Large area transparent ZnO photodetectors with Au wire network electrodes

Abstract

Transparent electronics as a futuristic technology are fast growing and expanding beyond conventional optoelectronics. Here we report the fabrication of a large area, visibly transparent ultraviolet (UV) photodetector employing a laser deposited ZnO active layer in between branched Au wire networks, with the latter serving as electrodes due to being transparent in UV as well as visible regions. The Au wire networks were prepared using a crack templating method which is not only cost effective but also enables heterostructuring with simple processes. When compared to Ag contact pads, these wire network electrodes seem to enhance photocurrent collection efficiency resulting in high UV sensitivity at low response times. Importantly, the visible light transparency was as high as 80%.

---

Introduction

Emerging transparent functional devices such as organic light-emitting diodes,¹ thin film loudspeakers,² capacitors,³ batteries,⁴ solar cells,⁵ etc. have paved the way for next generation “see-through” devices. Transparent devices have become increasingly attractive due to their high impact in various areas such as displays, sensors, photodetectors, etc.^6,7 Transparency is not just meant for visual appeal but also to gain crucial visual space, in much the same way that glass partitions do in lieu of opaque walls. Taking photodetectors as an example, a “visible-blind” photodetector will have an extended practical usage when combined with the variety of devices which require exposure to visible light.⁸ In this work, we have explored the fabrication of a transparent UV detector.

UV photodetectors play important roles in our day-to-day lives, from pollution monitoring to water purification.^9–11 Most commercially available photodetectors are based on semiconducting materials such as Si, GaN, InGaAs, etc. which are opaque and brittle, thus restricting their possible usage in flexible transparent detectors.¹² Further, such conventional detectors possess an active area of ∼1 mm², which serves essentially as a point detector. On the other hand, for large area detection, the active material should be relatively inexpensive yet maintain higher efficiencies. In other words, the material should be optically transparent with high photoconductivity in the UV region. In this direction, a few attempts have been made in recent literature to fabricate transparent photodetectors with materials such as SnO₂, TiO₂, and InGaZnO as well as their hybrids.^13–18 However, the demonstrated device areas are rather small. We aimed at producing large area photodetectors using a ZnO layer as the active material as it exhibits high transparency in the visible region and high absorbance in the UV region while being amenable for easy fabrication.¹⁹

Although the active materials for UV detection may be visibly transparent, the contact electrodes in the device obstruct visibility, which is yet another challenge. Consequently, a transparent UV detector cannot be realized as the electrodes occupy a considerable area in these devices. Thus, it is important that the electrodes themselves are made of a transparent conductor. Tin-doped indium oxide (ITO) is a well-known transparent conducting material widely employed in optoelectronic devices. Yu et al. made use of ITO as the contact electrode for a ZnO nanowire photodetector.⁸ However, the shortcomings associated with ITO such as its brittleness, high price and instability under high-temperature processing conditions have led to the usage of alternative transparent electrodes such as graphene, Ag nanowires etc.²⁰ The latter, however, pose other issues such as high surface roughness and contact resistance and demand further remedial steps such as mechanical pressing and thermal annealing.^21,22 In this work, we have employed fine Au wire meshes^23,24 with seamless junctions as transparent electrodes in an UV photodetector.

Results and discussion

The fabrication steps are shown in Fig. 1a and S1a.† An acrylic resin based crackle precursor (CP) was spin coated on glass substrate resulting in a dried layer of an interconnected crackle network with widths in the range of 5–15 μm (Fig. 1a(i) and S1b†). Using a physical vapor deposition process, Au was deposited on the crackle template. The CP layer was then removed via washing with chloroform, resulting in a fine interconnected Au network on the glass substrate (Fig. 1a(ii) and S1c†). The sheet resistance of the formed Au network was found to be ∼8 Ω □⁻¹ and it had a transmittance of ∼90% in the entire range, 200–3000 nm (Fig. S2†). A part of the Au network was then masked for use as the bottom electrical contact and the remaining part was coated with ZnO via pulsed laser deposition (PLD) with the glass substrate held at room temperature (Fig. 1a(iii)). ZnO layers of varying thickness, ∼100–250 nm, were then deposited on Au mesh coated substrates. The top Au network was then deposited using similar steps detailed in Fig. 1a(iv). SEM micrographs of the assembled device, (Fig. S1d†) along with optical micrographs (Fig. S1e†) and EDS elemental mapping (Fig. S1f–i†), clearly reveal the presence of top and bottom Au mesh electrodes. The ZnO layer being transparent is not discernible but the EDS spectrum confirms the presence of a uniform ZnO layer (Fig. S1 and S3a†). An XRD pattern of the as-deposited ZnO film on the glass substrate is shown in Fig. S3b.† The optical profilometric image in Fig. S3c† demonstrates that the as-deposited ZnO layer submerges the bottom Au wire mesh uniformly at a layer thickness of 100 nm. This fabricated Au mesh/ZnO/Au mesh (henceforth Au/ZnO/Au) photodetector is found to be highly transparent in the visible range as is evident from the photographs and the transmittance spectra shown in Fig. 1b and c. The presence of the fine Au wire network makes the device transparent in both the UV and visible regions and enables the illumination to fall upon the entire ZnO layer from both sides, as appropriate to the context. The transmittance spectrum of the device on the quartz substrate clearly indicates that absorption below 400 nm is mainly due to the ZnO layer (Fig. S4†) as the metal mesh and quartz are transparent at this wavelength (see Fig. S2†).

