Issue
Eur. Phys. J. Appl. Phys.
Volume 90, Number 3, June 2020
Disordered Semiconductors: Physics and Applications
Article Number 30102
Number of page(s) 9
Section Semiconductors and Devices
DOI https://doi.org/10.1051/epjap/2020190267
Published online 07 July 2020

© EDP Sciences, 2020

1 Introduction

Recently, TiO2 nanotubes (nt-TiO2) have aroused the interest of the scientific community due to their high photocatalytic activity, band gap and volume ratio; in addition to their good biocompatibility and low cost of synthesis [1]. nt-TiO2 are unidimensional nanostructures that are normally synthesized by a template of AlO3, hydrothermal process and electrochemical anodization [2]. Anodization is the most common method for the synthesis of nanotubes since it results in highly ordered and organized nanomaterials; it is inexpensive and easy to make [3]. In recent years, TiO2 nanostructures have been studied for different applications such as, gas sensors, solar cells, photonic crystals, and Li-ion batteries [4]. Due to the tube shape, high surface area, self-ordering and defect states, nt-TiO2 have been extensively investigated for gas sensing applications using target gases such as O2, H2, NO2, CO, SO2, NH3, ethanol, etc., because they have demonstrated a good and fast response as well as high reactivity and selectivity for reducing and oxidizing gases [5,6]. Anatase phase is a crystalline structure used to gas sensing since it presents a better chemical and structural stability, and higher catalytic activity than other structures of TiO2 (rutile and brookite). For this reason, thermal treatments are often used to change from amorphous to the crystalline phase since electrochemical anodization produces amorphous nanotubes [7]. On the other hand, gas sensors are used in different areas such as industries, research centers, hospitals and school because they monitor environmental and contaminants gases to prevent accidents [8,9]. Normally, gas sensors are heated from 100 °C to 400 °C to improve performance, but this process increases the cost and size of the device, hence the importance of developing a smaller, and consequently low-cost, gas detector at room temperature. In this sense, new materials have been synthesized to enhance the sensor response, responsivity, response times, selectivity, operating temperature and recovery behavior [9]. For example, Oomman et al. reported an excellent response when hydrogen was tested by nt-TiO2 synthesized with 0.5% hydrofluoric acid in the electrolyte solution; however, they used Pt electrodes and heated the film to increase chemisorption, thus increasing the manufacturing cost [10]. Comert et al. studied nt-TiO2 on n-type silicon for methane gas sensing, nevertheless, a good sensor response was obtained only when the device was heated at 200 °C [11].

It is well known that superficial and transversal morphology are modified by the water content in electrolyte solutions [12]. In this respect, Hazra et al. studied the nt-TiO2 morphologies with different parameters in the chemical composition of electrolyte. They produced disorganized nanotubes that were used for detecting: methanol, ethanol, and propanol, at room temperature and reported that the oxygen vacancies have an important function for gas sensing [13]. Perillo et al. showed that nt-TiO2 with different diameter do not improve the properties of the gas sensor when ammonium and ethanol are measured at room temperature, since their results showed that relative resistance is low [14]. Gas detection in different nanostructures is being widely studied but a study of nitrogen gas sensing with a good response has not been reported yet [15,16].

In this paper, we present a study of the effect of morphology on the optical and electrical properties of nt-TiO2 for gas sensing applications, for this, we synthesized four nanomaterials with several morphologies using different electrolyte solutions. Because a large number of oxygen vacancies change the band gap wide, increase the photo-absorption in the lattice of TiO2 and promote a better sensor response in gas sensors [17,18], we studied the relationship of the morphological, optical, structural properties and defect states as a function of the electrical resistance change of nt-TiO2.

2 Experimental details

2.1 Synthesis of TiO2 nanotubes

Before anodizing, titanium foils of 1.5 cm2 (100 μm in thickness, from American Elements, 99.95% purity) were cleaned with trichloroethylene, acetone and distilled H2O. The nanotubes were prepared with four different electrolyte solutions: 0.255 wt% NH4F with 1 wt%, 3 wt%, 6 wt%, and 9 wt% of deionized water in ethylene glycol. For the electrochemical anodization, we used titanium foils as an anode, and a platinum foil as a cathode with a voltage of 30 V. nt-TiO2 were synthesized by a typical a three-step anodization process during 1, 4 and 20 hours and with 2 detachment step of nanotubes after the first and second anodization. In order to get the best characteristics of nt-TiO2, a thermal treatment at 450 °C during two hours was carried out to change from the amorphous phase to the anatase phase. To study the change in the electrical resistance of our device, a film 20 nm of titanium was deposited on nt-TiO2 by e-beam evaporation, and during metallization was used a metallic mask to make two stripes (1 cm × 0.2 cm) on nanotubes. A schematic representation of the final structure of our device is shown in Figure 1.

thumbnail Fig. 1

Schematic representation of the gas sensor using nt-TiO2.

