Open Access
Add to BibSonomyAbstract
Arsenic doped ZnO films are prepared using a pre-deposited GaAs finite surface source on sapphire substrates by a MOCVD method. Their conductivity and optical properties are closely related to the annealing process. The as-grown ZnO film shows n-type conductivity with weak FA emission. The in situ annealed sample shows p-type conductivity. The AsZn–2VZn acceptor level is confirmed by low-temperature photoluminescence measurement. The post annealed ZnO film appears to be n-type, which is attributed to the arsenic surface enrichment and the compensation of introduced donor like defects. Our method could be widely used in fabricating arsenic doped p-type ZnO related photoelectric devices.
© 2016 Optical Society of America
1. Introduction
ZnO is an important II-VI compound semiconductor, which has a wide direct band gap of 3.37 eV at room temperature and large exciton binding energy of 60 meV. It has been extensively investigated for fabrication of advanced optical and electrical devices, such as UV photodetectors, transparent electronics and nanogenerator [1–3]. For the device fabrication, n-type and p-type ZnO materials are essential. High quality n-type ZnO could been obtained by doping with Ga and Al elements [4,5]. But it is still very difficult to grow stable p-type ZnO with reliable performance, which hinders the application of ZnO based UV LEDs. The difficulty of growing p-type ZnO can be summarized for low solubility and deep level of the acceptor dopants, as well as the strong compensation of donor defects like zinc interstitial and oxygen vacancy or other induced impurities [6].
In the past fifteen years, researchers have tried different dopants to fabricate p-type ZnO. Group-V elements, such as N, P, As and Sb, which are recognized as effective acceptor dopants. The N dopant is investigated deeply because its electronic structure and ionic radius are closer to that of replaced O. But, the solubility and chemical stability of N in ZnO are small and weak. It leads to the difficulty of fabricating high performance N-doped p-type ZnO based UV LEDs. However, people have found a complex doping method by co-doping N with other elements, such as Li, B and Be [7–9]. The effect of co-doping elements is to enhance the incorporation of N acceptor and lower its ionization energy through the strong attractive interaction. However, the challenge of this method is the uncertain behavior of co-doping elements. To further pursue p-type conductivity in ZnO, other group-V dopants have been tried, such as P, As and Sb elements [10–15]. And according to the first-principles calculation results, the acceptor for these large-size-mismatched dopants is XZn-2VZn (X denotes P, As or Sb). For example, the As atom substitutes on the Zn atom site and induces two Zn-vacancies.
Among these elements, the covalent radius of As atom (119 pm) is closer to that of the Zn atom (122 pm). This indicates smaller lattice distortion during the acceptor forming process. There are many methods that can be used to grow arsenic doped p-type ZnO, for example sputtering, pulsed laser deposition, molecular beam epitaxy and metal organic chemical vapor deposition (MOCVD). Among them, the MOCVD method has advantages in doping and multilayer growth and has become a mainstream growth method for fabricating commercial devices. Diethylzinc and oxygen gas are usually used as zinc and oxygen sources for growing ZnO films in MOCVD systems. In most of the previous reports, researchers chose GaAs substrates to grow arsenic doped p-type ZnO, where arsenic acceptors could diffuse into the ZnO film by a subsequent annealing process [16–18]. The choice of the optimum annealing temperature and ambient is crucial to minimize the diffusion of gallium donors from the GaAs substrate. According to C.R. Bayless’s report, if the GaAs material is annealed in vacuum, the evaporation of the arsenic atoms is observed at 300°C. Up to 700°C, the evaporation of the gallium atoms starts [19]. Moreover, H.P. Sun has found that the gallium atoms could diffuse only a few atomic layers into the ZnO film, which is grown on Ga modified (0001) sapphire surface, after annealing at 650°C for 3 hours [20]. And in our other experiment, we have detected the gallium related XPS signals near the GaAs/ZnO interface after post annealing at 700°C for 1 hour. Analyzing the relevant literature, the suitable range of annealing temperature and time are 500-550°Cand 30-60 min in oxygen atmosphere. However, this method of growing arsenic doped p-type ZnO material is restricted to GaAs substrates. And the narrow band gap characteristics of the GaAs substrates are not conducive to fabricate ZnO based short wavelength LEDs and detectors. In this paper, we present a method to grow arsenic doped ZnO films, using a pre-deposited GaAs layer (on wide band gap sapphire substrate) as finite surface doping source. We carefully studied the arsenic doping characteristics of the different ZnO samples, which are annealed in reaction chamber and tube furnace. Our experiment provides another way to realize practical arsenic doped p-type ZnO material.
