Open Access
Add to BibSonomy
Abstract
In this study, a ZnO/diamond structure ultraviolet (UV) photodetector was fabricated and investigated. ZnO films with thickness of 50 and 100 nm were deposited on half of diamond substrates by sputtering technique. Then, electrodes were patterned on ZnO and diamond areas to form photodetectors. The photocurrent gain in the UV region has been strongly influenced by ZnO film. ZnO films with thickness of 50 and 100 nm on diamond substrates reaches 14.3 and 308 A/W, respectively. Both of peak responsivities were located at 270 nm. Additionally, two shoulder peaks around 240 nm and 290 nm were observed for ZnO/diamond photodetector, which may stem from diamond and ZnO, respectively.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
It is noticeable that photodetectors operating in the UV region can be applied in ozone layer monitoring, high temperature flame and fire detection, missile warning system, UV astronomy, and especially in cooling devices [1]. The ideal photodetector should generally satisfy the 5S requirements of high sensitivity, high signal-to-noise ratio, high spectral selectivity, high speed, and high thermal stability [2,3]. UV photodetectors deployed for these purposes can be made using diamond owning to its 5.47 eV band gap, highest thermal conductivity, high radiation resistance and good temperature stability [3]. Moreover, improved single crystal diamond characteristics can offer the highest figure-of-merit for device performance [4]. However, high recombination between holes and electrons limits development.
ZnO, an II–VI group semiconductor, exhibits n-type conduction due to oxygen deficiency or excess of zinc [5,6]. Along with that, it has direct wide bandgap (3.37 eV), high transparency (>80%) in the visible wavelength region, large excitation binding energy (60 meV) and large saturation velocity [7]. Therefore, ZnO UV detectors have many important applications for satellite-based missile plume detection, in situ combustion monitoring gas, air quality monitoring, gas sensing, accurate measurement of radiation for the treatment of UV-irradiated skin, etc [8]. However, increased performance in the UV region is still a key issue for ZnO because ZnO photodetectors suffer heat issues due to emission of large amounts of heat in a small area during commercial applications [9].
In this paper, the photodetector based on the single crystal diamond and ZnO was investigated. The detection response can be further increased by diamond combined with ZnO. In addition, variety of wavelength under UV illumination and voltages has been revealed and analyzed.
2. Experiment
200 nm undoped single crystal diamond homoepitaxial layer was grown on high pressure and high temperature Ib-type diamond substrate by microwave plasma chemical vapor deposition. The conditions about CH4/H2 ratio, gas pressure, growth temperature, and microwave power were maintained at 5%, 85 Torr, 1000°C and 1000 W, respectively. Afterwards, these samples were oxidized through boiling in the HNO3/H2SO4 mixed acid. Next, standard photolithography was used to cover half of the diamond, then, 50 nm thick ZnO film was deposited on the uncovered diamond surface by reactive magnetron sputtering technique. Sputtering parameters such as working gas, pressure, power and time were Ar, 2 Pa, 120 W, and 5 minutes, respectively. After sputtering, two parts on diamond surface were achieved by lift-off technique: one is diamond surface, and the other is diamond covered by ZnO film. At last, the Ti/Pt/Au interdigitated electrodes were fabricated by photolithography and electron beam evaporation on both ZnO and diamond surface to form ZnO/diamond photodetector and diamond photodetector, as shown in Fig. 1. The total active area was about 0.0961 mm2. Additionally, ZnO/diamond photodetector with 100 nm thick ZnO was also fabricated to investigate effect of ZnO thickness.
The composition of ZnO film was analyzed by X-ray photoelectron spectroscopy thermos fisher ESCALAB Xi + . The surface of diamond and ZnO films was characterization by atomic force microscope INNOVA. An Agilent B1505A power device analyzer and a 1000 W xenon light source with a monochromator calibrated by a UV-enhanced Si detector were utilized to characterize the optoelectronic performance of the photodetectors.
3. Results and discussion
Figures 2(a) and 2(b) shows 1µm × 1µm AFM images of the diamond epitaxial layer and ZnO film surface, respectively. The surface roughness (Ra) of diamond film is 0.712 nm. Obviously, there are many particles on the surface of ZnO and the roughness (Ra) is 20 nm.
