ZnO homojunction UV photodetector based on solution-grown Sb-doped p-type ZnO nanorods and pure n-type ZnO nanorods

Abstract

A ZnO homojunction UV photodetector based on Sb-doped p-type ZnO nanorods (NRs) and n-type ZnO NRs was fabricated by a low temperature solution method. The fabricated homojunction shows well-defined rectifying characteristics, confirming the p-type conductivity of the Sb-doped ZnO NRs. Moreover, a high UV sensitivity of 3300% and a fast reset time to UV illumination are also achieved.

---

ZnO nanostructures have attracted increased attention in the burgeoning field of electronics and optoelectronics, such as solar cells, gas sensors and UV photodetectors,^1–6 which can be attributed to their low cost, strong radiation hardness, ease of manufacturing and numerous nanostructure forms.^7,8 However, such potential applications have been hindered by the difficulty in realizing reliable p-type ZnO. It is well-known that group-I (Li, Na, and K) atoms have been widely used as dopants to achieve p-type ZnO.^9–11 However, their small atomic radii may make them form interstitials to act as donors rather than acceptors.¹² Recently, group-V (N, P, As, and Sb) atoms were doped into ZnO, which have been broadly researched to realize stable p-type ZnO.^13–15 Among the group-V atoms, despite Sb has much larger atomic radius, Sb has been demonstrated as a promising dopant candidate to produce more stable p-type ZnO with higher hole concentrations, which can be attributed to the formation of Sb_Zn–2V_Zn complex with low formation energy and ionization energy rather than occupying the O site in ZnO.^16,17 Many methods have been reported to synthesize Sb-doped ZnO, such as vapor phase transport,¹⁸ pulsed laser deposition (PLD) and metal–organic chemical vapor deposition (MOCVD),^19,20 which require high temperatures or relatively expensive processes. Thus, it is urgent to introduce a suitable method with low cost, process simplicity and high yield of products to synthesize Sb-doped p-type ZnO. Furthermore, up to now, a very small amount of literature has reported the UV sensing performance based on the p-type Sb-doped ZnO/n-type ZnO axial homojunction.

In this work, we successfully synthesized Sb-doped p-type ZnO nanorods (NRs) using a low temperature hydrothermal method. The presence of Sb in the doped samples is confirmed by X-ray photoelectron spectroscopy (XPS) and secondary ion mass spectroscopy (SIMS). A p-type Sb-doped ZnO/n-type ZnO axial homojunction was fabricated and its I–V curve shows well-defined rectifying characteristic, confirming the p-type conductivity of the Sb-doped ZnO NRs. Furthermore, their potential applications as UV photodetector have also been explored. The UV sensitivity and the response time of the fabricated axial homojunction based UV photodetector are as high as 3300% and less than 16 s.

Pure and Sb-doped ZnO NRs were grown on silicon substrates via a facile hydrothermal method. Prior to the hydrothermal process, a 30 nm ZnO seed layer was spin-coated on substrates. The details of seed layer deposition are described in our previous work.²¹ Afterwards, the spin-coated substrates were placed on the surface of the nutrient solution and kept at 90 °C for 6 h in an oven. For pure ZnO NRs, the nutrient solution 1 consisted of 20 mM zinc acetate (Zn(CH₃COO)₂) and 20 mM hexamethylenetetramine (C₆H₁₂N₄, HMTA). For Sb-doped ZnO NRs, the nutrient solution 2 was prepared by adding required dopant solution into solution 1. The dopant solution was prepared by mixing equal molar glycolic acid (C₂H₄O₃) and sodium hydroxide (NaOH) in water to form sodium glycolate (NaC₂H₃O₃). The glycolate not only improves the Sb solubility but also helps the incorporation of Sb into ZnO.²² Then, antimony acetate (Sb(CH₃COO)₃) was added into this solution in a 1 : 12 molar ratio. The amount of Sb was 0.4%, 1.0% and 2.0% in molar ratio relative to zinc in our experiment. After the reaction was completed, the products were rinsed with deionized water and ethanol several times and finally annealed at 500 °C in air atmosphere for 2 h to activate the dopants.

