Transparent ZnO-nanowire-based device for UV light detection and ethanol gas sensing on c-Si solar cell

Abstract

In this study, a transparent ZnO nanowire (NW)-based device for ethanol gas sensing and ultraviolet (UV) detection was fabricated and deposited onto an indium tin oxide/crystalline silicon (c-Si) solar cell. For UV detection, the photocurrent increased rapidly with a time constant of about 137 s when UV excitation was applied. The photocurrent decreased from 3 × 10⁻⁶ to 1.2 × 10⁻⁷ A when the UV light was switched off. For ethanol gas sensing, UV light was used to increase the quantity of O₂⁻ species. The ZnO sensor response increased from 8% to 21% when the ethanol gas concentration was increased from 50 to 150 ppm at 53 °C (heat generated by the c-Si solar cell). The sensor response approximately zero without solar illumination. The sensor had almost no effect on the transfer efficiency of the solar cell.

---

Introduction

The Internet of Things (IoT) is an intelligent network that connects devices to the internet for information sensing. The concept of IoT can be divided into three layers, namely sensing, network, and application. Sensing semiconductor is including optical detection and non-optical detection sensor, optical detector such as environment, proximity, infra-red ray, laser, etc. On the other hand, non-optical detector such as pressure, gas, temperature, accelerated, etc. Gas sensors are considered to be especially important. IoT sensors must be small, low-cost, low-power, wearable, and wireless.¹

Gas sensors can be divided into four main types, namely those based on electrochemistry, catalytic combustion, optics, and metal oxide semiconductors. Metal oxides have attracted considerable attention due to their small dimensions, low cost, and high compatibility with Si-based microelectronics. ZnO is a chemically and thermally stable metal–oxide semiconductor with a large exciton binding energy of 60 meV and a large bandgap energy of ∼3.3 eV (∼380 nm) at room temperature. ZnO is thus widely used in nanoelectronics and UV detectors. ZnO is also a native n-type semiconductor with some oxygen vacancies. The electrical properties of oxide nanowires are governed by oxygen vacancy-related donor states, it can provide more charge on the surface of nanowires.² Oxygen-related gas sensing generally involves chemisorption of oxygen on the oxide surface, followed by charge transfer during the reaction between chemisorbed oxygen and target gas molecules. As a result, surface resistance of the sensor element changes.^3–5 Thus, ZnO is also widely used in detecting toxic and combustible gases.⁶ Notably, the chemisorbed oxygen species depend strongly on temperature. At temperatures of lower than 130 °C, O₂⁻ is commonly chemisorbed. At higher temperatures, O⁻ is normally chemisorbed while O²⁻ is rapidly.⁷ However, work temperature is an issue due to the required low power consumption of IoT sensors. One-dimensional (1D) nanowires (NWs) and nanorods have attracted great attention in the field of gas sensors. Compared with bulk and thin film gas sensors, 1D NW gas sensors are expected to provide a larger response due to their high length-to-diameter (aspect) ratio and high surface-to-volume ratio.

With the growth of IoTs, self-powered devices are becoming more extensively investigated. For example, M. R. Hasan et al. fabricated the p-NiO/n-ZnO heterojunction ultraviolet photodetectors on plastic substrates.⁸ Other group has reported the use of using microfiber nanowire hybrid structure and its application on self-powered UV detection.⁹ Y. Y. Zeng, et al. proposed using Ag nanoparticle-modified ZnO NWs as an active photoanode and H₂O as the electrolyte.⁶ In this work, transparent ZnO NWs were deposited on the surface of a commercial crystalline silicon (c-Si) solar cell to form a self-powered ultraviolet (UV) detector and ethanol gas sensor. UV illumination enhances the sensitivity of the gas sensor at room temperature. The detailed properties of the fabricated sensors are discussed.

Experiment

Prior to the growth of ZnO NWs, a 100 nm-thick ZnO thin film was deposited onto a glass substrate via radio-frequency (RF) magnetron sputtering. For ZnO NW growth, the source of zinc vapor was Zn metal powder with a purity of 99% (2N) (Strem Chemicals). The zinc metal power was placed onto an alumina boat and inserted into a tube furnace. Ar and O₂ were used as the working gases in the furnace, with flow rates of 54.4 and 0.8 sccm, respectively. The work pressure was kept at 10 Torr, the reaction temperature was 600 °C, and the total growth time was 1 hour. The as-synthesized ZnO NWs were scraped and immersed in ethanol solution. Finally, the ZnO NWs were dropped onto a patterned indium tin oxide (ITO)/c-Si solar cell, as shown in Fig. 1.

Fig. 1 Schematic diagram of gas sensor fabrication and deposition onto c-Si solar cell.

To fabricate the patterned ITO/c-Si solar cell, a 300 nm-thick ITO thin film was deposited on a commercial c-Si solar cell via RF sputtering. Then, photo-resist was used to define an interdigital transducer pattern and the ITO thin film was wet-etched (2% HCl). Finally, the photo-resist was removed using an acetone solution.