Figure 3(a) shows XPS full spectrum of 50 nm ZnO/diamond photodetector. The test area is located at the interface of ZnO film and diamond. Figure 3(b) illustrates that the peak of carbon C1s is located at 283 eV from diamond sp3 bond. As shown in the Fig. 3(c), the peaks of oxygen are located at 530.7 eV and 528.8 eV, corresponding to O1s level, which are mainly derived from zinc oxide and oxygen terminated diamond surface, respectively. Figure 3(d) presents the XPS spectrum of Zn 2p core level. It can be seen that there are two peaks at 1020.02 eV and 1044.22 eV, corresponding to Zn 2P3/2 and Zn 2P1/2 level, respectively. In addition, the peak of gold element is located at 84 eV Au 4f. It is noted that combination of ZnO and diamond is used to fabricate MSM photodetector.
Fig. 3. (a) XPS full spectrum of 50 nm thick ZnO film on diamond. (b) XPS spectrum of C 1S core level. (c) XPS spectrum of O 1S core level. (d) XPS spectrum of Zn 2p core level.
Figure 4 illustrates the IV characteristics of 50 nm ZnO/diamond photodetector and diamond photodetector in dark condition and under the illumination of 220, 270, 330, and 660 nm light respectively. In dark condition, as shown in Fig. 4(a), currents at 30 V bias for 50 nm ZnO/diamond photodetector and diamond detector are 0.482 pA and 2.55 nA, respectively. When illuminated under 220 nm, 270 nm, 330 nm and 660 nm light, 50 nm ZnO/diamond photodetector currents get to 3.98×10−7, 3.912×10−6, 6.22×10−6 and 4.10×10−8 A. In comparison to ZnO/diamond photodetector, as shown in Fig. 4(b), currents for diamond photodetector are 2.02×10−10, 3.11×10−10, 4.49×10−10 and 5.6×10−11 A, respectively. The UV/visible rejection ratios at bias 30 V on 50 nm ZnO/diamond photodetector for 220/400 nm and 270/400 nm are 350.63 and 905.06. Therefore, it should be noted that this photocurrent gain is responsible for existence of ZnO.
The responsivity is calculated by R = Iph/A×P, where R is the responsivity. Iph is pure photocurrent. A is active area of the photodetector. P is light power intensity [10]. Spectral response of 50 nm ZnO/diamond photodetector and diamond photodetector at 10 V bias are illustrated in Fig. 5(a). The peak of diamond photodetector is located at 220 nm with a value of 0.004 A/W, by contrast, 50 nm ZnO/diamond photodetector peak responsivity at 270 nm is measured up to 4.09 A/W. It is thought that combination of diamond with ZnO enhances responsivity and extends the detection range. Interestingly, other shoulder peaks for 50 nm ZnO/diamond photodetector at 290 nm and 240 nm were observed. Further exploration was taken under different applied bias, as shown in Fig. 5(b). It can be seen that when the applied bias increased, 240 nm and 290 nm shoulder peak become more distinct. In addition, one unusual behavior is that the responsivity gap between 20 V and 30 V was larger than that between 20 V and 10 V. This may be originated from saturation carrier effect under higher bias.
Fig. 5. (a) Spectral response of 50 nm ZnO/diamond photodetector and diamond photodetector at 10 V bias. (b) Spectral response of 50 nm ZnO/diamond photodetector at different bias voltages.
The shoulder peaks at 240 and 290 nm and main peak at 270 nm are related to diamond excitons generation, diamond nitrogen defect level response and ZnO response, respectively. Figure 6 reveals energy band diagram of ZnO/diamond, where bandgap of ZnO is 3.37 eV and that of diamond is 5.47 eV. Besides, electron affinity of O-diamond (1.7 eV) is much less than that of ZnO (4.2 eV) [11,12]. Therefore electrons transfer from diamond to ZnO, whereas, holes inject from ZnO to diamond which was main injection current. When light irradiation was applied on the photodetector, considerable photo-electrons and holes generated in the ZnO film at first and these carriers were collected by electrodes [13], resulting in 290 nm peak. As the energy of incident light increased, large amount of electrons and holes from diamond participated in, contributing to the peak around 270 nm. Relating with previous work [14,15], the peak is attributed to sub-bandgap response which originates from the charged nitrogen regarded as deep trap for electrons in diamond layer under UV illumination. In fact, 240 nm is approaching to the exciton recombination emission wavelength (235 nm) of diamond [13]. When incident light wavelength is shorter than 240 nm, interaction between photon and diamond will generate excitons. These excitons have a large recombination ratio, which leads to less collected carriers by electrodes. Therefore, rapid decline in responsiveness is associated with incident light wavelength shorter than 240 nm. With applied bias increasing, electric field becomes larger and more excitons can be gathered that results in more apparent of 240 nm shoulder peak. Moreover, in contrast to diamond photodetector, collaboration effect between ZnO and diamond strengthens the separation of photogenerated carriers, which is suggested as the reason for responsiveness improvement.