2.2 Characterization

We used several analytical techniques in order to evaluate the physical, chemical, optical, and electrical properties of nt-TiO2: the morphology and topography were studied and analyzed by SEM (FEI, model SCIOS), the elemental composition as atomic percentage (At%) of oxygen, titanium, fluorine, and nitrogen were measured by energy dispersive X-ray spectroscopy at 10 keV (EDS, EDAX, model APOLLO); the UV-Vis optical properties were measured by a Perkin Elmer spectrophotometer (model Lambda 3B) and the phase identification of TiO2 was determined by X-ray diffraction (XRD, D8 Discover Bruker diffractometer). Besides, a photoluminescence (PL) study to determine the main mechanisms that increase the emission was carried out by fluorescence spectroscopy (Fluoromax-3, Jobin Ybon). The electrical current and resistance were measured by an electrometer Keithley (model 2400) applying a voltage from −1 V to 1 V to bias the sensing device. The gas detection was carried out in a gas sensing chamber operating at 5 psi at room temperature, with a flux of N2 and a mixture of 90% N2 and 10% H2, controlled by mass flow controllers (MFC) from FATHOM technologies GR series (GR116-1-A-PV). The gas sensor was used in two tests, the first (1 a) consisting of three gas flow steps: air, followed by nitrogen and, finally air again, and the second test (1 b) consisting of three steps, too: air, a mixture of 90% nitrogen and 10% hydrogen, and air.

3 Results and discussion

Figure 2 shows images of the superficial morphology of each of the nanotubes films synthesized with four different electrolyte solutions. As we can see, different morphologies were obtained, the wall thickness increases, and both the inner diameter and the control in the organization of the nanostructures decrease as the concentration of water in the electrolyte solutions increases. Also, the nanotubes trend to form random packets without a particular shape on the top of the nanotubes film. On the contrary, a low concentration of water promotes self-organization and self-ordering of the TiO2 nanostructures, causing the wall thickness to be thinner and the circular diameter well defined. Thereby, the nanotubes synthesized with 1 wt% H2O DI (nt-TiO2-1, Fig. 2a) obtained a better ordering and organization (like a honeycomb) than the nanotubes synthesized with a higher content of water in electrolytic solution (nt-TiO2-9, Fig. 2d).

Figure 3 shows SEM images of the cross-section of nanotubes. As we can see, the length of the nanotubes is also affected by the water content. In this instance, the nanomaterials prepared with lower water content (nt-TiO2-1) show a shorter length than those synthesized with higher water content. Furthermore, a higher content of water (nt-TiO2-9) increased the ripples in the nanotubes compared to the samples prepared with lower water content. A summary of the average length for each sample is shown in Table 1, these results are in agreement with other literature reports [19,20]. Additionally, Table 1 shows the elemental composition of the TiO2 films on the titanium substrate after electrochemical anodization. For all samples, the ratio between Ti and O is similar, however, the TiO2 film closest to its stoichiometry is the sample with the lowest content of water (nt-TiO2-1) because it has an approximated ratio of 1:1.7 (Ti ≤ Ox), this ratio is attributed to titanium substrate and incomplete oxidation [21]. Likewise, we can see that the films contain elements of the electrolyte solution such as F, N, and C, as previously reported [21,22].

thumbnail Fig. 2

SEM images of TiO2 nanotubes synthesized by electrochemical anodization with (a) 1 wt%, (b) 3 wt%, (c) 6 wt%, and (d) 9 wt% of deionized water and 0.255 wt% NH4F in ethylene glycol.

thumbnail Fig. 3

SEM images of cross-section of TiO2 nanotubes synthesized by electrochemical anodization with (a) 1 wt%, (b) 3 wt%, (c) 6 wt%, and (d) 9 wt% of deionized water and 0.255 wt% NH4F in ethylene glycol.

Table 1

Length and elemental composition of the TiO2 nanotubes prepared by electrochemical anodization.