The surface morphologies of the products were characterized using field emission scanning electron microscope (FE-SEM Hitachi S-4800) operated at 5 kV. X-ray diffraction (XRD, Bede D1) with a Cu Kα radiation source (λ = 1.54056 Å) was used to analyze the crystalline structure of the products. XPS was employed to study the elemental composition of the samples. The depth profile of Sb-doped ZnO NRs was investigated by SIMS using 250 eV O₂⁺ ions at normal incidence. Photoluminescence (PL) measurements of these products were carried out by an Edinburgh Instruments FLS-920 fluorescence spectrophotometer. A Xe lamp emitting at 325 nm was used as the excitation source at room temperature during PL measurement. The electrical properties of the samples were studied by current–voltage (I–V) measurements, which were carried out using a semiconductor parameter analyzer (Agilent E5270B). The UV photoresponse of the fabricated device was measured by illuminating the devices with a UVA-LED (60 mW cm⁻²).

Fig. 1a–c display the typical FE-SEM images of pure ZnO NRs and Sb-doped ZnO NRs grown on silicon substrates, respectively. All images show that both pure ZnO NRs and Sb-doped ZnO NRs are highly aligned, vertical and uniform. The hexagonal cross-section of these NRs is indicative of a wurtzite structure with c-axis growth direction. Moreover, the diameter and length of pure ZnO NRs are 80–100 nm and 1.73 μm (Fig. 1a). With the increase of Sb doping concentration, the length of NRs decreases dramatically, while the diameter keeps nearly unchanged (Fig. 1b and c). The observed change in aspect ratio after Sb doping can be attributed to the incorporated dopant, which provides an energetic barrier to crystal growth reported by Pradel et al.²³ Fig. 1d depicts the XRD patterns of pure ZnO NRs and Sb-doped ZnO NRs. Except for one peak from the silicon substrates, only one sharp peak at 34.5° is detected in all samples, corresponding to hexagonal wurtzite (002), which reveals that no Sb-related phases have been formed. Furthermore, the intensity of (002) peak decreases gradually with the increase of doping concentration. This phenomenon is attributed to the deteriorative crystallinity after the incorporation of Sb dopant into ZnO.

Fig. 1 Typical SEM images of (a) pure ZnO NRs, (b) 1.0% Sb-doped ZnO NRs, and (c) 2.0% Sb-doped ZnO NRs. The lengths of the scale bars are 100 nm for all images. (d) XRD patterns of pure ZnO NRs, 0.4% Sb-doped, 1.0% Sb-doped, and 2.0% Sb-doped ZnO NRs.

In order to confirm the presence of Sb in the doped samples, the XPS measurements were carried out on the pure ZnO NRs and Sb-doped ZnO NRs, as shown in Fig. 2a. It can be seen that there is only one peak at 530.2 eV in the pure ZnO NRs, which corresponds to O1s peak. For Sb-doped ZnO NRs, there are two peaks at 539.6 eV, corresponding to Sb3d_3/2, and at 530.2 eV, attributing to O1s and Sb3d_5/2.^24,25 To further investigate the element composition distribution in the Sb-doped ZnO NRs, a depth profile was carried out by SIMS measurements, as shown in Fig. 2b. Quantitative results could not be obtained due to the lack of reference samples for the SIMS measurement. It is evident that Sb has been well detected. Note that the content of dopant Sb has a little increase and fluctuation during the sputtering time from 0 to 320 s. In the initial growth stage of ZnO NRs, the Sb dopant solution is sufficient. Thus, more Sb can be incorporated into ZnO NRs during initial growth stage. With the further growth of ZnO NRs, the nutrient solution is consumed gradually which results in the decrease of Sb content. What's more, the Sb salt is not very compatible with the ZnO growth solution due to the low solubility. Some Sb salt may precipitate onto the substrate, which causes the increase of Sb at the interface. Therefore, it is concluded that above factors are responsible for the non-uniformity and fluctuation of dopant Sb in ZnO NRs. The observed feature confirms again that Sb has been successfully doped into ZnO by a facile hydrothermal method and post-annealing process.