2. Experiments
On the first step, the GaAs layer, acting as finite surface doping source, was deposited on sapphire substrates by RF magnetron sputtering method at room temperature. High-purity polycrystalline GaAs and argon were used as target and sputtering gas. The sputtering power and working pressure were 80 W and 1 Pa, respectively. The thickness of GaAs layer is ~30 nm.
On the second step, the ZnO film was grown on this composite substrate by MOCVD method. The zinc source, oxygen source, reacting temperature, working pressure and growth time were Diethylzinc, oxygen, 450°C, 200 Pa and 20 min, respectively. The film thickness is ~600 nm. The thermal annealing was done following two different processes. For in situ annealing treatment in reaction chamber, after growing ZnO film, we shut off the zinc source and raised the temperature up to 500°C. The thermal annealing was performed in an oxygen atmosphere for 60 min. The annealing pressure is 200 Pa. For post annealing treatment in tube furnace, the ZnO film was put into the tube furnace and annealed at 500°C for 60 min in oxygen atmosphere under normal pressure. Here, we marked the as-grown, the in situ annealed and post annealed ZnO films as sample A, B and C, respectively. Hall-effect measurements were carried out to estimate the film conductivity. The structural quality of the ZnO films were characterized by X-ray diffraction (LabX XRD-6100, SHIMADZU) with Cukα radiation. The voltage and current of the X-ray tube were 40 kV and 30 mA. The scan speed and sampling pitch were 2.0000 degree/min and 0.0200 degree, respectively. The local chemical states and doping characteristics were investigated by X-ray photoelectron spectroscopy (XPS). The instrument model of the XPS system was ESCALAB 250xi. Its energy resolution was 0.45eV (Ag 3d5/2). And it could analyze the elements (except H and He) which were more than 0.1% atomic percentage. The optical quality and the impurity levels were studied by temperature-dependent photoluminescence (PL) measurement. The PL spectra were measured with a He-Cd CW laser operating at 325 nm. The mode, power, beam diameter, beam divergence and incident angle were TEM00, 30 mW, 1.2 mm, 0.5 mrad and 45 degree, respectively. And the PL emission was detected using iHR-320 Jobin-Yvon monochromator with a charge-coupled device (CCD) detector. The spectral resolution of the spectrometer was 0.06nm.
3. Results and discussion
Hall Effect measurements were carried out to characterize the electrical properties of the ZnO samples using a van der Pauw configuration. The results are summarized in Table 1. The as-grown sample A has an n-type conductivity with an electron concentration of 4.2 × 1017 cm−3 and a mobility of 2.3 cm2/V·s. The in situ annealed sample B shows p-type conductivity. The sample exhibits a hole concentration of 1.6 × 1017 cm−3 and a mobility of 1.3 cm2/V·s. The post annealed sample C shows n-type conductivity. Comparing with sample A, its electron concentration increases to 2.7 × 1018 cm−3 while the mobility decreases to 0.7 cm2/V·s.
Table 1. Electrical properties of the as grown and annealed ZnO films
XPS measurements were performed to well understand the elements distribution in ZnO films. For the as-grown ZnO film, neither arsenic nor gallium related signals are detected, which indicates the weak atomic diffusion from the pre-deposited GaAs layer. Because of the influence of native donor defects [6], the as-grown ZnO film shows n-type conductivity. For the in situ annealed ZnO film, the Ar+-etching dependent XPS profiles are implemented to carefully investigate the local chemical states and doping characteristics of the potential dopants. As expected, the gallium related signals are still not detected. It is consistent with the results of the ZnO films prepared on GaAs substrates [16–18,21]. However, the arsenic related signal is obviously detected. Figure 1 shows the As-3d profiles versus Ar+-etching time. The strength of these As-3d peaks do not vary much, which indicates the distribution of arsenic atoms is relatively uniform from the pre-deposited GaAs source. It is found that the As-3d binding energy decreases with increasing Ar+-etching time. On the film surface, the peak center is located at 45.0 eV, which indicates the oxidized state of As in As2O3 form. Then, it shifts to 44.4 eV after 10 min etching, and remains stable at 44.2 eV after 20 and 30 min etching. According to previous reports [12,21], the binding energy of AsZn-2VZn is near 44.0 eV. So the p-type conductivity of the in situ annealed ZnO film is attributed to the effect of AsZn-2VZn acceptor.