3.1 The effect of electrolyte solution on structure of TiO2 nanotubes

To confirm the change from the amorphous a crystalline phase, an XRD study was carried out. Before the thermal treatment, an amorphous phase in the sample nt-TiO2-1 was observed because the sample did not present peaks related to TiO2. However, two titanium peaks at 38° and 40° associated with the planes (002) and (101) were detected, see Figure 4. These peaks of titanium are caused by the substrate [23,24].

Through thermal treatment, the amorphous structure is converted to the anatase crystalline phase, the typical peaks were observed at 25°, 48°, 54°, 55°, and 62.8° corresponding to the planes (101), (200), (105), (211), and (204), respectively [24,25]. For the titanium peaks, we can see that the intensity of planes (002) and (101) increases with the lowest concentration of water in the electrolyte solution (see Fig. 3). When comparing the intensity of the anatase peaks of the nt-TiO2-1, nt-TiO2-3, and nt-TiO2-9 samples, we saw that nt-TiO2-1 showed the highest peak intensity than other samples, which means that the intensity of the anatase peaks decreases as both the organization of nanostructures and the wall thickness increase. The highest crystalline intensity is the plane (101) of nt-TiO2-1 when the nanotubes are more organized as a closed-packed, because a reduction of water content modifies the Ti:Ox ratio and it is closer to the stoichiometry of the TiO2 film, as we can see it in Table 1.

Therefore, we improved the crystalline structure of the nanomaterial by modifying the chemical composition of electrolyte solution, as previously, other studies showed that annealing temperature [26], time [27] and voltage [28] during the anodization process can also improve it.

thumbnail Fig. 4

XRD pattern of TiO2 nanotubes synthesized by electrochemical anodization.

3.2 The photoluminescence study of TiO2 nanotubes

The PL spectra for the four samples are illustrated in Figure 5 when the TiO2 films have amorphous and crystalline structure. Figure 5a shows the PL intensities of nt-TiO2 films after anodization and Figure 5b presents the PL spectra after thermal treatment. Consecutively, Figures 5c through 5f display the deconvolution of PL spectra corresponding to the defect states peaks for each sample.

As we can see, the four samples have the same characteristic curve (Fig. 5a) where it can be observed that nt-TiO2-9 had the highest emission in blue band while nt-TiO2-6 had the lowest one. The main defect states of these samples are self-trapped excitons (STE), the single-ionized oxygen vacancies (Vo*), the doubly-ionized oxygen vacancies (Vo**), and excited states of Ti3+ that are related to the rutile phase, at 416 nm, 443 nm, 531 nm, and 802 nm, respectively [2931].

In Figure 5b, we can observe the spectra of the crystalline films. Specifically, for nt-TiO2-1 sample, the emission increased six times in violet and blue bands with respect to the amorphous nt-TiO2-1 sample before thermal treatment, because the morphology of nanotubes is ordered and organized. Conversely, nt-TiO2-9 had a poor change in the PL intensity because this has a morphology with a low control of periodicity. Another important emission in violet and blue bands is the produced by crystalline nt-TiO2-6 since it had an increasing of about four times higher compared to the amorphous nt-TiO2-6 sample.

To analyze the defect states and their behavior with respect to thermal treatment, we deconvolved the PL spectra for each of the samples. Figure 5c shows the main defect states for nt-TiO2-1, first STE at 415 nm which is the maximum peak of spectrum, secondly, Vo* at 448 nm, thirdly, Vo** at 531 nm and, finally, excited states of Ti3+ at 800 nm. The defect states area percentage of this sample is presented in the pie chart, the area with the highest percentage is Vo** (48.7%), while Vo* is only 22.3%. As we can see in bar graph of Figure 5b, nt-TiO2-1 sample has an area of Vo* larger than the other three samples (nt-TiO2-3, nt-TiO2-6, and nt-TiO2-9). Figure 5d shows the spectra of nt-TiO2-3, here, the largest area is observed for Vo** (57.1%), followed by Vo* (19.4%), STE (19%), and excited states of Ti3+ (4.5%). For nt-TiO2-6 sample, STE were not detected, nevertheless, Vo* presented an area similar to (0.997 ratio) nt-TiO2-3 (Fig. 5e). In particular, nt-TiO2-9 did not have Vo* (Fig. 5f) but it had the largest area of self-trapped excitons than the other samples. Thus, the increase in Vo* in nanotubes can directly affect the change in electrical resistance and improve the sensor response. In summary, we increased oxygen vacancies as an effect of morphology by modifying the concentration of the synthesis electrolyte solutions

thumbnail Fig. 5

PL intensity of TiO2 nanotubes (a) after anodization and (b) after thermal treatment; and the deconvolution of PL spectra for (c) nt-TiO2-1, (d) nt-TiO2-3, (e) nt-TiO2-6, and (f) nt-TiO2-9.

