Abstract

The p-type ZnO nanorod arrays were prepared by doping Na with hydrothermal method. The structural, electrical, and optical properties were explored by XRD, Hall-effect, PL, and Raman spectra. The carrier concentrations and the mobility of Na-doped ZnO nanorod arrays are arranged from  cm−3 to  cm−3 and 0.45 cm2 v−1 s−1 to 106 cm2 v−1 s−1, respectively.

1. Introduction

Zinc oxide (ZnO) is a promising material candidate for photoelectronic devices due to its direct wide electron energy bandgap of 3.37 eV at room temperature and large exciton binding energy of 60 meV. ZnO is conventionally used as a substrate for GaN and has advantages over GaN such as larger exciton binding energy, lower refraction index, and shallower acceptor levels. Despite its advantages over GaN, the application of ZnO in optoelectronics is hampered by the lack of stable p-type doping due to self-compensation by donor-like native defects, low solubility of p-type dopants, and formation of deep acceptor levels [1]. The n-type ZnO is available even without any intentional doping, but it is very difficult to dope ZnO p-type. Many groups have realized p-type ZnO conversion by doping group V elements (N [2], P [3], As [4], and Sb [5]) and groups I and IB (Li [6], Na [79], K [10], Ag [11], and Cu [12]), but the stability was not good. Among the group V elements, N is the most suitable dopant because it has about the same ionic radius as O and thus should readily be substituted on O sites. Indeed, it remains difficult to achieve good quality p-type conduction in N-doped ZnO due to either a low dopant solubility or a high defect ionization energy. Theoretically, group I element substituted on Zn sites is shallow acceptors [13], but instead these dopants tend to occupy the interstitial sites, partly due to their small atomic radius, and therefore act mainly as donors. On the other hand, reports of Na-doped p-type ZnO films are rather poor [14]. Up to now, various methods have been employed to prepare p-type ZnO including molecular beam epitaxy (MBE) [15], chemical vapor deposition [16], and pulsed laser deposition [17], in which complicated, high temperature deposition and expensive vacuum systems have been employed. In contrast, the hydrothermal method was proven to be an ideal technique for the development of p-type ZnO films owing to its simplicity and easy control of the composition at low cost [1820]. To date, there is no report of Na doping of p-type ZnO nanorods by a hydrothermal method. Here, we report a simple, reliable, and low-cost method for growing and doping p-type ZnO nanorod arrays in aqueous solution, with Na as the p-type dopant.

2. Experimental Section

A high-resistance Si (100) wafer, the resistivity of which is above 500 Ω cm, was used as the substrate. Precleaned silicon substrate was coated with a ZnO seed layer by sol-gel spin coating. Zn(AC)2·2H2O were first dissolved in the mixture of 2-methoxyethanol and monoethanolamine. The molar ratio of monoethanolamine to Zn(AC)2·2H2O was maintained at 1 : 1 and the concentration of Zn(AC)2·2H2O was 75 mM. The sol was aged for 24 h at room temperature. The above solution was then spin-coated at 500 rpm for 15 s and then at 5000 rpm for 60 s, followed by drying at 120°C for 15 min. The procedure from coating to drying was repeated five times before the substrates were placed into a 360°C tube furnace in air for 20 min to form ZnO seed layer, evaporate the solvent, and remove any organic residue. Subsequently, ZnO nanorod arrays were grown directly by placing the substrate at the bottom of the beaker in 20 mL of aqueous solution of 12.5 mM Zn(AC)2·2H2O, 12.5 mM hexamethylenetetramine (HMTA, C6H12N4), and different amounts of NaCl in a 50 mL autoclave. Different Na doping levels were obtained by varying the molar ratio of Na/Zn () (0, 2.8, 5.6, 8.4, and 11.2) in the precursor solution. Finally the autoclave was sealed and heated at 95°C for 2 h. Subsequently, the samples are washed with deionized water for several times to remove the residual salt and amino complex and dried in a flow of nitrogen gas.

Powder X-ray diffraction (XRD) analyses were performed on a Philips PW-1830 X-ray diffractometer with Cu Kα irradiation ( Å) at a scanning speed of 0.014°/sec over the 2θ range of 30–40°. The electronic morphology of the samples was examined by Hitachi S-4800 scanning electron microscope (SEM). The electrical properties were characterized by a Hall effect measurement system. Photoluminescence and Raman spectra of samples on Si substrate were recorded using Jobin-Yvon Lab-Ram high-resolution spectrometer with He-Cd laser ( nm).