Fig. 1 (a) Schematic illustration of the fabrication steps for the UV photodetector. (b) Photographic images of the obtained photodetectors. (c) Transmittance spectra of the photodetecting device, along with the Au top and bottom layers, with a varying ZnO thickness (in nm). The inset shows the schematic of the photodetector.

I–V measurements have been performed on the fabricated device in the presence and absence of UV-light (365 nm), as shown in Fig. 2. A typical I–V curve for a photodetector with 100 nm of ZnO (see Fig. 2a) shows that, upon UV illumination, the photocurrent increases by ∼6.5 times that of the dark current. The I–V curves are symmetrical and quite linear,²⁵ indicating that the carrier density in ZnO is high and the Au–ZnO contacts are ohmic. The photoinduced change in the current that is observed is in line with the known behaviour of ZnO.²⁶ The ZnO layer absorbs the incident photons and generates electron–hole pairs (Fig. 2b). These photogenerated carriers in the ZnO film are transported through the fine Au network and are finally collected under the external bias voltage. As the process of UV detection strongly depends on the carrier separation and collection, the presence of the fine Au network enhances the net efficiency of the device by increasing the transparency as well as the surface area involved in charge collection which is clearly evident from the I–V characteristics obtained from the thin film of ZnO with Ag paint electrodes (Fig. S5†). The band bending at the heterojunction suppresses exciton recombination in the presence of light, thereby encouraging discharge as well as the acceleration of trapped electrons.²⁷ The photocurrent response of the detectors was also found to scale linearly with the applied bias. The barrier height for Au/ZnO (100 nm)/Au was calculated to be 0.77 eV (see note S1†).

Fig. 2 (a) Typical I–V characteristics measured in the dark and under UV light illumination; the inset shows a schematic of the photodetection process. (b) Schematic of the band diagram. (c) Percentage change in the current with illumination with respect to the different thicknesses of the ZnO film in the photodetector. (d) Current vs. time plot for various ON and OFF positions for the photoconductive device with 125 nm thick ZnO at an applied bias of 1 V.

Detectivity is one of the important photodetector characteristics of merit and is given by the expression²⁷

D = (J_ph/AE_i)/(2qJ_d)^1/2 (1)

where J_ph and J_d are the photocurrent (125 μA) and dark current (10 μA), respectively, E_i is the light intensity (250 mW), q is the coulombic charge (1.6 × 10⁻¹⁹ coulomb) and A is the active area exposed to UV light (0.5 × 2 cm²). Another important parameter is responsivity.

Responsivity (A W⁻¹) is defined by

R = J_ph/AE_i (2)

For the given device, a detectivity of ∼10¹⁰ Jones (cm Hz^1/2 W⁻¹) and responsivity of ∼10⁻² A W⁻¹ were estimated. The obtained responsivity value is comparable to the literature value (0.0013 A W⁻¹) obtained with a ZnO nanoparticle/RGO composite with an active area of 10 × 5 mm².²⁸

