Abstract

ZnO nanorods were grown on thermally oxidized p-type silicon substrate using sol-gel method. The SEM image revealed high-density, well-aligned, and perpendicular ZnO nanorods on the oxidized silicon substrate. The XRD profile confirmed the c-axis orientation of the nanorods. PL measurements showed the synthesized ZnO nanorods have strong ultraviolet (UV) emission. The electrical characterization was performed using interdigitated silver electrodes to investigate the stability in the current flow of the fabricated device under different ultraviolet (UV) exposure times. It was notified that a stable current flow was observed after 60 min of UV exposure. The determination of stable current flow after UV exposure is necessary for UV-based gas sensing and optoelectronic devices.

1. Introduction

Zinc oxide (ZnO) belongs to II–VI semiconductor compound which has a wide bandgap energy (~3.37 eV) and stable wurtzite structure with lattice spacing = 0.325 nm and = 0.521 nm. The direct wide bandgap favours ZnO for the UV detection applications. Among various types of photodetectors, metal-semiconductor-metal (MSM) photoconductive detectors are generally the simplest in terms of fabrication and also easy to obtain higher gain [1]. The large exciton binding energy (~60 meV) makes ZnO suitable for optoelectronic applications, especially at temperature near and above room temperature [2]. Zinc oxide has attracted much research attention because of its unique properties and versatile applications in transparent electronics [3], ultraviolet (UV) light emitters [4], and piezoelectric devices [5]. UV photoresponse in ZnO was first observed by Mollow in 1940s [6]. After his study, many researchers studied the UV region photoresponse by utilizing single crystal, poly-, and crystalline films as well as nanowires [79]. A slow response time of UV photodetector based on polycrystalline ZnO thin films, ranging from a few minutes to several hours, is commonly observed [1015]. Due to large surface-to-volume ratio and miniaturized dimensionality of the active area, ZnO nanostructures are expected to have high photon conductance [16]. ZnO nanostructures can synthesized with different morphologies such as nanowires [17], nanotubes [18], nanorings [19], nano-tetrapods [20], and nanoflakes [21] via variety of methods including MOCVD [22], thermal evaporation [23], molecular beam epitaxy (MBE) [24], electrochemical deposition [25], spray pyrolysis [26], and sol-gel [27]. Among these methods, the sol-gel technique is the simplest and least expensive. One-dimensional ZnO nanostructures are gaining great attention because of their potential applications in nanoscale electronic and optoelectronic devices [28, 29]. Thus, a well-characterized and easily preparable ZnO nanostructure has long been expected for its application in specialized optoelectronic devices. Most of the studies on UV sensor have been focused on the mechanism investigation [30, 31] and improving the sensitivity [32, 33]. However, little attention has been paid to monitor the stability time of the sensor. Keem et al. [34] studied the photocurrent in ZnO nanowires and tested the sensor after 3000 s of UV illumination.

In this study, ZnO nanorods with photodetector functions were synthesized using an easily performable sol-gel method. The current-to-voltage (-) measurements were performed at different time intervals to investigate the effect of timing on the sensor at a fixed UV power.

2. Experimental

ZnO thin films were prepared on a thermally oxidized p-type silicon substrate of resistivity 1–100 Ω cm−1 using a sol-gel method. Before coating the seed layer, the silicon substrate was cleaned with RCA1 and RCA2 [27] to remove the contaminants. RCA1 solution was prepared by mixing DI water, ammonium hydroxide [NH4OH (27%)], and hydrogen peroxide [H2O2 (30%)] by maintaining the ratio of 5 : 1 : 1. For the RCA2 preparation, hydrochloric acid [HCL (27%)] and hydrogen peroxide [H2O2 (30%)] were mixed in DI water by maintaining the composition at 6 : 1 : 1. The residual oxide layer was removed by dipping the substrate into a BOE solution followed by washing with DI water and drying under N2 flow. After the cleaning process, the silicon wafer was thermally oxidized to generate a 1 μm layer of SiO2. The oxide layer is used to isolate the deposited ZnO film from silicon substrate in order to neglect the influence of current flow from silicon substrate. All the chemicals were used without further purification. The seed solution preparation and hydrothermal growth were performed following the procedure described in literatures [21, 27]. Briefly, 3.84 gm zinc acetate dihydrate was dissolved in 50 mL 2-methoxyethanol and 1.069 gm monoethanolamine. Monoethanolamine acts as a stabilizer and was added dropwise under stirring and constant temperature. Thus a ZnO solution of concentration 0.35 M was prepared. The solution was stirred for 2 h at 60°C until a transparent and homogeneous solution was obtained. The prepared solution was stored overnight at room temperature. The aged solution was spin coated on SiO2 grown p-type silicon substrate using spin coater which was rotated at 3000 rpm for 30 s. The film was preheated at 250°C for 20 min after each coating. The coating-to-drying process was repeated for five times and thus produced film was annealed at 550°C for 1 hr.