A JEOL JSM-6500F field-emission scanning electron microscope (FESEM) operated at 10 keV was then used to characterize the structural properties of the as-grown ZnO NWs. The as-prepared products were electronically characterized using a Keithley 2400 source meter and a personal computer. The source meter provided an electric source and measured the output current. The transmittance and reflection spectra were measured using a spectrophotometer (Hitachi U-4100). A steady-state solar simulator was used for photovoltaic measurements (SS150, Science Tech Inc.).

Results and discussion

Fig. 2 shows the transmittance spectra of ITO/glass structures. It can be found that the transmittances are over 85% in the visible-light region.

Fig. 2 Optical transmittance spectra of ITO/glass.

Fig. 3(a) shows a cross-section FESEM image of the ZnO NWs grown on the ZnO/glass substrate. It can be seen that high-density vertically well-aligned ZnO NWs were grown on the ZnO/glass substrate. The average diameter and length of the NWs were approximately 100 nm and 21 μm, respectively. Fig. 3(b) shows a FESEM image of ZnO NWs dropped onto a patterned ITO/c-Si solar cell. A lot of ZnO NWs cover the solar cell surface. Fig. 3(c) plots current–voltage (I–V) characteristics measured from two ITO electrodes in air. The linear behavior reveals that good ohmic contacts were formed between the nanowires and the electrodes.

Fig. 3 (a) Cross-section FESEM image of as-grown ZnO NWs. (b) Top-view FESEM image of ZnO NWs dropped onto c-Si solar cell solar cell. (c) I–V of characteristics measured from the two electrodes of the ZnO nanowire samples.

Fig. 4 shows the reflectance spectra of the c-Si solar cell and ZnO NWs/patterned ITO/c-Si solar cell. The reflectance spectra are very similar. The ZnO NWs/patterned ITO/c-Si solar cell had lower reflectance in the UV range (360–450 nm), but the c-Si solar cell had lower reflectance in green-light to red range (500–800 nm), as shown in the inset of Fig. 4.

Fig. 4 Reflectance of c-Si solar cell surface with and without ZnO sensor. Inset shows magnified view of indicated range.

Fig. 5 shows the illuminated (AM 1.5G) current–voltage (I–V) characteristics of the c-Si solar cell structure with and without the ZnO NW sensor. The short-circuit current densities (J_sc) of solar cells with and without the ZnO NW sensor are 36.54 and 36.68 mA cm⁻², respectively. In addition, the open-circuit voltage (V_oc) of all samples is ∼0.6 V. The efficiencies of the c-Si solar cell with and without the ZnO NW sensor are about 16.29% and 16.49%, respectively. The results show that the c-Si solar cell with the ZnO NW sensor almost no have influence.

Fig. 5 I–V characteristics of c-Si solar cell structure with and without ZnO NW sensor (current density: J_sc; open-circuit voltage: V_oc; efficiency: η).

Fig. 6(a) shows the transient response of the fabricated ZnO NW photodetector prepared on patterned ITO/c-Si solar cells with UV (365 nm) excitation switched on and off. Upon exposure to UV light, the photocurrent initially rapidly increased, followed by a slow increase without reaching saturation. When the UV light was turned off, the photocurrent decreased rapidly initially value. The photocurrent during this period decreased from 3 × 10⁻⁶ to 1.2 × 10⁻⁷ A. The same experiment was repeated several times, with the same results obtained. Fig. 6(b) shows spectral responses of the ZnO nanowire photodetector using a 300 W Xe lamp dispersed by a monochromatic as the excitation source at room temperature. It should be noted that photo-responses of the fabricated photodetector were flat in the short wavelength region while sharp cutoff occurred at about 360 nm.

Fig. 6 (a) Photocurrent response of ZnO NW photodetector recorded with and without illumination. (b) Spectral responses of the crabwise ZnO nanowire photodetector measured with various applied bias.

Fig. 7 shows the response of the ZnO NW sensor measured for various ethanol concentrations at room temperature with and without illumination (AM 1.5G) (note: the temperature of the substrate was 53 °C). The sensor response increased from 8% to 21% when the ethanol gas concentration was increased from 50 to 150 ppm. During these measurements, ethanol vapor of various concentrations was introduced into a sealed chamber and the resistivity of the sensor was measured both in air (R_a) and in the presence of ethanol vapor (R_b). To quantify sensor performance, the response of the sensor is defined as [((R_a − R_b)/R_a) × 100%]. It was also found that the sensor response was approximately zero at room temperature without solar illumination.

Fig. 7 Sensor response of ZnO NW ethanol gas sensor on c-Si solar cells measured with and without illumination (AM 1.5G) for ethanol concentrations of from 50 to 150 ppm.