Fig. 2 (a) XPS spectra of pure ZnO NRs and Sb-doped ZnO NRs. (b) SIMS depth profile of Sb-doped ZnO NRs grown on silicon substrate. (c) The PL spectra of pure ZnO NRs and Sb-doped ZnO NRs with different doping concentration. (d) Spectra highlighted in (c).

The optical property of the products was characterized by room temperature PL measurement. Fig. 2c shows the normalized PL spectra of the pure ZnO NRs and Sb-doped ZnO NRs with different doping concentration, and Fig. 2d shows the spectra highlighted of Fig. 2c. The pure ZnO NRs show a strong near band emission (NBE) at 377 nm and a weak broad visible emission. The NBE peak is attributed to the recombination of free excitons (FEs) through exciton–exciton collision in ZnO. For Sb-doped ZnO NRs with different doping concentration, it is obvious that the NBE peak gradually red-shifts to 382 nm compared with pure ZnO NRs and the PL intensity of visible emission becomes much stronger after Sb doping. Kao et al. have reported that the red-shift effect of the NBE peak in Sb-doped ZnO can be attributed to the recombination of FEs.²⁶ The red-shift effect also indicates that Sb doping can slightly reduce the band gap of ZnO. So far, for the visible emission, there has been no consensus on its PL mechanism even for pure ZnO. Many researchers suggest that intrinsic defects such as oxygen vacancies are responsible for the visible emissions. Thus, the much stronger visible emission in Sb-doped ZnO NRs indicates that more intrinsic defects have been introduced by Sb doping. Furthermore, Sb-related complex defects (Sb_Zn–2V_Zn) are formed after post-annealing process owing to the low formation energy and ionization energy.¹⁷ It is, therefore, concluded that Sb-related complex defects and intrinsic defects are responsible for the much stronger visible emission.

Fig. 3a shows the schematic diagram of the fabricated p-type Sb-doped ZnO/n-type ZnO axial homojunction, and Fig. 3b shows the corresponding cross-section SEM image. The NR growth process consists of two steps: the growth of an intrinsic n-type ZnO NRs section on FTO (SnO₂:F) substrate, followed by a p-type ZnO NRs section (via 1.0% Sb-doping). For the measurement of electrical and UV detection properties, a patterned Au electrode (∼100 nm thick) was deposited onto the p-type layer using a shadow mask through electron beam evaporation technique. Then sweeping voltage was applied through the Au/p-type/n-type/FTO device with measurements of current. Fig. 3c shows the typical I–V characteristics curve of the axial homojunction diode under dark and 365 nm UV illumination conditions. It can be seen that the fabricated p-type Sb-doped ZnO/n-type ZnO axial homojunction shows obvious asymmetric rectifying behavior and a negligible leakage current (28.3 μA) in the dark. The ohmic contact behaviors of Au and In to p-type and n-type layers are confirmed by the fairly linear I–V dependencies, as shown in the inset of Fig. 3c. The well-defined rectifying characteristic indicates that the Sb-doped ZnO NRs exhibit p-type conductivity. The origin of the p-type behavior in Sb-doped ZnO NRs is attributed to the replacement of Zn by Sb accompanying two Zn vacancies (Sb_Zn–2V_Zn), rather than occupying the O site in ZnO. Under UV light irradiation (365 nm), both the forward and reverse current increase obviously. We usually define the sensitivity of the UV photodetector as (I_photo − I_dark)/I_dark in percentage (where I_photo is the photocurrent and I_dark is dark current). It is found that the sensitivity could reach to 3300% at −3 V, indicative of high sensitivity of the fabricated axial homojunction UV photodetector. In our present work, the obtained UV sensitivity is much higher than single Sb-doped ZnO nanowire based UV photodetector (∼14%) and p-type Sb doped ZnO nanowires/n-type ZnO film based UV photodetector (∼100%) reported by other research groups.^15,26 The time-resolved UV photocurrent on/off measurement of the fabricated UV photodetector was operated at −3 V, as shown in Fig. 3d. It can be seen that the device shows a single exponential rise and a double exponential decay under illumination which can be attributed to the recombination of the electron–hole pairs and the readsorption of oxygen molecules. The growth time constant (τ_g) is 15.2 s. The 1st (τ_d1) and 2nd (τ_d2) decay time constants are 7.3 s and 20.3 s, respectively. Furthermore, four repeated cycles are displayed in Fig. 3d, in which the photocurrent is observed to be consistent and repeatable with no degenerate effect found during the detection process. Thereby, the p-type Sb-doped ZnO/n-type ZnO axial homojunction based UV photodetector can be one of the most promising candidates for the applications in UV detection field.