Similar I–V characteristics were measured for different thicknesses of the ZnO active layer, as shown in Fig. S6†, and the percentage change in photocurrent (also known as sensitivity), (I–I₀)/I₀% where I₀ is the forward current in the dark at a forward bias of 5 V and I is the current under illumination at 5 V, is presented in Fig. 2c. This also gives the photoconductive gain (G, calculated from the ratio of the photocurrent to the dark current at the same bias voltage) value as 7. An Au electrode in the form of a network facilitates a reduced gap between the two electrodes shortening the carrier transit time (Fig. S5†). Consequently, a long lifetime and short transit time of photogenerated carriers can cause a photoconductive gain. For the device with a ZnO layer thickness of 100 and 125 nm, the percentage change value was as high as 630 and 290, respectively, whereas the value decreased to 73, 41 and 9, for increasing thicknesses of 190, 220 and 250 nm respectively (see Fig. 2b). These sensitivity values are nearly 3 times higher compared to the values obtained with Ag paint contacts (Fig. S5†). The thickness of the ZnO film is known to influence the overall properties significantly²⁹ accompanied by morphological changes as well. The defect states also increase with increasing film thickness, while the optical band gap is known to decrease only marginally.^30,31 With an increased thickness the higher energy photons are absorbed in or near the surface of the semiconductor, where the photogenerated electron–hole pairs recombine at a much higher rate than in the bulk region. This leads to a lower contribution of photogenerated electrons and holes to the photocurrent. Hence, the performance of the photodetector device decreases with increasing film thickness.

Another important parameter of photodetectors is their response speed. Hence, time-dependent studies of photoresponse were recorded to study the response as well as recovery times. The transient response of the present photodetectors was measured by turning on and off an UV LED with a centre peak wavelength at 365 nm, as shown in Fig. 2d. The response time, τ_r, (the time required to reach 63% of the maximum photocurrent) and recovery time (the time required to fall to 37% of the maximum photocurrent)³² were measured to be around 3.3 s and 20 s, respectively. These response and recovery times for the present device are much smaller compared to those based on polycrystalline ZnO photodetectors, where the response times range between minutes to hours.^33–35 It is worth mentioning that the time constant for the rise time is always faster than the fall time, again suggesting the involvement of traps and other defect states in this process. Such a transient response is expected for large area UV photodetectors that are fabricated from wide band gap metal oxides such as ZnO and is very desirable for their practical applications in devices such as light-wave communications or optoelectronic switches. The increase in light-induced conductivity is extremely sensitive to ultraviolet light and is also a reversible phenomenon (Fig. 2d). The reproducible response observed implies the excellent stability of the device.

Flexibility is an attractive feature of a device. Using similar design principles, we were motivated to fabricate a transparent and flexible photodetector, for which we used PET as the substrate in place of glass (both exhibit similar absorption for 365 nm) (Fig. 3). Although the net photodetection effect was minimal even with 100 nm ZnO thin film (∼10% change) perhaps due to different interfacial phenomena between PET and the ZnO heterostructures, it does give way to a new possibility of these materials being used as large area transparent and flexible photodetectors with optimized thickness. These devices can be employed as transparent windows for atop devices which need visible light. The nature of the ZnO active material makes the device transparent in the entire range of 350 nm and beyond. Further optimization of the ZnO morphology may lead to transparent strain sensors as the film does show a small change in resistance during bending to different radii, as shown in Fig. 3b.

Fig. 3 (a) Optical transmittance characteristics of the photodetector device on a PET substrate. The insets are a schematic of the fabricated photodetector (left) and a real photographic image of the device (right). (b) The change in current of the flexible detector on a PET substrate when bent to various radii under a fixed bias of 5 V.

Conclusions

In this work, we have successfully fabricated UV photodetectors with a 2 cm² active area with a visible light transparency of 80%. The fabrication involved depositing Au wire networks sandwiching a laser ablated ZnO film in between, while using glass as a substrate. Such heterostructuring has been possible thanks to the crackle template method pioneered in this laboratory. The main advantage of using metal wire networks as electrodes has been their high degree of transparency, from UV to beyond the visible range, as well as the enhanced electrical contact with the active layer. The fabricated devices showed typically a sensitivity of 630%, detectivity of ∼10¹⁰ Jones and responsivity of ∼10⁻² A W⁻¹. Using PET as a substrate in place of glass, a flexible photodetector was also made and demonstrated. It is noteworthy that the devices fabricated in this study do not require a complex lithography process for electrode fabrication and the resulting electrodes exhibit transparency in spite of being on the top and bottom of the active layer.