3.3 The band gap energy of TiO2 nanotubes with different morphology

The band gap energy was calculated by linear extrapolation of the Tauc plot method as a function of photon energy from the transmittance measurement when TiO2 nanostructures had an anatase phase. Using 2 as the value of the indirect allowed transitions in the TiO2 semiconductor, the optical band gap was calculated based on equation [32]:(1)where α is the absorption coefficient, A is a constant, h is constant of Planck and Eg is the optical band gap energy and v is the photon frequency.

In order to understand the effect of the morphology on optical properties, band gap energy was calculated for each sample when they had crystalline structure (Fig. 6). The band gap energy of nt-TiO2-1 sample was 3.2 eV corresponding to the value of the energy of the phase anatase as previously reported [32,33]. Also, in Figure 6 it can be seen that the band gap energy decreases with increasing the content of water. The band gap energy of disordered nanotubes decreases until 2.78 eV (nt-TiO2-9). For nt-TiO2-3 and nt-TiO2-6, a similar value of band gap energy was observed. Therefore, the high water content in the electrolyte solutions lends to lower values in the band gap energy, and it is attributed to morphology and stoichiometry [32]. In this way, the disorganization, geometrical shape and no-stoichiometry of the TiO2 nanostructures can modify the band gap energy.

Several studies have reported that oxygen vacancies and dopants can move the band gap to low values (low transmittance in the visible spectrum) [33] or high values of energy because oxygen vacancies act as donor states within the band gap [34]. In this way, oxygen vacancies have narrower the band gap because these create localized states under the edge of the conduction band [35].

thumbnail Fig. 6

Band gap energy of TiO2 nanotubes by Tauc plot.

3.4 The effect of electrolyte solution on electrical resistance of TiO2 nanotubes for gas sensing

To correlate the morphology effect on the physical, optical and chemical properties of the nt-TiO2 films, we measured electrical resistance as a function of time for each sensing device with different morphology. Figure 7a shows the change in electrical resistance when gas sensors were used in the two different tests. As we observe, in t = 0 s, all devices had a high resistance in the presence of air (several tens kilo-ohms). After a while, when the gas sensors were exposed to N2 or N2 + H2 gases, a reduction in the resistance was observed (a few ohms). This means that electrical resistance decreases with increasing the amount of N2 or N2 + H2. Due to partial oxidation of the TiO2 film, electrical resistance is lower than other studies [3641]. The best and fastest change in electrical resistance occurred in nt-TiO2-1 device, decreasing by three orders of magnitude during 160 s of testing when the sensor was tested with N2 + H2 because this device had a better control of the geometric shape with self-ordered and self-organized nanostructures. Besides, we observed a good response from sensing device by the highest crystallinity of the anatase phase as well as the highest amount of the defect states (oxygen vacancies).

Oxygen vacancies are key defect states in the sensing mechanism since they act as donor states close to the edge of the conduction band, increasing the adsorption of the target gas. They have been studied using either different oxidizing or reducing gas and nanostructure films [4244]. Consequently, nt-TiO2-3 and nt-TiO2-6 had a good response, because they had a similar amount of oxygen vacancies but less than nt-TiO2-1 device. By contrast, nt-TiO2-9 had a poor change in electrical resistance in the different atmospheres because this did not present the peak corresponding to Vo* in PL spectrum after thermal treatment, but they had STE, Vo**, and excited states of Ti3+. In this way, these defect states did not enhance the response of the device. Also, nt-TiO2-9 showed a low intensity of crystallinity in the anatase phase and the lowest band gap energy. Accordingly, these results demonstrate that the amount of Vo* improves the reactivity of metal oxide surfaces because they create donor states in band gap [4244]. Another important characteristic in Figure 7a is the repeatability of each device for N2 (2 a) and 90% N2 + 10% H2 (2 b). The results show that an excellent repeatability was obtained from different gas sensors. From Figure 7a it can be seen that the response time is shorter when the device tests N2 + H2 because H2 is more reactive and smaller than N2 and, therefore, can be more easily adsorbed by Vo*.

