than 24 h, in order to complete the gelation step, after which transparent xerogels (monoliths) were obtained. We proceeded to heat treatment of these powders. Firstly, they were dried at 110°C for 12 h, then annealed at 500°C for 5 h to obtain a thermally stable crystalline phase. The gelation time for undoped TiO 2 was 60 min, while for Zn-doped TiO 2 , the gelation needed more much time (about 15 days).

Characterization techniques
First, X-ray diffraction (XRD) patterns were recorded to characterize the phase and crystal structure of nanoparticles using a Phillips powder diffractometer operating with copper Kα radiation source (λ= 1.54056 Å) over a wide range of Bragg angles (10°\2h \ 60°). Also, the structural analysis was carried out using the standard Rietveld method [38]. The morphology of the TiO 2 powders was analyzed by a Hitachi S4100-1scanning electron microscope (SEM). Second, Fourier transformed infrared (FTIR) analysis was carried out using Perkin Elmer spectrometer in the wavenumber range 400-4000 cm −1 . The optical absorption measurements of the powders were recorded using a Shimadzu UV-3101 PC spectrophotometer in the UV-visible range (from 200 to 2000 nm wavelength). On the other hand, the electrical properties of TiO 2 nanoparticles calcined at 500°C were investigated using the impedance spectroscopy [39]. Prior to measurements, the TiO 2 powders were pressed into pellets of 8 mm diameter and 1 mm thickness to form a plate capacitor configuration. Aluminum films of 6 mm diameter were deposited on both sides of the pellet. Afterwards, electrical impedances were measured using a Tega-m3550ALF impedance analyzer operating over a frequency range 40 Hz-1 MHz and at temperature 480-600 K. Finally, the photocatalytic degradation experiments were carried out by monitoring the evolution of the main absorption of RhB aqueous solution (λ max =554 nm) after irradiation with UV light for 10-140 min. The absorbance spectra were recorded by means of Shimadzu UV3100 spectrophotometer, using a high-pressure mercury lamp (300 W) as a light source. Before illumination, the solution was magnetically stirred for 60 min (in dark), in order to ensure an adsorption equilibrium between the sample surface and the organic dye [40]. A detailed description of the photocatalytic reactor was previously reported [41,42]. The RhB dye photodegradation efficiency (η) was calculated for the main absorption of RhB solution as follows [43]: where C is the RhB dye concentration at time t of the reaction and C 0 is the concentration of the initial dye solution (C 0 =30 mg/l), deduced from C/C 0 =A/A 0 (A is the absorbance of the RhB dye solution).

Results and discussions
3.1 Structural analysis Figure 1 shows the XRD patterns of undoped and zinc-doped titanium dioxide nanopowders as a function of Zn content. The diffraction peaks in all samples matched well with the typical anatase (A) and rutile (R) phases of tetragonal TiO 2 . Besides, in 3 mol% Zn-doped TiO 2 sample, a small trace of brookite phase was formed, confirmed by the characteristic (200) peak. All the diffraction peaks agreed with the JCPDS cards (153-0152 for anatase, 900-7532 for rutile and 900-4139 for brookite phases). No additional peaks characteristic of secondary phases such as zinc oxide were detected. Moreover, specific surface area (S a ), was estimated for all the samples using the following equation: where D is the average crystallite size estimated according to Scherrer model. ρ is the density of samples calculated by the following relation [44]: where n=4 for anatase and n=2 for rutile phase, M is molecular weight, N is Avogadro's number and V is the unit cell volume.
The obtained values are listed in Table 1. The density (ρ) of all the samples was found to have almost the same value (about 3.9 g/cm 3 for anatase phase and about 4.25 g/cm 3 for rutile phase), while the specific surface area increased with the increase of doping concentration. This may be attributed to the decrease of the crystallite size. In fact, from Fig. 2, we noticed that Zn-doped samples have larger specific surface area, which is promising for high photocatalytic efficiency. This will be confirmed in the following sections.