Fig. 3 (a) A schematic diagram of the p-type Sb-doped ZnO/n-type ZnO axial homojunction. (b) The cross-section SEM image of the Sb-doped ZnO NRs grown on the FTO substrate. (c) I–V characteristics of a typical p-type Sb-doped ZnO/n-type ZnO axial homojunction diode under dark and 365 nm UV illumination conditions, and the inset shows the ohmic contact behavior of Au and In/FTO to Sb-doped ZnO and pure ZnO, respectively. (d) The growth and decay of the photocurrent of the p-type Sb-doped ZnO/n-type ZnO axial homojunction under 365 nm UV light illumination.

The enhanced UV sensing performance of the fabricated axial homojunction UV photodetector can be explained by the photoconduction mechanisms as follows. In the dark conditions, the oxygen chemisorbs on the surface of ZnO NRs as O₂⁻ by capturing free electrons from n-type ZnO [O₂(g) + e⁻ → O₂⁻(ad)], leading to the reduction of conductivity of ZnO NRs by forming a depletion region near the surface of ZnO NRs. When the 365 nm UV light shines on the fabricated device, the electron–hole pairs are generated [hν → e⁻ + h⁺]. The photogenerated electron–hole pairs are effectively separated by the built-in electrical field at the interface of p–n homojunction. The positive holes are driven to the surface of ZnO NRs and oxidize the adsorbed oxygen molecules [O₂⁻(ad) + h⁺ → O₂(g)], inducing desorption of oxygen from the ZnO surface.²⁷ At the same time, the trapped electrons are released back to the conduction band to increase the conductivity. Consequently, the current under UV illumination is higher than that in the dark conditions.

In conclusion, we have synthesized p-type Sb-doped ZnO NRs by a low temperature hydrothermal method. The XPS and SIMS measurements confirm the incorporation of Sb into ZnO NRs. In addition, a red-shift effect of the NBE peak occurs in Sb-doped ZnO NRs compared to that of pure ZnO NRs in the PL spectra. We also fabricated the p-type Sb-doped ZnO/n-type ZnO axial homojunction diode and further investigated the application in UV photodetection. The fabricated p-type Sb-doped ZnO/n-type ZnO homojunction shows well-defined rectifying characteristic with high rectification factor. Furthermore, the fabricated UV photodetector exhibits a high UV sensitivity of 3300% and a fast reset time. It indicates that the p-type Sb-doped ZnO NRs act as one of promising candidates for electronic and optoelectronic devices in the future.

Acknowledgements

This work was supported by National Natural Science Foundation of China under Grant nos 51302244 and 91333203, and Zhejiang Provincial Natural Science Foundation of China under Grant no. LQ13E020001.