In order to demonstrate the role of diamond in ZnO/diamond photodetector, a 50 nm ZnO film deposited on the quartz substrate is used to form a contrast. The technology and parameters are identical with fabrication of 50 nm ZnO/diamond photodetector. The responsivities of ZnO/diamond photodetector and ZnO photodetector at 10 V are illustrated in Fig. 7. For the ZnO photodetector, the responsivities are 0.000224 and 0.000470 A/W with illumination wavelength of 270 nm and 400 nm. The UV/visible rejection ratios for 270/400 nm is 4.76, and the highest responsivity value is reached at 310 nm (0.000416A/W). There is a weaker peak at 360 nm which is cutoff wavelength. However, the responsivities of ZnO/diamond photodetector are 4.09 and 0.00145 A/W with illumination wavelength of 270 nm and 400 nm, and the UV/visible rejection ratios at bias 10 V on 50 nm ZnO/diamond photodetector for 270/400 nm are 2820.6. The 290 nm response peak is attributed to superimposition between diamond and ZnO responsivity. Diamond films as for an additional trapping centers allows more free electrons to be easily injected from conduction band of the diamond to that of ZnO film because diamond film has higher conduction and valence band edges than ZnO film. Based on these results, the ZnO/diamond photodetector can greatly improve responsiveness especially in the UV region.
In order to clarify if the thickness of ZnO would take effect on the photodetector performance, the responsivity of 100 nm ZnO/diamond photodetector was also calculated and shown in Fig. 8. Similar to 50 nm ZnO/diamond photodetector, the highest responsivity was measured at 270 nm and as the voltage increases, 240 nm peak become more obvious. However, responsivity of 100 nm ZnO/diamond photodetector has been significantly improved compared to that of 50 nm ZnO/diamond photodetector. The 270 nm responsivity of 50 nm ZnO/diamond photodetector at 30 V is 14.3 A/W, while 100 nm ZnO/diamond photodetector responsivity attains 308 A/W. In addition, spectral range solar-blind region has been raised to 400 nm, which indicates that increase of ZnO film thickness widen spectral response range.
4. Conclusion
In summary, photocurrent gain can be obtained by single crystal diamond with ZnO film. There exist three peaks around 240, 270 and 290 nm, which are ascribed to diamond excitons generation, diamond nitrogen defect level response and ZnO response, respectively. In contrast with diamond photodetector, 50 nm ZnO/diamond shows better performance. The photoresponse is three orders of magnitude higher than that of diamond photodetector. Moreover, the responsiveness is also significantly better than that of ZnO photodetector. In addition, thickness of ZnO film also has an impact on responsivity. The photocurrent gain has been significantly improved from 14.3 A/W to 308 A/W. The UV detector made up for ZnO film on diamond substrate is a promising device for high photoresponse.
Funding
National Natural Science Foundation of China (61705176); China Postdoctoral Science Foundation (2017M620449, 2018T111056); Dongguan Introduction Program of Leading Innovative and Entrepreneurial Talents; Innovation Capability Support Program of Shaanxi (2018PT-28, 2019PT-05).