Based on the results of electrical resistance and PL, the gas sensing mechanism of the nt-TiO2 devices by oxygen vacancies is proposed in Figures 7b and 7c when the sensor was tested in air and reducing gas, respectively. The mechanism starts, when oxygen is adsorbed on the surface of nanotubes by Vo*, then, the electrons from the conduction band are trapped by the defect states and by adsorbed oxygen. Afterward, O, , and O2− are created by adsorption; and hence, electrical resistance is increased. In fact, a good change in electrical resistance can be associated with the adsorption and desorption mechanisms of the gas on the surface of the nanotubes. In our device, another sensing mechanism is the Van der Waals forces because the sensor works at room temperature. Furthermore, oxygen is crucial to increase the electrical resistance, when N2 + H2 is adsorbed by physisorption, this process creates the surface states in the band gap and modifies the Fermi-Dirac Distribution [44]. But, this sensing mechanism has a small contribution to resistance change compared to oxygen vacancies [44]. Because N2 and H2 act as reducing agents and nanotubes are considered as n-type semiconductor, electrons density increases with increasing of the reducing gas flux because the trapped electrons return to conduction band (Fig. 7c). Therefore, the electrical resistance decreases in the presence of N2 or N2 + H2. The Schottky barrier is another important sensing mechanism to get a good sensor response because the final structure is Ti/TiO2 (metal/semiconductor) [45]. As we can see in Figures 7b and 7c, the barrier height is high when oxygen is adsorbed on the surface of the TiO2 film while it is lower when reducing gas (N2 or H2) is adsorbed [41,46,47]. Wall thickness is another important factor to increase sensor response since a thinner wall thickness usually has a high gas sensor response [48,49]. This point is very important because nt-TiO2-1 device has the thinnest wall thickness and the best electrical resistance change of our devices.

In order to compare our results with other reports, the sensor response was calculated with the following equation for a reducing gas [41,46,48,49]:(2)where Ra is the resistance in air and Rg is the resistance of target gas.

As it can be seen in Figure 7b, nt-TiO2-1 device got the best response of gas sensor when tested with N2 + H2. In this case, the response is mainly influenced by the greater amount of Vo* in the nanotubes due to their better organization and thinner wall thickness, thus, the sensor response was 713 when the sensing device was tested with N2 + H2 and 562 when it was tested with N2. The sensor response of nt-TiO2-1 is significant because another study reported a response of 3% for sensing N2 at 500 Torr and a poor change in electrical resistance (17 Ohm) when they used three-terminal carbon nanotubes [49]. Also, it is evident that the same response is obtained for nt-TiO2-3 and nt-TiO2-6 devices because they had a similar amount of Vo*, which operates as donor states in the band gap. The response time is a parameter that should be improved since industrial applications require fast responses; however, the response time obtained in this work is hardly comparable to other devices with good characteristics [50,51]. Figure 7c shows the sensor response and concentration of Vo* of each device. As we can see, the sensor response increases with increasing of amount of Vo*. In conclusion, the nanomaterials that we synthesize improve the sensor response due to the number of oxygen vacancies generated as an effect on the morphology of the nanotubes.

thumbnail Fig. 7

(a) The electrical resistance of TiO2 nanotubes, the schematic of gas sensing mechanism of nt-TiO2-1 and their energy bands in (b) air and (c) reducing gas, (d) sensor response of crystalline TiO2 nanotubes, and (e) oxygen vacancies and the sensor response of each device. Here titanium, oxygen and oxygen vacancies are represented by gray circles, blue circles and red circles, respectively.

4 Conclusions

In this research work, we studied the effect of morphology on the optical, structural, electrical, and chemical properties of TiO2 nanotubes for gas sensing applications. For that, we synthesized four nanomaterials using electrolyte solutions with different water content and we found that nanotubes made with a low concentration of water have a more ordered morphology with a thinner wall thickness, a shorter length and a more defined circular diameter than nanomaterials synthesized with high water concentration solutions. EDS analysis showed that the elemental ratio of Ti:O for the sample prepared with the lowest water content, nt-TiO2-1, was 1:1.7, the closest to the stoichiometry and it is attributed to titanium substrate and incomplete oxidation.

As an effect of the morphology, after thermal treatment, we observed, by XRD, an increase in the intensity of the peak corresponding to the anatase phase when the TiO2 films were more organized. Furthermore, the emission improved and increased six times in violet and blue bands when the nanotubes had the thinnest wall thickness and the best organization compared to the other TiO2 films. Photoluminescence analysis showed that the main state defect related to the control of morphology is single-ionized oxygen vacancies, and this increases as the wall thickness decreases.