3. Results and Discussion

XRD patterns of the hydrothermally grown samples were shown in Figure 1. The strong peak corresponding to ZnO (002) is observed in each film, which indicates high -axis oriented in all samples. It can be seen that the diffraction peak shows an obvious shift to a smaller angle of about 34.56, 34.52, 34.50, 34.38, and 34.52 for the sample with of 0, 2.8, 5.6, 8.4, and 11.2, respectively. This demonstrates that Na doping increases the lattice constant as a result of larger ionic radius of Na+ (0.089 nm) compared with that of Zn2+ (0.070 nm) [8].

Figure 2 shows the surface morphologies of samples grown by simple aqueous solution approach on Si substrates. The nanorods are vertically well-aligned on the substrate surface. The diameter and length of pure ZnO are in the range of around 30 nm and 754 nm, respectively. Figure 3 shows a plot of diameter and length of ZnO nanorod arrays as functions of . With the increase of , the diameter of ZnO nanorod increased to 50 nm, 60 nm, 70 nm, and 70 nm, and the length decreased to 150 nm, 112 nm, 98 nm, and 71 nm, respectively.

Table 1 summarizes the electrical properties of ZnO films. Under Hall effect measurements, the undoped sample showed n-type conductivity, while the doped sample with of 2.8–8.4 showed p-type conductivity, but when was increased to 11.2, the film showed n-type conductivity again. In contrast, the electrical properties of samples with of 2.8–8.4 showed higher mobility and carrier concentration. Mobility was up to 106 cm2 v−1 s−1 when was 2.8, and the carrier concentration was  cm−3 when was 5.6. Mobility and carrier concentration were significantly improved by doping. We think this is due to the right amount of Na ions in the Zn lattice sites acting as acceptors, compensation , and intrinsic donor defects in the lattice. Most of the Na+ occupies the lattice position, and the lattice expansion is small. The above XRD analysis further illustrates the rationality of our analysis. Electrical properties of Na-doped ZnO films by different processes are shown in Table 2. It can be seen that Na-doped nanorod arrays in this study showed excellent electrical properties.

Usually, PL spectra of ZnO at room temperature were composed of a near band edge UV emission peak at around 375 nm and a deep level green emission peak at 450–550 nm. Researchers generally believe that the near band edge UV emission peak was caused by the recombination luminescence of the band edge exciton, while the deep level green emission peak was caused by the recombination of the defects such as O vacancy and Zn interstitial. Strong excitonic emission at 375 nm was observed, as shown in Figure 4. The UV emission was so strong, which indicates that all the 1D ZnO arrays were of high UV emission efficiency and that the green emission could hardly be observed in the PL spectrum, indicating that the crystalline quality of ZnO was high and lattice oxygen vacancy concentration is very low.

Figure 5 shows the doped samples before and after 325 nm excitation of Raman spectra. Raman spectra can be sensitive to detect the crystallization of the material and the defect state. Undoped Al (LO) mode peak at 578.1 cm−1 corresponds to the El longitudinal optical phonon (LO) mode. When the molar ratio of Na and Zn was 2.8, 5.6, 7.8, and 11.2, the peak position was at 577.3 cm−1, 574.9 cm−1, 574.2 cm−1, and 575.8 cm−1, which moved 0.8 cm−1, 3.2 cm−1, 3.9 cm−1, and 2.3 cm−1, respectively. We observed that the peaks move to lower frequency, because this peak and the oxygen atom vacancies and zinc atoms interstitial related to its move reflect the changes in the concentration of free carriers in the ZnO crystal. So the concentration of free carriers in the ZnO crystals doped Na changed, and when the doping concentration of Al (LO) peak of the amount of movement is also different, this also shows the degree of change of free carrier concentration to be different.

4. Conclusion

In summary, Na-doped ZnO nanorod arrays with high crystalline quality were successfully fabricated on Si substrate by a simple hydrothermal method for the first time. With the increase of , the diameter of ZnO nanorod gradually increased from 30 nm to 70 nm, and the length decreased from 754 nm to 71 nm. Structural, electrical, and optical properties were investigated. The p-type carrier concentrations and mobility of Na-doped ZnO nanorod arrays arranged from  cm−3 to  cm−3 and 41.9 cm2 v−1 s−1 to 106 cm2 v−1 s−1, respectively.

Competing Interests

The authors declare that there are no competing interests regarding the publication of this paper.

Acknowledgments

This study is supported by the National Natural Science Foundation of China (no. 51402252) and the Natural Science Foundation of Jiangsu Province (no. BK20140463).