SEM observations
The surface morphology of TiO 2 powders was analyzed through SEM investigations. Figure 3a, b displays SEM images for undoped and 1 mol% Zndoped TiO 2 NPs as an example. One can clearly notice changes in shape and size of TiO 2 nanoparticles with doping. The SEM micrographs revealed the presence of a small irregular spherical and slightly elongated morphology of the undoped TiO 2 particles with an average particle size in the range of 140 nm. However, Zn-doped nanopowders exhibit randomly distributed 2D nanosheets with a nanometric thickness. Furthermore, the porous texture of these samples was not uniform and it is obviously doping dependent. It is clear that both macropores and mesopores were included and interconnected. This porous morphology would lead to an enhancement in photocatalytic activities of Zn-doped TiO 2 nanopowders.

FTIR spectroscopy
FTIR spectra of all samples are displayed in Fig. 4. The broad band located around 3450 cm −1 was attributed to the stretching vibration mode (ν(OH)) originated from the adsorbed water [45]. The later was confirmed by the band around 1622 cm −1 which attributed to the bending vibrations of H-O-H (δ(OH)) [46]. Indeed, the porous surface morphology proved by SEM images facilitate the water adsorption at the surface of TiO 2 nanoparticles. Furthermore, the bands spread in the frequency domain between $ 500 and $ 700 cm −1 and are related to the presence of metal-oxygen bonds (Ti-O-Ti), (Ti-O, Zn-), and (Ti-O-Zn) [47]. In addition, we noticed a change in the vibrational band shape with doping mainly for the sample with 3% Zn doping. This can be related to the disorder created by the incorporation of Zn atoms in TiO 2 network.

UV-visible analysis
In order to investigate the changes in optical transitions of TiO 2 NPs due to Zn doping, UV-visible absorption measurement was performed. The absorption spectra of the samples (Fig. 5a) were computed from the diffuse reflectance spectra using the Kubelka-Munk equation. The absorption at around 285 nm corresponds to the electronic transition from 2p state of oxygen atoms in the valence band, to the 3d state of Ti in the conduction band [48]. Moreover, from this figure, it is noted that the UV light absorption increases with increasing Zn content, which may be due to sp-d exchange interactions between the band electrons and the localized d-electrons of Zn 2? ions.
The band gap energies E g of the Zn-doped TiO 2-NPs were estimated from the Tauc's plots (Fig. 5b) according to the following formula [49]: where α is the absorption coefficient given by α= 2.303/A.d, with A is the absorbance and d is the thickness of the sample, B is a constant and hm is the photon energy. The exponent n value depends on the nature of the electronic transition. It is equal to 1/2 for TiO 2 as it is a direct band gap semiconductor [50]. The optical band gap was obtained by extrapolating the linear part of the curve to the hm axis (Fig. 5b).
Results showed a slight red-shift of the bandgap of Zn-doped TiO 2 samples. It shifted from 3.37 eV for undoped sample to 3.32 eV for 3 mol% Zn-doped one (insert of Fig. 5b)

Impedance analysis
The Nyquist plots (− Z″ vs Z′) of TiO 2 samples as a function of temperature are shown in Fig. 6a-d. At selected temperatures ranging from 480 to 600 K, the figures exhibit deformed semicircles. They indicated the existence of non-Debye relaxation phenomenon with distribution of relaxation times. Non-Debye-like behavior of individual electro-active regions in the material can be modeled by a variety of circuit elements. The more common model is formed by combinations of resistance (R) and constant phase element (CPE). The CPE component was proposed by Abram et al. [52]. Its impedance is expressed as follows: where j is the imaginary unit (j 2 =−1), Q is a constant, and α is a dimensionless parameter (0\α\1). It characterizes the degree of deviation from ideal Debye-like illustration. The experimental data of Fig. 6a-d are fitted using the Z-view software. So, the best fit of experimental data was obtained with an equivalent circuit that consists of two parallel combinations of resistance and CPE elements (R g //CPE g in serial with R gb // CPE gb ) as presented in the inset of this figure. The equivalent circuit parameters estimated from the best fit of experimental data are gathered in Table 2a-d. These parallel combinations described the volume (grains) and interfacial (grain boundaries) effects. Therefore, the overlapped semicircles confirm the dispersal nature of the relaxations and the strong heterogenic nature of materials [53]. This result can be related to the multiphase nature of TiO 2 NPS (anatase and rutile phases), as detailed in XRD analysis section.
The a values are close to the unit, which implies the capacitive behavior of the Zn-doped TiO 2 .
When increasing temperature, the low-frequency intersection point of the diagrams with the x-axis shifts towards the origin of the diagram indicating the decrease in the resistivity of the samples, assisted by the increase of the mobility of charge carriers and thus the conduction process [54]. In addition, the presence of free charge carriers and impurities at the grain boundaries can influence the electrical conductivity. At high frequencies, the second semicircle is very weak, which indicates the dominance of the grain boundary contributions in conductivity [55]. Figure 7a-d shows the variation of real part (Z′) of impedance as the function of frequency at different temperatures for the all TiO 2 samples. It is found that the magnitude of Z′ decreases smoothly with the  [56]. Furthermore, the effect of doping on the resistivity can be highlighted through the analysis of the values of Z′ at low frequencies at a given temperature. It is obvious that it increases by factor of about ten in Zn-doped samples compared to undoped TiO 2 .
In order to understand the charge motion mechanism and the relaxation process as a function of temperature, the frequency dependence of imaginary part (Z″) of the complex impedance was plotted at different temperatures (Fig. 8a-d). The (Z″) spectra clearly show the relaxation peak which the frequency and intensity are strongly temperature dependent.
Indeed, the relaxation peak shifts towards higher frequencies with increasing temperature, whereas (Z ″) max value decreases. This indicates a thermally activated dielectric relaxation process.