Figure 2 shows the XPS spectrum of As-3d core level of the post annealed ZnO film. The results are measured at the film surface and at the new exposed face after Ar+-etching for 30 min. Similarly, the binding energy of arsenic atoms is 45.3 eV at the film surface, and shifts to 44.2 eV at the new exposed face. Differently, the peak intensity obtained at the surface is much stronger than that obtained inside the film, which indicates the surface enrichment phenomenon of the arsenic atoms. Although the AsZn-2VZn acceptor has formed and the gallium related donors are not detected, the post annealed ZnO film does not yet show p-type conductivity. In order to find the compensation mechanism, the XRD and room-temperature PL measurements are used to characterize the structural and optical properties.
The XRD patterns are plotted in Fig. 3. The strongest diffraction peaks (for sample A to C) are located round 34.44°, which indicates the ZnO films have wurtzite-type structure with (002) orientation. The other peaks located round 36.28°, 47.62° and 62.94° are caused by the diffraction of ZnO (101), (102) and (103) planes, respectively. In order to characterize the preferred-orientation changes, we calculate the intensity ratio of ZnO(002)/ZnO(101). The values are 19.2, 22.4 and 3.2 for as-grown, in situ annealed and post annealed ZnO films. As seen, the small value indicates a crystal quality decline in the post annealed ZnO film. Figure 4 shows room-temperature PL spectra. Both of the as-grown and in situ annealed ZnO films have a strong ultraviolet (UV) emission peak round 377-378 nm. As reported, this UV emission is caused by the radiative recombination between electrons and holes near the band edge. But, for the post annealed ZnO film, the visible emission (round 495 nm) is much stronger than the ultraviolet emission (round 385 nm). Because visible emission is ascribed to donor like defects (such as VO or Zni [6]), the PL results indicate a large amount of donor like defects existing in the post annealed ZnO sample. Therefore, the acceptor compensation phenomenon in post annealed n-type ZnO film is attributed to the effect of introduced donor like defects caused by the deterioration of ZnO crystal quality. Furthermore, it is found that the UV peak positions have little difference. In order to investigate the internal mechanism, temperature-dependent PL measurement are implemented from 300K to 10K.
Fig. 3 XRD patterns of as-grown (Sample A), in situ annealed (Sample B) and post annealed (Sample C) ZnO films.
Fig. 4 PL spectra of the as-grown (Sample A), in situ annealed (Sample B) and post annealed (Sample C) ZnO films.
Figure 5-7 shows the temperature-dependent PL spectra of the as-grown, in situ annealed and post annealed ZnO films, respectively. In Fig. 5, the PL spectrum (10K) of the as-grown ZnO sample exhibits five peaks located around 3.355, 3.308, 3.227, 3.145 and 3.053 eV, respectively. The peak located at 3.355 eV can be assigned to a neutral acceptor bound exciton (A0X) [22,23]. And the peak at 3.308 is attributed to the recombination emission between free electrons and acceptor holes (FA) [22,23]. The intensity of the FA related peak is weaker than that of A0X related peak. It indicates the amount of effective acceptor is relatively small. The acceptor could not totally compensate the effect of donor like defects. So the as-grown ZnO film still shows n-type conductivity. Meanwhile, the peak located at 3.227 eV is ascribed to the recombination of donor acceptor pair (DAP) [22,23]. The emission around 3.145 and 3.053 eV are first-order and second-order longitudinal optical (LO) phonon replicas of the DAP emission [22,23]. Figure 6 shows the PL spectrum (10K) of the in situ annealed ZnO film, five peaks located round 3.352, 3.304, 3.212, 3.138 and 3.021 eV could also be found. They are ascribed to the recombination of A0X, FA, DAP, DAP-1LO and DAP-2LO, respectively. The FA emission is much stronger, which indicates the dominant effect of the arsenic related acceptor. So this ZnO film shows p-type conductivity. The arsenic related acceptor level can be calculated from the following equation: EA = Eg-EFA + kBT/2, where Eg is the intrinsic band gap, EFA is the transition between free electrons and acceptors. In our results, the value of EFA has been regarded as 3.304 eV, and Eg is evaluated as 3.437 eV [22]. So the value of EA is derived to be about 133 meV, which is quite consistent with the AsZn–2VZn acceptor model [12]. However, for post annealed ZnO films, the inhomogeneity of arsenic distribution and the deterioration of crystal quality make the energy level states become very complicated. Two broad peaks are found around 3.251 eV and 3.207 eV at 10K in Fig. 7. Besides, the peak intensity is strengthened and then weakened from 300K to 80K. These phenomenon could not be perfectly explained now. It still needs further research.