References
1. A. BenMoussa, A. Soltani, U. Schühle, K. Haenen, Y. M. Chong, W. J. Zhang, R. Dahal, J. Y. Lin, H. X. Jiang, H. A. Barkad, B. BenMoussa, D. Bolsee, C. Hermans, U. Kroth, C. Laubis, V. Mortet, J. C. De Jaeger, B. Giordanengo, M. Richter, F. Scholze, and J. F. Hochedez, “Recent developments of wide-bandgap semiconductor based UV sensors,” Diamond Relat. Mater. 18(5-8), 860–864 (2009). [CrossRef]
2. R. D. McKeag, S. S. M. Chan, and R. B. Jackman, “Polycrystalline diamond photoconductive device with high UV-visible discrimination,” Appl. Phys. Lett. 67(15), 2117–2119 (1995). [CrossRef]
3. M. Liao, L. Sang, T. Teraji, M. Imura, J. Alvarez, and Y. Koide, “Comprehensive Investigation of Single Crystal Diamond Deep-Ultraviolet Detector,” Jpn. J. Appl. Phys. 51(9R), 090115 (2012). [CrossRef]
4. J. Liu, Y. Xia, L. Wang, P. Zhao, Q. Su, and W. Shi, “A New ZnO Film Photodetector Fabricated on Diamond Substrate,” in 2006 8th International Conference on Solid-State and Integrated Circuit Technology Proceedings. IEEE, pp. 950–952.
5. L. Luo, Y. Zhang, S. S. Mao, and L. Lin, “Fabrication and characterization of ZnO nanowires based UV photodiodes,” Sens. Actuators, A 127(2), 201–206 (2006). [CrossRef]
6. J. Liu, Y. Xia, L. Wang, Q. Su, and W. Shi, “Electrical characteristics of UV photodetectors based on ZnO/diamond film structure,” Appl. Surf. Sci. 253(12), 5218–5222 (2007). [CrossRef]
7. L. Qi, B. Sun, W. Wang, J. Zhang, H. Yang, and Y. Qi, “Microstructure of oxygen-deficient annealing-induced non-polar Zno films with extremely low electrical resistivity,” CrystEngComm 21(5), 791–795 (2019). [CrossRef]
8. R. R. Kumar, D. Punetha, R. Shukla, and S. K. Pandey, “Performance Optimization and Analysis of ZnO based Ultraviolet Photodiode,” in 2018 International Conference on Numerical Simulation of Optoelectronic Devices (NUSOD). IEEE, pp. 5–6.
9. J. Liu, Y. Xia, L. Wang, Q. Su, and W. Shi, “Effect of grain size on the electrical properties of ultraviolet photodetector with ZnO/diamond film structure,” J. Cryst. Growth 300(2), 353–357 (2007). [CrossRef]
10. Z. Liu, J.-P. Ao, F. Li, W. Wang, J. Wang, J. Zhang, and H.-X. Wang, “Photoelectrical characteristics of ultrathin TiO2/diamond photodetector,” Mater. Lett. 188, 52–54 (2017). [CrossRef]
11. A. Saravanan, B. R. Huang, J. Lin, G. Keiser, and I. N. Lin, “Fast Photoresponse and Long Lifetime UV Photodetectors and Field Emitters Based on ZnO/Ultrananocrystalline Diamond Films,” Chem. - Eur. J. 21(45), 16017–16026 (2015). [CrossRef]
12. S. Gao, S. Jiao, B. Lei, H. Li, J. Wang, and Q. Yu, “Efficient photocatalyst based on zno nanorod arrays/p-type boron-doped-diamond heterojunction,” J. Mater. Sci.: Mater. Electron. 26(2), 1018–1022 (2015). [CrossRef]
13. X. Chang, Y. Wang, X. Zhang, Z. Liu, J. Fu, S. Fan, R. Bu, J. Zhang, W. Wang, H. Wang, and J. Wang, “UV-photodetector based on NiO/diamond film,” Appl. Phys. Lett. 112(3), 032103 (2018). [CrossRef]
14. M. Liao and Y. Koide, “High-performance metal-semiconductor-metal deep-ultraviolet photodetectors based on homoepitaxial diamond thin film,” Appl. Phys. Lett. 89(11), 113509 (2006). [CrossRef]
15. M. Liao, Y. Koide, J. Alvarez, M. Imura, and J. P. Kleider, “Persistent positive and transient absolute negative photoconductivity observed in diamond photodetectors,” Phys. Rev. B 78(4), 045112 (2008). [CrossRef]