Dielectric analysis
The dielectric losses can be deduced from impedance components as follows: Figure 9a-d displays the frequency dependence of the imaginary part of dielectric constant. The dielectric losses rise sharply at low frequencies, reach high values at high temperatures, and decreases as frequency increases. The decrease in high frequencies is attributed to ionic inertia hindering the dipole to follow the frequency response. Thus, the dielectric losses approach a zero value. Furthermore, the charge carriers are trapped by defects in grain boundaries, forming dipole moments. Hence, the hopping of charge carriers between defect centers (Titanium interstitials and oxygen vacancies) at the grain boundaries constitutes the dielectric polarization at lower frequencies.
In the explored frequency range, ε″(f) plots do not show any loss peaks, due to the conduction phenomena or/and electrode polarization which may obscure a dielectric relaxation [57][58][59]. To overcome this difficulty, we used the electric modulus formalism, in which the electric modulus is defined as follows:

Electrical modulus analysis
The complex electric modulus representation M* (ω) developed by Provenzano et al. [60] is one of the methods to analyze the space charge relaxation phenomena which is not clearly detected, when the complex permittivity formalism is used. The electric modulus M* is given by the following equation: where M′, M″ are the real and the imaginary parts of the electric modulus and e 00 is expressed as follows:  r 0 is the static conductivity, α is a parameter between 0 and 1, ε s and ε ∞ are, respectively, the static and infinite frequency dielectric constants, ω is the angular frequency (ω=2πf), τ is the relaxation time, and ε 0 =8.8510 -12 F/m is the permittivity of vacuum.
As it can be seen, the electric modulus can greatly decline the electrode polarization and conduction effects, which appear to obscure relaxation in the permittivity presentation [61]. Figure 10 shows the variations of imaginary part of electric modulus M″ with frequency at different temperatures. It exhibits a relaxation peak that shift toward the higher frequency side with an increase in temperature. This indicates a thermally activated behavior of the relaxation.
The peak is assigned to the transition from long range to short range mobility of charge carriers with the increase in frequency [62]. In fact, it is associated to the dipolar relaxation of the TiO 2 grains [63,64]. This peak is usually related to oxygen interstitial O i , oxygen vacancy V O , and titanium interstitial Ti i . These observations are in good agreement with the results obtained by Ben Taher et al. [65]. Besides, the asymmetry in the relaxation peak indicates distribution of the relaxation time, and hence the relaxation in the material is of non-Debye type. Furthermore, it is obvious that Zn-doped TiO 2 exhibit more dispersive relaxation behavior vs temperature, compared to the undoped TiO 2 (Fig. 10), due to structural disorder induced by insertion of Zn ions into TiO 2 network.