Notes and references

1. S. H. Ko, D. Lee, H. W. Kang, K. H. Nam, J. Y. Yeo, S. J. Hong, C. P. Grigoropoulos and H. J. Sung, Nano Lett., 2011, 11, 666–671.
2. Q. Zhang and G. Cao, Nano Today, 2011, 6, 91–109.
3. Z. Wen, L. P. Zhu, L. Li, L. W. Sun, H. Cai and Z. Z. Ye, Dalton Trans., 2013, 42, 15551–15554.
4. C. L. Hsu, Y. D. Gao, Y. S. Chen and T. J. Hsueh, Sens. Actuators, B, 2014, 192, 550–557.
5. W. Dai, X. H. Pan, S. S. Chen, C. Chen, Z. Wen, H. H. Zhang and Z. Z. Ye, J. Mater. Chem. C, 2014, 2, 4606–4614.
6. P. N. Ni, C. X. Shan, S. P. Wang, X. Y. Liu and D. Z. Shen, J. Mater. Chem. C, 2013, 1, 4445–4449.
7. Y. Jin, J. Wang, B. Sun, J. C. Blakesley and N. C. Greenham, Nano Lett., 2008, 8, 1649–1653.
8. D. C. Look, D. C. Reynolds, J. W. Hemsky, R. L. Jones and J. R. Sizelove, Appl. Phys. Lett., 1999, 75, 811.
9. Y. J. Zeng, Z. Z. Ye, W. Z. Xu, D. Y. Li, J. G. Lu, L. P. Zhu and B. H. Zhao, Appl. Phys. Lett., 2006, 88, 062107.
10. H. H. Zhang, X. H. Pan, Y. Li, Z. Z. Ye, B. Lu, W. Chen, J. Y. Huang, P. Ding, S. S. Chen, H. P. He, J. G. Lu, L. X. Chen and C. L. Ye, Appl. Phys. Lett., 2014, 104, 112106.
11. M. K. Gupta, N. Sinha and B. Kumar, J. Appl. Phys., 2011, 109, 083532.
12. J. Anderson and G. V. d. W. Chris, Rep. Prog. Phys., 2009, 2, 126501.
13. S. C. Su, X. D. Yang and C. D. Hu, Phys. Rev. B: Condens. Matter, 2011, 406, 1533–1535.
14. V. Vaithianathan, B. T. Lee and S. S. Kim, Appl. Phys. Lett., 2005, 86, 062101.
15. G. Wang, S. Chu, N. Zhan, Y. Lin, L. Chernyak and J. Liu, Appl. Phys. Lett., 2011, 98, 041107.
16. U. Wahl, J. G. Correia, S. Decoster and T. Mendonça, Phys. B, 2009, 404, 4803–4806.
17. S. Limpijumnong, S. B. Zhang, S. H. Wei and C. H. Park, Phys. Rev. Lett., 2004, 92, 155504.
18. S. H. Kim, A. Umar, Y. K. Park, J. H. Kim, E. W. Lee and Y. B. Hahn, J. Alloys Compd., 2009, 479, 290–293.
19. I. A. Palani, D. Nakamura, K. Okazaki, M. Higashihata and T. Okada, Mater. Sci. Eng., B, 2011, 176, 1526–1530.
20. J. W. Kang, Y. S. Choi, M. Choe, N. Y. Kim, T. Lee, B. J. Kim, C. W. Tu and S. J. Park, Nanotechnology, 2012, 23, 495712.
21. W. Dai, X. H. Pan, C. Chen, S. S. Chen, W. Chen, H. H. Zhang and Z. Z. Ye, RSC Adv., 2014, 4, 31969–31972.
22. W. Fei, S. Jung-Hun, B. Dylan, S. Jian, M. Hongyi, M. Zhenqiang, Z. Deyin, S. Yichen, Z. Weidong and W. Xudong, Nanotechnology, 2011, 22, 225602.
23. K. C. Pradel, W. Wu, Y. Zhou, X. Wen, Y. Ding and Z. L. Wang, Nano Lett., 2013, 13, 2647–2653.
24. I. A. Palani, K. Okazaki, D. Nakamura, K. Sakai, M. Higashihata and T. Okada, Appl. Surf. Sci., 2012, 258, 3611–3616.
25. W. E. Mahmoud, F. Al-Marzouki, S. Al-Ameer and F. Al-Hazmi, J. Appl. Crystallogr., 2012, 45, 182–185.
26. T. H. Kao, J. Y. Chen, C. H. Chiu, C. W. Huang and W. W. Wu, Appl. Phys. Lett., 2014, 104, 111909.
27. C. Soci, A. Zhang, B. Xiang, S. A. Dayeh, D. P. R. Aplin, J. Park, X. Y. Bao, Y. H. Lo and D. Wang, Nano Lett., 2007, 7, 1003–1009.

---