In addition, low control of the organization of nanotubes decreases the band gap energy because defect states and non-stoichiometry create localized states within the band gap. The gas sensing tests showed that the best and fastest change in the electrical resistance occurred in nt-TiO2-1 device, when the sensor was tested with N2 + H2 due to the highest crystallinity of the anatase phase, as well as the highest amount of oxygen vacancies.

Morphology of TiO2 nanotubes is a key parameter for gas sensing applications, as the most ordered nanostructures showed the best response to gas detection tests.

These devices are simple, easy to make and inexpensive. In order to evaluate the compatibility with CMOS technology, our devices are going to be made on corning glass and silicon, and we are going to evaluate the selectivity with other gases.

Author contribution statement

Alba Arenas Hernandez and Carlos Zúñiga Islas designed, and planned the synthesis and characterization of TiO2 nanotubes. Alba Arenas Hernandez prepared, synthesized, and characterized TiO2 nanotubes. Alba Arenas Hernandez and Julio César Mendoza Cervantes analyzed the results of the morphological study, EDS study, PL study, transmittance spectra, structural study, and the sensor response of the TiO2 films. Alba Arenas Hernandez and Julio César Mendoza Cervantes wrote the manuscript. Carlos Zúñiga Islas reviewed and evaluated the results of the characterization and manuscript.

Acknowledgments

Alba Arenas Hernandez acknowledges to the National Council of Science and Technology of Mexico, CONACYT, for its support for this work through a PhD scholarship.