4. Conclusion
In conclusion, arsenic doped ZnO films were prepared using pre-deposited GaAs finite surface source. The arsenic chemical states and diffusion characteristics are closely related to the annealing process, which decide the film conductivity. The As-grown ZnO film shows n-type conductivity because of the lack of effective acceptors. For the in situ annealed ZnO film, the arsenic atoms diffuse uniformly. The binding energy is around 44.2 eV, which indicates the formation of the AsZn–2VZn acceptor. According to temperature-dependent PL measurement, the AsZn–2VZn acceptor level is about 133 meV above the valence band maximum. For the post annealed ZnO film, the conductivity appears to have an n-type characteristic in spite of the detection of arsenic related acceptor. One reason is that the arsenic atoms present a surface enrichment phenomenon, which reduces the amount of effective AsZn–2VZn acceptor in the film. Another reason is that the crystal quality of the post annealed ZnO film became worse than that of the as-grown and in situ annealed ZnO films. The introduced donor like defects could totally compensate the effect of the AsZn–2VZn acceptor. As a result, the n-type ZnO film is formed. Our experiments suggest that the in situ annealing process is more suitable to prepare p-type ZnO films using a pre-deposited GaAs finite surface source. And this method could be widely used in fabricating arsenic doped p-type ZnO related photoelectric devices.
Funding
National Natural Science Foundation of China (NSFC) (11601069); Scientific Research Fund of Liaoning Provincial Education Department (L2014457).
References and links
1. Z. P. Yang, Z. H. Xie, C. C. Lin, and Y. J. Lee, “Slanted n-ZnO nanorod arrays/p-GaN light-emitting diodes with strong ultraviolet emissions,” Opt. Mater. Express 5(2), 399–407 (2015). [CrossRef]
2. G. Nam, M. S. Kim, J. Kim, S. O. Kim, and J. Y. Leem, “Regrowth method for the enhancement in the photoluminescence, UV photoresponse, and electrical properties of n-ZnO:Sn films,” Opt. Mater. Express 6(1), 220–229 (2016).
3. S. Lu, Q. Liao, J. Qi, S. Liu, Y. Liu, Q. Liang, G. Zhang, and Y. Zhang, “The enhanced performance of piezoelectric nanogenerator via suppressing screening effect with Au particles/ZnO nanoarrays Schottky junction,” Nano Res. 9(2), 372–379 (2016). [CrossRef]
4. Z. Szabó, Z. Baji, P. Basa, Z. Czigány, I. Bársony, H. Y. Wang, and J. Volk, “Homogeneous transparent conductive ZnO:Ga by ALD for large LED wafers,” Appl. Surf. Sci. 379, 304–308 (2016). [CrossRef]
5. S. Yin, J. Li, M. M. Shirolkar, M. Li, and H. Wang, “Oxygen interstitial mediated effective doping of Al in ZnO:Al films prepared by magnetron sputtering,” Mater. Lett. 179, 146–149 (2016). [CrossRef]
6. S. B. Zhang, S. H. Wei, and A. Zunger, “Intrinsic n -type versus p -type doping asymmetry and the defect physics of ZnO,” Phys. Rev. B 63(7), 075205 (2001). [CrossRef]
7. F. Sun, C. X. Shan, B. H. Li, Z. Z. Zhang, D. Z. Shen, Z. Y. Zhang, and D. Fan, “A reproducible route to p-ZnO films and their application in light-emitting devices,” Opt. Lett. 36(4), 499–501 (2011). [CrossRef] [PubMed]
8. X. Chen, Z. Zhang, B. Yao, Y. Zhang, Y. Gu, P. Zhao, B. Li, and D. Shen, “The effect of boron on the doping efficiency of nitrogen in ZnO,” J. Alloys Compd. 672, 260–264 (2016). [CrossRef]