Experimental

The Au/ZnO/Au based UV photodetector was fabricated on glass substrates (2 × 2 cm²) after stepwise cleaning via ultrasonication in detergent, acetone, distilled water and isopropanol. A schematic of the different steps involved in the fabrication of the photodetector is presented in Fig. 1a. An acrylic resin based crackle precursor (Ming Ni Cosmetics Co., Guangzhou, China) diluted to 0.5 g mL⁻¹ was spin coated (2000 rpm, Technoscience, Bangalore) on a glass substrate to obtain a crackle template.^36–38 Over the template, Au was thermally evaporated (Hindvac, India) at a base pressure of 10⁻⁶ Torr, which was followed by template lift-off using chloroform. The formed Au wire network was highly transparent and conducting. Later, a ZnO layer was deposited using the pulsed laser deposition (PLD) technique at an oxygen partial pressure of 2 × 10⁻⁵ Torr over the Au wire network. A frequency tripled pulsed Nd: YAG laser (Quanta-Ray GCR-170, Spectra-Physics, USA) was used to ablate the zinc oxide target (obtained via thermal decomposition of zinc acetate at 1200 °C, see Fig. S7 in ESI†), which was fixed at 45° to the laser beam, with a pulse width of ∼5 ns and a repetition rate of 10 Hz. The laser beam having ∼180 mJ of energy per pulse was focused onto the target using a convex lens that had a 50 cm focal length through a quartz window. In order to increase the width of the plasma plume and also to avoid ablation at a point, the target holder was driven by a linear-rotary motion controlled by the stepper motor. The Au wire network/glass was held at ∼4 cm from the target and deposition was carried out at room temperature.

Characterization

The transmittance spectrum of the device was measured using an UV-vis spectrophotometer (Perkin-Elmer Model Lambda 900 UV/visible/near-IR spectrophotometer). Scanning electron microscopic (SEM) images of the photodetector were taken using a Nova NanoSEM 600 instrument (FEI Co., The Netherlands). Using an EDAX Genesis instrument (Mahwah, NJ) attached to the SEM column, energy-dispersive spectroscopy (EDS) analysis was performed. An optical microscope from Laben, India was used for this study. An UV LED from Hamamatsu (C10559) was used for the photodetection studies. The thickness of the material was measured using a Wyko NT9100 Optical Profiling system (Bruker, USA). The typical I–V characteristics of the photodetector were measured using a Keithley source meter (236). The sheet resistance of the Au network was measured using a 4-point probe Station from Techno Science Instruments, India. XRD measurements of the ZnO films were performed on a Miniflex (Rigaku, Japan).

Acknowledgements

Authors thank Prof. C. N. R. Rao for his encouragement. The financial support from DST, India is gratefully acknowledged. SK acknowledges DST-INSPIRE for the fellowship. SS would like to acknowledge DST-INSPIRE, UGC-FRP program and JNCASR.