Conductivity study
The ac conductivity (σ) of the material can be calculated using the following relation: Fig. 9 Variation of the imaginary part of dielectric constant of a undoped, b 1 mol% Zn, c 2 mol% Zn, and d 3 mol% Zn-doped TiO 2 The (σ) plots of undoped and Zn-doped TiO 2 samples are displayed in Fig. 11a-d. The conductivity is almost constant at low frequencies, then increases with the increase of frequency. At low frequencies, the frequency-independent values of conductivity correspond to the dc conductivity. Moreover, an increase in σ dc is noticed when increasing temperature. In fact, at low frequencies, few charge carriers can tunnel through the potential barrier at the grain boundaries, related with low conductivity. More charge carriers tunneling let to enhance the conductivity at higher temperatures and frequencies.
Beyond a definite value of frequency, the charge carriers get sufficient energy to overcome the potential barrier and hence a rapid increase in conductivity occurs at higher frequencies [66]. Furthermore, it is clearly seen from Fig. 11e that dc conductivity sharply decreases in the doped samples from about 5.10 -9 Ω −1 cm −1 in the undoped TiO 2 to 7.10 -10 Ω −1 cm −1 in 3% Zn-doped TiO 2 . This can be related to Zn 2? incorporated into a TiO 2 lattice. Indeed, as their ionic radius is almost equal (r Ti 4? =0.068 nm and r Zn 2? = 0.074 nm), Zn can easily substitute Ti ions. Therefore, Zn ions cause defects like titanium interstitials and oxygen vacancies in the host TiO 2 system, which act as deep charge traps [67], thus reducing the charge mobility and consequently decreasing the conductivity. Furthermore, several reports [68,69] approved that the dopants confined in interstitial sites can act as trapping or recombination centers for excited electrons and holes [70]. The ac conductivity can be described by the following equation: known as Jonscher's universal power law [71], where A is a constant and s is a temperature and frequency dependent exponent (0\s\1). The exponent "s" characterizes the degree of interactions between charge carriers and their environment. The second term of Eq. (11) (Aω s ) describes the frequency dependence of ac conductivity and characterizes all dispersion phenomena; besides "s" gives information about charge transport. At low frequencies, the temperature dependence of dc conductivity is given as follows: where σ 0 is a pre-exponential factor which includes the charge carrier mobility and density of states, E a is the activation energy associated with dc conductivity, K B is Boltzmann constant, and T is the absolute temperature. Figure 12 displays the Arrhenius plots of the dc conductivity (Eq. (12)), in which the slopes permit the evaluation of the activation energy for the all TiO 2 Fig. 11 Frequency dependence of the AC conductivity for: a undoped, b 1% Zn-doped TiO 2 , c 2% Zn-doped TiO 2 , and d 3% Zn-doped TiO 2 , as a function of temperature, and e the dc conductivity at 480 K, as well as its activation energy vs Zn doping samples. So, the activation energy was found to increase from 0.37 eV for undoped sample to 1.09 eV for 3% Zn-doped TiO 2 , as seen in Fig. 11e, in accordance with the decrease of the dc conductivity. This can be explained by a reduction in the charge carrier's mobility in doped TiO 2 due to the creation of deep level defects induced by the insertion of Zn in TiO 2 structure. It is reported that the major defects that arise in TiO 2 are titanium interstitials (Ti i ) and oxygen vacancies (V O ) [72]. According to the literature [73], oxygen vacancies form two energy levels at 0.75 eV and 1.18 eV below the conduction band, corresponding, respectively, to singly and doubly  Generally, the variation of the exponent "s" (Eq. 11) is related to the conduction mechanism. This exponent "s" was calculated from the linear parts of (σ=f(f)) at high frequencies (Fig. 12). The decrease of "s" with temperature ( Fig. 13) suggests that the correlated barrier hopping (CBH) mechanism governs the ac conduction [74][75][76]. This reflects the hopping of charge carriers between two sites over a barrier separating them.