The detailed mechanism of this change is not yet clear. A possible explanation is related to the photoconduction of ZnO NWs being governed by the desorption and adsorption of oxygen.^10–13 In the dark, oxygen molecules on the surface of the NWs carry negative charges by capturing free electrons from n-type ZnO, creating a depletion layer with low conductivity near the surface. UV light (hν) absorption generates electron–hole pairs. The photo-generated holes oxidize the adsorbed negatively charged oxygen ions on the surface, while the remaining electrons in the conduction band increase conductivity. Thus, the photo response of the oxide semiconductor including the adsorption and photo desorption of oxygen on the surface gives rise to the very rapid change in conductivity, which was caused by photo-generated hole–electron pairs and their extinction. The surface of ZnO NWs has a large quantity of O₂⁻ species. The adsorption reactions are as follows:

O₂ (gas) ↔ O₂ (ads.) (1)

O₂ (ads.) + e_hν⁻ ↔ O₂⁻ (2)

In the metal oxide semiconductor gas sensing process, the chemisorbed oxygen species depend strongly on temperature (T). At temperatures of lower than 130 °C, O₂⁻ is commonly chemisorbed. At higher temperatures, O⁻ is normally chemisorbed, while O²⁻ is rapidly.⁷

O₂ (ads.) + e_T⁻ ↔ O₂⁻ (low T) (3)

O₂⁻ + e_T⁻ ↔ O⁻ (high T) (4)

Thus, the conductance of the ZnO NWs increased when the reducing gas was introduced into the test chamber due to the exchange of electrons between the ionosorbed species and ZnO itself. The reaction between the reducing gas and the surface of oxide sensor can be described as:¹⁴

R + O⁻ (ads.) ↔ RO + e⁻ (5)

where R is the reducing gas. To make the ZnO NW gas sensor work at low ambient temperature, UV light is used to increase the quantity of O₂⁻ species, as shown in Fig. 8. The results indicate that the ZnO NW sensor can act as both a UV light detector and an ethanol gas sensor on c-Si solar cells.

Fig. 8 Schematic diagram of the ZnO nanowire absorbed O₂⁻ by low temperature and UV illumination.

Conclusions

This study combined a ZnO NW sensor and solar cells. The ZnO NW sensor acts as both a gas sensor and a UV detector. For UV detection, the photocurrent increased rapidly with a time constant of about 137 s when UV excitation was applied. The photocurrent decreased from 3 × 10⁻⁶ to 1.2 × 10⁻⁷ A when the UV light was turned off. For ethanol gas sensing, UV light is used to increase the quantity of O₂⁻ species. The ZnO NW sensor response increased from 8% to 21% when the ethanol gas concentration was increased from 50 to 150 ppm at 53 °C (heat generated by c-Si solar cell). In addition, the ZnO NW sensor has no effect on the transfer efficiency of the solar cell.

Acknowledgements

The authors would like to thank the National Science Council and the National Nano Devices Laboratories, Tainan, Taiwan, for financially supporting this research under Contract No. 103-2221-E-492-047-MY3.

Notes and references

1. X. C. Chen, R. Woogeun and Z. H. Wang, China Communications, 2015, 12, 42–54.
2. T. J. Hsueh, Y. W. Chen, S. J. Chang, S. F. Wang, C. L. Hsu, Y. R. Lin, T. S. Lin and I. C. Chen, Sens. Actuators, B, 2007, 125, 498.
3. Z. Q. Zheng, J. D. Yao, B. Wang and G. W. Yang, Sci. Rep., 2015, 5, 11070.
4. C. S. Chou, Y. C. Wu and C. H. Lin, RSC Adv., 2014, 4, 52903–52910.
5. Q. J. Yu, C. L. Yu, J. Z. Wang, F. Y. Guo, S. Y. Gao, S. J. Jiao, H. T. Li, X. T. Zhang, X. Z. Wang, H. Gao, H. B. Yang and L. C. Zhao, RSC Adv., 2013, 3, 16619–16625.
6. Y. Y. Zeng, X. H. Pan, W. Dai, Y. C. Chen and Z. Z. Ye, RSC Adv., 2015, 5, 66738–66741.
7. M. N. Rumyantseva, E. A. Makeeva and A. M. Gas'kov, Russ. J. Gen. Chem., 2008, 78, 2556–2565.
8. M. R. Hasan, T. Xie, C. Sara, S. C. Barron, G. N. Liu, N. V. Nguyen, A. Motayed, M. V. Rao and R. Debnath, APL Mater., 2015, 3, 106101.
9. L. Zhang, S. Bai, C. Su, Y. B. Zheng, Y. Qin, C. Xu and Z. L. Wang, Adv. Funct. Mater., 2015, 25, 5794–5798.
10. S. S. Zhang, C. S. Xie, G. Z. Zhang, Q. Zhu and S. P. Zhang, Phys. Chem. Chem. Phys., 2015, 17, 18045–18054.
11. C. C. Yu, Y. T. Hsu, W. H. Lan, M. C. Shih, J. H. Hong, K. F. Huang and C. J. Huang, J. Nanomater., 2013, 2013, 531328.
12. J. C. Belmonte, J. Manzano, J. Arbiol, A. Cirera, J. Puigcorbe, A. Vila, N. Sabate, I. Gracia, C. Cane and J. R. Morante, Sens. Actuators, B, 2006, 114, 881–892.
13. Y. W. Heo, B. S. Kang, L. C. Tien, D. P. Norton, F. Ren, J. R. La Roche and S. J. Pearton, Appl. Phys. A: Mater. Sci. Process., 2005, 80, 497–499.
14. Y. Wang, J. Chen and X. Wu, Mater. Lett., 2001, 49, 361–364.

---