Notes and references

1. G. Gu, V. Bulović, P. E. Burrows, S. R. Forrest and M. E. Thompson, Appl. Phys. Lett., 1996, 68, 2606–2608.
2. C. Keplinger, J.-Y. Sun, C. C. Foo, P. Rothemund, G. M. Whitesides and Z. Suo, Science, 2013, 341, 984–987.
3. R. Gupta, K. D. M. Rao and G. U. Kulkarni, RSC Adv., 2014, 4, 31108–31112.
4. Y. Yang, S. Jeong, L. Hu, H. Wu, S. W. Lee and Y. Cui, Proc. Natl. Acad. Sci. U. S. A., 2011, 108, 13013–13018.
5. Y. Zhao, G. A. Meek, B. G. Levine and R. R. Lunt, Adv. Opt. Mater., 2014, 2, 606–611.
6. A. Facchetti and T. J. Marks, Transparent Electronics, John Wiley & Sons, Ltd, 2011, pp. 443–448.
7. D. Chen, Z. Liu, B. Liang, X. Wang and G. Shen, Nanoscale, 2012, 4, 3001–3012.
8. H. Yu, E. Ali Azhar, T. Belagodu, S. Lim and S. Dey, J. Appl. Phys., 2012, 111, 102806.
9. T. V. Blank and Y. A. Gol’dberg, Semiconductors, 2003, 37, 999–1030.
10. S. Dali, Y. Mingpeng, L. Jie and S. Shayla, Nanotechnology, 2013, 24, 295701.
11. H.-G. Li, G. Wu, M.-M. Shi, L.-G. Yang, H.-Z. Chen and M. Wang, Appl. Phys. Lett., 2008, 93, 153309.
12. Y.-L. Liu, C.-C. Yu, K.-T. Lin, T.-C. Yang, E.-Y. Wang, H.-L. Chen, L.-C. Chen and K.-H. Chen, ACS Nano, 2015, 9, 5093–5103.
13. C. Gao, X. Li, X. Zhu, L. Chen, Y. Wang, F. Teng, Z. Zhang, H. Duan and E. Xie, J. Alloys Compd., 2014, 616, 510–515.
14. S. Huang, C. F. Guo, X. Zhang, W. Pan, X. Luo, C. Zhao, J. Gong, X. Li, Z. F. Ren and H. Wu, Small, 2015, 11, 5712–5718.
15. H. K. Li, T. P. Chen, S. G. Hu, X. D. Li, Y. Liu, P. S. Lee, X. P. Wang, H. Y. Li and G. Q. Lo, Opt. Express, 2015, 23, 27683–27689.
16. X. Liu, X. Liu, J. Wang, C. Liao, X. Xiao, S. Guo, C. Jiang, Z. Fan, T. Wang, X. Chen, W. Lu, W. Hu and L. Liao, Adv. Mater., 2014, 26, 7399–7404.
17. M. Patel, H.-S. Kim and J. Kim, Adv. Electron. Mater., 2015, 1, 1500232.
18. S. W. Shin, K.-H. Lee, J.-S. Park and S. J. Kang, ACS Appl. Mater. Interfaces, 2015, 7, 19666–19671.
19. S. J. Young, L. W. Ji, S. J. Chang and Y. K. Su, J. Cryst. Growth, 2006, 293, 43–47.
20. W. Tian, T. Zhai, C. Zhang, S.-L. Li, X. Wang, F. Liu, D. Liu, X. Cai, K. Tsukagoshi, D. Golberg and Y. Bando, Adv. Mater., 2013, 25, 4625–4630.
21. H. Lu, D. Zhang, X. Ren, J. Liu and W. C. H. Choy, ACS Nano, 2014, 8, 10980–10987.
22. Z. Siwei, G. Yuan, H. Bin, L. Jia, S. Jun, F. Zhiyong and Z. Jun, Nanotechnology, 2013, 24, 335202.
23. S. Kiruthika, K. D. M. Rao, K. Ankush, G. Ritu and G. U. Kulkarni, Mater. Res. Express, 2014, 1, 026301.
24. K. D. M. Rao and G. U. Kulkarni, Nanoscale, 2014, 6, 5645–5651.
25. K. K. Manga, J. Wang, M. Lin, J. Zhang, M. Nesladek, V. Nalla, W. Ji and K. P. Loh, Adv. Mater., 2012, 24, 1697–1702.
26. W. Dai, X. Pan, S. Chen, C. Chen, W. Chen, H. Zhang and Z. Ye, RSC Adv., 2015, 5, 6311–6314.
27. Z. Yang, M. Wang, X. Song, G. Yan, Y. Ding and J. Bai, J. Mater. Chem. C, 2014, 2, 4312–4319.
28. K. Liu, M. Sakurai, M. Liao and M. Aono, J. Phys. Chem. C, 2010, 114, 19835–19839.
29. V. Kumar, N. Singh, R. M. Mehra, A. Kapoor, L. P. Purohit and H. C. Swart, Thin Solid Films, 2013, 539, 161–165.
30. R. Ghosh, D. Basak and S. Fujihara, J. Appl. Phys., 2004, 96, 2689–2692.
31. Y. C. Lin, B. L. Wang, W. T. Yen, C. T. Ha and C. Peng, Thin Solid Films, 2010, 518, 4928–4934.
32. Z. Wang, X. Zhan, Y. Wang, S. Muhammad, Y. Huang and J. He, Nanoscale, 2012, 4, 2678–2684.
33. J. Liu, R. Lu, G. Xu, J. Wu, P. Thapa and D. Moore, Adv. Funct. Mater., 2013, 23, 4941–4948.
34. P. Sharma, K. Sreenivas and K. V. Rao, J. Appl. Phys., 2003, 93, 3963–3970.
35. K. Liu, M. Sakurai and M. Aono, Sensors, 2010, 10, 8604–8634.
36. S. Kiruthika, R. Gupta, A. Anand, A. Kumar and G. U. Kulkarni, ACS Appl. Mater. Interfaces, 2015, 7, 27215–27222.
37. S. Kiruthika, R. Gupta, K. D. M. Rao, S. Chakraborty, N. Padmavathy and G. U. Kulkarni, J. Mater. Chem. C, 2014, 2, 2089–2094.
38. K. D. M. Rao, R. Gupta and G. U. Kulkarni, Adv. Mater. Interfaces, 2014, 1, 1400090.

---

Footnotes

† Electronic supplementary information (ESI) available: Schematic of the fabrication process and microscopic images of the Au wire network/ZnO/Au wire network, transmittance of Au wire meshes on glass and quartz substrates, SEM, XRD and optical profilometry of the ZnO film over Au mesh, transmittance of Au/ZnO/Au on a quartz substrate, photocurrent measurement of ZnO with Ag paint contacts, I–V characteristics of the Au mesh/ZnO/Au mesh with different ZnO film thicknesses, XRD of a ZnO pellet, and calculations of the barrier height, note S1. See DOI: 10.1039/c6ra07118j

‡ On lien from JNCASR Bangalore 560064, India.

---