Photocatalytic tests
The photocatalytic activities of undoped and Zndoped TiO 2 were evaluated using the degradation of the standard organic RhB dye, under UV irradiation. For that the absorption of RhB dye solution with TiO 2 NPs was measured and typical absorption spectra are plotted over the time in Fig. 14a. It is obvious that TiO 2 NPs react well leading to a decrease of the main absorption band of RhB solution over the time. Indeed, RhB is a phenothiazine dye in which the chromophore part contains amino and thiocarbonyl bonds and exhibits a strong absorbance in the visible region (554 nm). However, the vanishing of this band reveals that the dye is degraded. Also, the photocatalytic activity exhibits a monotonous improvement by increasing Zn content and reaches the highest efficiency with 3% Zn doping (Fig. 14b). Moreover, Fig. 14c clearly shows that all TiO 2 samples have high photocatalytic performances against RhB after 140 min under UV irradiation, which was clearly improved by 3% Zn doping. In fact, for this sample the photodegradation efficiency reached 95% while it was of 82% for the undoped TiO 2 . This fact may be related to the small average crystallite size (Table 1) and a large specific area (Fig. 2) of the 3% Zn-doped  sample. The latest can help diffusion of the reactive molecules to the active sites, as the samples have porous morphology, which leads to an increase in the charge of the dye molecules and thus activates dye photodegradation.
Furthermore, from Fig. 14b, one can notice a difference in photodegradation kinetics of RhB using undoped and Zn-doped TiO 2 nanopowders. Indeed, the kinetic rate is an important parameter in photodegradation studies because it can calculate the rate at which pollutant is removed from the aqueous solution. So, it is reasonable to quantify the photodegradation kinetics of the RhB dye. In all cases, the degradation kinetics are well adjusted to pseudo-first-order process, which can be evaluated using Langmuir and Hishelwood formula [77]: where K app is the apparent rate constant of the firstorder reaction (min −1 ). A plot of (Eq. (13)) is given in Fig. 14d. It exhibits a straight line in which the slope equals the apparent first-order rate constant K app . The obtained values are plotted versus Zn doping content and presented in the insert of Fig. 14d. As shown, the 3% Zn-doped TiO 2 sample has the highest rate constant, which degrades the RhB dye faster. This proves the enhancement of photocatalytic performance of TiO 2 by Zn doping.
Following the above discussion, the proposed mechanism of dye degradation is shown in Fig. 14e. When irradiated by UV light, valence electrons of TiO 2 were excited to the conduction band (CB), resulting on generation of electron-hole pairs. The electrons in the CB react with dissolved oxygen and produce reactive oxygen species, which react with water molecules to form hydroxyl radicals (·OH). The dye is then degraded by the hydroxyl groups. While holes at the valence band of TiO 2 react with absorbed water or hydroxyl groups to form surface hydroxyl radicals which then degrade dye. The holes can oxidize the dye molecules directly. The main reactions are shown below [78]: TiO 2 h þ ð Þ þ H 2 O=OH À ! TiO 2 þ OH Á ð16Þ the electron in the CB from reacting with molecular.

Conclusion
In summary, undoped and Zn-doped TiO 2 has been successfully prepared by sol-gel method. All synthesized nanopowders are formed by a mixture of rutile and anatase crystalline phases with tetragonal structure. Zn doping highly modified the powder morphology and leads to the formation of nanosheets. The specific surface area increased with the increase of doping concentration. The optical band gap decreases with the increase of Zn content, due to defects creation and the decrease of the average crystallite size. Dielectric properties were analyzed using complex impedance for the sample at various temperatures. The impedance plots are well fitted to a (R g //CPE g ) − (R gb //CPE gb ) equivalent electrical circuit. The equivalent circuit parameters estimated from theoretical fit proved the enhancement of capacitive behavior of the Zn-doped TiO 2. The frequency dependence of the conductivity is interpreted in terms of Jonscher's law. As well, the conduction mechanism was dominated by the correlated barrier hopping model (CBH). Finally, photocatalytic activities of the Zn-doped TiO 2 nanoparticles were evaluated via the degradation of RhB dye aqueous solution under UV light irradiation. Also, the results showed that Zn-doped TiO 2 nanopowders exhibit a noticeable improvement of photodegradation of RhB dye typically for 3% mol Zn doping, compared to the undoped TiO 2 nanoparticles. Its photodegradation efficiency reaches 95% after irradiation time of 140 min. The kinetic study revealed that 3% mol Zndoped TiO 2 sample is a promising photocatalyst for the degradation of RhB pollutant. Further works are planned to test these samples in other sensitivity applications such as gas and bio-sensors.