References

  1. Z. Jedi-Soltanabadi, N. Pishkar, M. Ghoranneviss, J. Theor. Appl. Phys. 12, 135 (2018) [Google Scholar]
  2. L.B. Arruda, C.M. Santos, M.O. Orlandi, W.H. Schreiner, P.N. Lisboa-Filho, Ceram. Int. 41, 2884 (2015) [Google Scholar]
  3. P. Roy, S. Berger, P. Schmuki, Angew. Chem. Int. Ed. 50, 2904 (2011) [Google Scholar]
  4. K. Lee, A. Mazare, P. Schmuki, Chem. Rev. 114, 9385 (2014) [PubMed] [Google Scholar]
  5. S. Park, S. Kim, S. Park, W.I. Lee, C. Lee, Sensors 14, 15849 (2014) [Google Scholar]
  6. L. Yang, S. Luo, Q. Cai et al., Chin. Sci. Bull. 55, 331 (2010) [Google Scholar]
  7. K. Indira, U.K. Mudali, T. Nishimura, N. Rajendran, J. Bio. Tribo. Corros. 1, 28 (2015) [Google Scholar]
  8. K.W. Cheung, J. Yu, D. Ho, Sensors 18, 3770 (2018) [Google Scholar]
  9. Z. Li, H. Li, Z. Wu, M. Wang, J. Luo, H. Torun, P. Hu, C. Yang, M. Grundmann, X. Liu, Y.Q. Fu, Mater. Horiz. 6, 470 (2019) [Google Scholar]
  10. O.K. Varghese, D. Gong, M. Paulose, K.G. Ong, C.A. Grimes, Sens. Actuators B 93, 338 (2003) [Google Scholar]
  11. B. Comert, N. Akin, M. Donmez, S. Saglam, S. Ozcelik, IEEE Sens. J. 16, 24 (2016) [Google Scholar]
  12. M. Kulkarni, A. Mazare, P. Schmuki, A. Iglic, Adv. Mater. Lett. 7, 23 (2016) [Google Scholar]
  13. A. Hazra, K. Dutta, B. Bhowmik, P.P. Chattopadhyay, P. Bhattacharyya, Appl. Phys. Lett. 105, 081604 (2014) [Google Scholar]
  14. P.M. Perillo, D.F. Rodríguez, Sens. Actuators B 171, 639 (2012) [Google Scholar]
  15. Y. Wang, T. Wu, Y. Zhou, C. Meng, W. Zhu, L. Liu, Sensors 17, 1971 (2017) [Google Scholar]
  16. J. Dai, J. Yuan, P. Giannozzi, Appl. Phys. Lett. 95, 232105 (2009) [Google Scholar]
  17. H.F. Lu, F. Li, G. Liu, Z.G. Chen, D.W. Wang, H.T. Fang, G.Q. Lu, Z.H. Jiang, H.M. Cheng, Nanotechnology 19, 405504 (2008) [PubMed] [Google Scholar]
  18. A.A. Valeeva, I.B. Dorosheva, E.A. Kozlova, R.V. Kamalov, A.S. Vokhmintsev, D.S. Selishchev, A.A. Saraev, E. Yu. Gerasimov, I.A. Weinstein, A.A. Rempel, J. Alloys Compd. 796, 293 (2019) [Google Scholar]
  19. H. Song, K. Cheng, H. Guo, F. Wang, J. Wang, N. Zhu, M. Bai, X. Wang, Catal. Commun. 97, 23 (2017) [Google Scholar]
  20. M. Naghizadeh, S. Ghannadi, H. Abdizadeh, M.R. Golobostanfard, Adv. Mater. Res. 829, 907 (2014) [Google Scholar]
  21. W. Wei, S. Berger, C. Hauser, K. Meyer, M. Yang, P. Schmuki, Electrochem. Commun. 12, 1184 (2010) [Google Scholar]
  22. J. Hu, G. Xu, J. Wang, J. Lv, X. Zhang, Z. Zheng, Y. Wua, J. Electrochem. Soc. 161, H529 (2014) [Google Scholar]
  23. K. Indira, S. Ningshenb, U. Kamachi Mudali, N. Rajendran, Mater. Charact. 71, 58 (2012) [Google Scholar]
  24. S.H.M. Suhaimy, S.B.A. Hamid, C.W. Lai, M.R. Hasan, M.R. Johan, Catalysts 6, 167 (2016) [Google Scholar]
  25. J.V. Pasikhani, N. Gilani, A.E. Pirbazari, Nano-Struct. Nano-Objects 8, 7 (2016) [Google Scholar]
  26. J.M. Macak, H. Tsuchiya, A. Ghicov, K. Yasuda, R. Hahn, S. Bauer, P. Schmuki, Curr. Opin. Solid State Mater. Sci. 11, 3 (2007) [Google Scholar]
  27. J.A. Díaz-Real, G.C. Dubed-Bandomo, J. Galindo-de-la-Rosa, L.G. Arriaga, J. Ledesma-García, N. Alonso-Vante, Beilstein J. Nanotechnol. 9, 2628 (2018) [PubMed] [Google Scholar]
  28. M.C. Nevárez-Martínez, M.P. Kobylanski, P. Mazierski, J. Wółkiewicz, G. Trykowski, A. Malankowska, M. Kozak, P.J. Espinoza-Montero, A. Zaleska-Medynska, Molecules 22, 564 (2017) [Google Scholar]
  29. C. Mercado, Z. Seeley, A. Bandyopadhyay, S. Bose, J.L. McHale, ACS Appl. Mater. Interfaces 3, 2281 (2011) [Google Scholar]
  30. H. Zhang, M. Zhou, Q. Fu, B. Lei, W. Lin, H. Guo, M. Wu, Y. Lei, Nanotechnology 25, 275603 (2014) [PubMed] [Google Scholar]
  31. L. Zhang, W. Yu, C. Han, J. Guo, Q. Zhang, H. Xie, Q. Shao, Z. Sun, Z. Guod, J. Electrochem. Soc. 164, H651 (2017) [Google Scholar]
  32. R.G. Freitas, M.A. Santanna, E.C. Pereira, J. Power Sources 251, 178 (2014) [Google Scholar]
  33. N. Pishkar, Z. Jedi-soltanabadi, M. Ghoranneviss, Results Phys. 10, 466 (2018) [Google Scholar]
  34. B. Bharti, S. Kumar, H.N. Lee, R. Kumar, Sci. Rep. 6, 32355 (2016) [PubMed] [Google Scholar]
  35. H. Zhao, F. Pan, Y. Li, J. Materiomics 3, 17 (2017) [Google Scholar]
  36. T. Plecenik, M. Mosko, A.A. Haidry, P. Durina, M. Truchly, B. Granci, M. Gregor, T. Roch, L. Satrapinskyy, A. Mosková, M. Mikula, P. Kús, A. Plecenik, Sens. Actuators B Chem. 207, 351 (2015) [Google Scholar]
  37. M. Cerchez, H. Langer, M.E. Achhab, T. Heinzel, D. Ostermann, H. Luder, J. Degenhardt, Appl. Phys. Lett. 103, 033522 (2013) [Google Scholar]
  38. M. Hübner, C.E. Simion, A. Tomescu-Stanoiu, S. Pokhrel, N. Bârsan, U. Weimar, Sens. Actuators B Chem. 153, 347 (2011) [Google Scholar]
  39. A. Gurlo, M. Sahm, A. Oprea, N. Barsan, U. Weimar, Sens. Actuators B Chem. 102, 291 (2004) [Google Scholar]
  40. F. Lange, S. Perl, S. Hofling, M. Kamp, Appl. Phys. Lett. 106, 063905 (2015) [Google Scholar]
  41. A.A. Haidry, L. Sun, B. Saruhan, A. Plecenik, T. Plecenik, H. Shen, Z. Yao, Sens. Actuators B Chem. 274, 10 (2018) [Google Scholar]
  42. M. Al-Hashem, S. Akbar, P. Morris, Sens. Actuators B Chem. 301, 126845 (2019) [Google Scholar]
  43. V. Galstyan, E. Comini, G. Faglia, G. Sberveglieri, Sensors 13, 14813 (2013) [Google Scholar]
  44. Y. Ling, F. Ren, J. Feng, Int. J. Hydrogen Energy 41, 7691 (2016) [Google Scholar]
  45. K. Chen, K. Xie, X. Feng, S. Wang, R. Hu, H. Gu, Y. Li, Int. J. Hydrogen Energy 37, 13602 (2012) [Google Scholar]
  46. A.A. Haidry, A. Ebach-Stahl, B. Saruhan, Sens. Actuators B Chem. 253, 1043 (2017) [Google Scholar]
  47. M. Paulose, O.K. Varghese, G.K. Mor, C.A. Grimes, K.G. Ong, Nanotechnology 17, 398 (2006) [Google Scholar]
  48. A. Hazra, B. Bhowmik, K. Dutta, P.P. Chattopadhyay, P. Bhattacharyya, ACS Appl. Mater. Interfaces 7, 18 (2015) [Google Scholar]
  49. C.S. Huang, B.R. Huang, Y.H. Jang, M.S. Tsai, C.Y. Yeh, Diamond Related Mater. 14, 1872 (2005) [Google Scholar]
  50. Z. Li, Z. Yao, A.A. Haidry, T. Plecenik, L.J. Xie, L.C. Sun, Q. Fatima, Int. J. Hydrogen Energy 43, 1 (2018) [Google Scholar]
  51. Y. Wang, T. Wu, Y. Zhou, C. Meng, W. Zhu, L. Liu, Sensors 17, 1971 (2017) [Google Scholar]