The ZnO nanorods were grown on seeded oxidized p-type silicon substrate using hydrothermal growth method. The growth solution was prepared by mixing zinc nitrate hexahydrate (25 mM) and hexamethyltetramine (25 mM) in 150 mL deionized water (DI). The seeded substrate was placed vertically so that the seeded side facing down in the growth solution. The growth was done in a preheated oven at 93°C for 6 hrs. After the growth, sample was washed with DI water to remove the residual salts and dried using N2 gun. The UV detector was fabricated as an MSM configuration using silver as the electrode. Interdigitated Ag electrodes were deposited on the SiO2/Si substrate using vacuum thermal evaporator. IDE device with the size of 8.75 × 5.00 mm was successfully deposited on the SiO2/Si substrate using hard mask. In this work, IDE with 8 fingers was used whereby the width and length of each finger was 0.25 mm and 4.75 mm, respectively. The spacing between the two fingers was 0.5 mm.

The surface morphology and structure of ZnO nanorods were examined using scanning electron microscopy (SEM- JEOL 6460LA). X-ray diffractometer (XRD) from Bruker D8 was used for phase identifications. The PL spectra of the ZnO nanorods were measured using He-Cd laser from Horiba Jyobin (HR 800) with an excitation wavelength of 325 nm at a power of 20 mW at room temperature. A UV LED with a wavelength of 365 nm and power of 2 mW was used as the illuminating source. A Keithley 2400 sourcemeter was used to measure the - characteristics. All the measurements were carried out under ambient environment.

3. Results and Discussion

The XRD pattern of the ZnO nanorods is shown in Figure 1. It consisted of several peaks at (100), (002), (101), (102), (103), and (112) planes. All the diffraction peaks were indexed to the hexagonal structure of the ZnO according to JCPDS Card no. 036- 1451. The lattice constants “” and “” for the ZnO nanostructures were calculated using the Bragg’s law [35] as: where is the order of diffraction, is the X-ray wavelength and is the spacing between planes of given Miller indices , , and . The plane spacing is related to the lattice constants “” and “” and the Miller indices by the following relation [35]: The calculated values for lattice parameters are and , indicating that the products were pure ZnO. The strongest peak (002) at with full-width half-maximum (FWHM) of showed good quality crystal and -axis orientation. No other peaks characterizing the impurities of the precursors, such as zinc nitrate and hexamethyltetramine, were detected. Table 1 shows the structural parameters of ZnO nanorods, whereas Table 2 displayed the calculated lattice constant values in comparison with the reported values. The grain size of the ZnO nanorods was estimated using the Scherer method [36] as follows: where is the grain size, is the wavelength, is the FWHM of the observed peak, and is the diffraction angle. It is well known that several aspects contribute to the strain in the films, like intrinsic strain induced by impurities and defects in the lattice, expansion coefficient mismatch between the film and substrate. The stress of the ZnO thin films was estimated using the following formula [36]: where is the lattice constant obtained from the (002) diffraction peak and is 5.205 Å for bulk ZnO [36].

The SEM image of ZnO nanorods is demonstrated in Figure 2. The SEM study clearly revealed the well-aligned, perpendicular, and high-density ZnO nanorods on oxidized silicon substrate. The SEM image strongly correlated with the XRD findings.

The photoluminescence spectrum of ZnO nanorods at room temperature is shown in Figure 3. The photoluminescence spectra demonstrated a strong UV emission at 381.97 nm and a weak visible emission at 565.94 nm. This UV band was due to the recombination of free excitons through an exciton-exciton collision process. It is suggested that deep level emissions are associated with the singly ionized oxygen vacancy in ZnO and results from the recombination of a photogenerated hole with the singly ionized charge state of this defect [39]. Usually, the visible emission from ZnO is attributed to different defects such as oxygen vacancies (), zinc vacancies (), or a complex defect involving interstitial zinc () and interstitial oxygen () [40]. The low intensity of the visible emission as compared to the near band edge (UV) emission in nanorods was an indication of minimum defect in ZnO nanorods.