9. A. Chen, H. Zhu, Y. Wu, M. Chen, Y. Zhu, X. Gui, and Z. Tang, “Beryllium-Assisted p-Type Doping for ZnO Homojunction Light-Emitting Devices,” Adv. Funct. Mater. 26(21), 3696–3702 (2016). [CrossRef]
10. X. H. Pan, J. Jiang, Y. J. Zeng, H. P. He, L. P. Zhu, Z. Z. Ye, B. H. Zhao, and X. Q. Pan, “Electrical and optical properties of phosphorus-doped p-type ZnO films grown by metalorganic chemical vapor deposition,” J. Appl. Phys. 103(2), 023708 (2008). [CrossRef]
11. W. J. Lee, J. Kang, and K. J. Chang, “Defect properties andp-type doping efficiency in phosphorus-doped ZnO,” Phys. Rev. B 73(2), 024117 (2006). [CrossRef]
12. S. Limpijumnong, S. B. Zhang, S. H. Wei, and C. H. Park, “Doping by large-size-mismatched impurities: the microscopic origin of arsenic- or antimony-doped p-type zinc oxide,” Phys. Rev. Lett. 92(15), 155504 (2004). [CrossRef] [PubMed]
13. H. S. Kang, G. H. Kim, D. L. Kim, H. W. Chang, B. D. Ahn, and S. Y. Lee, “Investigation on the p-type formation mechanism of arsenic doped p-type ZnO thin film,” Appl. Phys. Lett. 89(18), 181103 (2006). [CrossRef]
14. F. X. Xiu, Z. Yang, L. J. Mandalapu, D. T. Zhao, and J. L. Liu, “Photoluminescence study of Sb-doped p-type ZnO films by molecular-beam epitaxy,” Appl. Phys. Lett. 87(25), 252102 (2005). [CrossRef]
15. U. Wahl, J. G. Correia, T. Mendonça, and S. Decoster, “Direct evidence for Sb as a Zn site impurity in ZnO,” Appl. Phys. Lett. 94(26), 261901 (2009). [CrossRef]
16. J. K. Dangbégnon, K. Talla, and J. R. Botha, “Effect of the annealing environment on the optical properties of ZnO/GaAs grown by MOCVD,” J. Lumin. 131(12), 2457–2462 (2011). [CrossRef]
17. P. Biswas, S. Kundu, and P. Banerji, “A study on electrical transport vis-à-vis the effect of thermal annealing on the p-type conductivity in arsenic-doped MOCVD grown ZnO in the temperature range 10–300K,” J. Alloys Compd. 552, 304–309 (2013). [CrossRef]
18. W. Lee, M. C. Jeong, and J. M. Myoung, “Catalyst-free growth of ZnO nanowires by metal-organic chemical vapour deposition (MOCVD) and thermal evaporation,” Acta Mater. 52(13), 3949–3957 (2004). [CrossRef]
19. C. R. Bayliss and D. L. Kirk, “The compositional and structural changes that accompany the thermal annealing of (100) surfaces of GaAs, InP and GaP in vacuum,” J. Phys. D Appl. Phys. 9(2), 233–244 (1976). [CrossRef]
20. H. P. Sun, X. Q. Pan, X. L. Du, Z. X. Mei, Z. Q. Zeng, and Q. K. Xue, “Microstructure and crystal defects in epitaxial ZnO film grown on Ga modified (0001) sapphire surface,” Appl. Phys. Lett. 85(19), 4385–4387 (2004). [CrossRef]
21. P. Biswas, P. Nath, D. Sanyal, and P. Banerji, “An alternative approach to investigate the origin of p-type conductivity in arsenic doped ZnO,” Curr. Appl. Phys. 15(10), 1256–1261 (2015). [CrossRef]
22. Y. R. Ryu, T. S. Lee, and H. W. White, “Properties of arsenic-doped p-type ZnO grown by hybrid beam deposition,” Appl. Phys. Lett. 83(1), 87–89 (2003). [CrossRef]
23. D. K. Hwang, H. S. Kim, J. H. Lim, J. Y. Oh, J. H. Yang, S. J. Park, K. K. Kim, D. C. Look, and Y. S. Park, “Study of the photoluminescence of phosphorus-doped p-type ZnO thin films grown by radio-frequency magnetron sputtering,” Appl. Phys. Lett. 86(15), 151917 (2005). [CrossRef]