Cite this article as: Alba Arenas-Hernandez, Carlos Zúñiga-Islas, Julio César Mendoza-Cervantes, A study of the effect of morphology on the optical and electrical properties of TiO2 nanotubes for gas sensing applications, Eur. Phys. J. Appl. Phys. 90, 30102 (2020)

All Tables

Table 1

Length and elemental composition of the TiO2 nanotubes prepared by electrochemical anodization.

All Figures

thumbnail Fig. 1

Schematic representation of the gas sensor using nt-TiO2.

In the text
thumbnail Fig. 2

SEM images of TiO2 nanotubes synthesized by electrochemical anodization with (a) 1 wt%, (b) 3 wt%, (c) 6 wt%, and (d) 9 wt% of deionized water and 0.255 wt% NH4F in ethylene glycol.

In the text
thumbnail Fig. 3

SEM images of cross-section of TiO2 nanotubes synthesized by electrochemical anodization with (a) 1 wt%, (b) 3 wt%, (c) 6 wt%, and (d) 9 wt% of deionized water and 0.255 wt% NH4F in ethylene glycol.

In the text
thumbnail Fig. 4

XRD pattern of TiO2 nanotubes synthesized by electrochemical anodization.

In the text
thumbnail Fig. 5

PL intensity of TiO2 nanotubes (a) after anodization and (b) after thermal treatment; and the deconvolution of PL spectra for (c) nt-TiO2-1, (d) nt-TiO2-3, (e) nt-TiO2-6, and (f) nt-TiO2-9.

In the text
thumbnail Fig. 6

Band gap energy of TiO2 nanotubes by Tauc plot.

In the text
thumbnail Fig. 7

(a) The electrical resistance of TiO2 nanotubes, the schematic of gas sensing mechanism of nt-TiO2-1 and their energy bands in (b) air and (c) reducing gas, (d) sensor response of crystalline TiO2 nanotubes, and (e) oxygen vacancies and the sensor response of each device. Here titanium, oxygen and oxygen vacancies are represented by gray circles, blue circles and red circles, respectively.

In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.