The optical bandgap energy () of ZnO nanorods is calculated from photoluminescence spectroscopy using the following relation: where is the energy bandgap, is the Plank’s constant, is the speed of light, and is the wavelength. By using the above relation the calculated value for the bandgap of obtained ZnO nanorods was 3.2459 eV. Many attempts have been made to relate the refractive index and the energy gap through simple relationships [4146]. However, these relations of are independent of temperature and incident photon energy. Here the various relations between “” and “” will be reviewed. Ravindra et al. [46] had presented a linear form of “” as a function of : where and  eV−1. Light refraction and dispersion will be inspired. Hervé and Vandamme [47] proposed an empirical relation as follows:

where = 13.6 eV and = 3.4 eV. For group-IV semiconductors, Ghosh et al. [48] have published an empirical relationship based on the band structure and quantum dielectric considerations of Penn [55] and van Vechten [56] as: where = 25 + 212, = 0.21 + 4.25, and () refers to an appropriate average energy gap of the material. The calculated refractive indices of the end-point compounds and energy gaps are listed in Table 3, and give a good agreement with experimental [49, 54] and theoretical studies [5053]. This is verified by the calculation of the optical dielectric constant which depends on the refractive index. Note that the optical dielectric constant [57]. It is clear that the investigated refractive index “” using model of Ravindra et al. [46] is important for nanostructure ZnO in enhancing the detecting and sensing. It means high absorption may be attributed to increase sensors efficiency.

To investigate the effect of UV exposure time on the sensor, current-to-voltage characteristics are shown in Figure 4. It was observed that with exposure to UV light, the current drastically increased from 4.46 × 10−6 A to 2.23 × 10−5 A. However, with the increment of UV exposure time from 30 min to onward, the changes in current flow gradually became static and no significant changes in current flow were observed after 60 min of UV exposure as shown in Figure 4(b). The stability in current flow upon UV exposure is important to get a reliable signal in the gas sensing applications. The standard equation for the diode current can be written as [58]: where is the saturation current, is the applied voltage, is the ideality factor, is the electronic charge, is the Boltzmann constant, and is the temperature in Kelvin. The barrier height can be calculated using the following relation [58]: where is the contact area, is the effective Richardson’s constant (32 cm−2 K−2) [59], is the absolute temperature, is the electron charge, and is the barrier height. It was observed that the barrier height decreased from 0.718 eV to 0.65 eV from dark to 65 min of UV illumination.

Figure 5 shows the log() versus voltage curve of the fabricated sensor with dark and under various UV exposure times. Our log() versus voltage curve followed the same behaviour as reported by Li et al. [60]. The calculated value for ideality factor was 18 under the dark condition and that value decreased to 9.21 when the sensor was illuminated for 1 min. For the rest of the UV exposure times, there were no significant variations in the ideality factor of the fabricated sensor.

The current versus time relationship as well as the repeatability curve of ZnO nanorod-based sensor are shown in Figure 6. When the UV light is on, the resistance of the device sharply decreases. This may be ascribed to the surface oxygen adsorption of ZnO NWs and the electron-hole pairs generated by UV light in NWs [61]. The repeatability of the device was tested by switching UV light on and off at the same time intervals. A good reproducibility is noted and approximately identical current rise and decay was observed over 3 cycles. This indicated that UV sensing mechanism involves some reversible interactions among ZnO nanorods, ambient gases, and UV light [61].

Some theoretical and experimental studies suggested that photoresponse of ZnO is primarily governed by desorption and adsorption of oxygen [61]. Thus the overall photo-response can be divided into two separate processes: one is a bulk material-related process that is fast and reproducible. The other is a surface-related process that is slow. The latter can be explained in the following way: first, an electron can be trapped by oxygen to form a chemically adsorbed surface state as: When photon energy equal to or higher than the bandgap hits on the ZnO surface, an electron-hole pair is generated. The positively charged hole neutralizes the chemisorbed oxygen (O2): The freed electrons during the photodesorption process contribute to the increase in photoconductivity. The bulk-related process is the exciton generation by absorption of high-energy photons [62].

4. Conclusions

ZnO nanorods were successfully grown on surface exposed SiO2 of silicon substrate by a sol-gel method. The combined analysis by XRD, SEM, and photoluminescence demonstrated good crystallinity and orientation of the synthesized ZnO nanorods. The XRD revealed that was the major growth direction of the ZnO nanorods. The SEM image demonstrated that the nanorods are densely populated on the entire silicon substrate. The UV and visible emissions profile of ZnO nanorods was confirmed by photoluminescence spectra. The current-to-voltage profile showed stability in photocurrent after 60 min of UV illumination at room temperature, reflecting its potential applications in miniaturized optoelectronic and gas sensing devices.

Acknowledgments

The authors would like to acknowledge the financial support from the FRGS Grant no. 9003-00276 from the Ministry of Higher Education (MOHE). The authors would also like to thank the technical staff of Institute of Nano Electronic Engineering and School of Microelectronic Engineering, Universiti Malaysia Perlis for their kind support to smoothly perform the research.