The highest-energy UV light from the sun, called UVC, has a wavelength shorter than 280nm and, fortunately for us, cannot penetrate the earth's protective atmosphere. As a result, this region of the solar electromagnetic spectrum constitutes a ‘black background’ that can be used to detect and control artificial UVC-emitting sources such as flames (generated by any kind of combustion) with very high sensitivity. The accuracy of these measurements can be further increased by using a detector that is ‘solar blind,’ i.e., insensitive to light above 280nm: see Figure 1.1 Conventional applications of solar-blind photodetection include fire sensing and missile interception. More recently, the technology has been applied to decontaminate water and air using UVC, which must be monitored to prevent unwanted scattering of light. But the solar-blind detectors presently on the market are costly and require a high operating voltage, which discourages their widespread use.
Figure 1. The principle of solar-blind photodetection. The solar spectrum has no radiation in the deep-UV region below 280nm (UVC). This so-called black background enables highly sensitive detection of flames or other nonsolar light sources that produce a UVC signature.
As a low-voltage alternative, researchers have synthesized materials known as UVC-sensitive wide-bandgap semiconductors by growing them either as epitaxial (layered) films or as nanowires. However, epitaxy is limited by cost, and nanowires are difficult to integrate into a working device. For example, fabricating photodetectors requires transferring nanowires randomly to an insulating substrate and connecting them to metal electrodes using a lift-off process2—see Figure 2(a)—that is not amenable to efficient production.
Figure 2. Nanowire device assembly techniques. (a) The conventional technique relies on multistep photolithography for fabricating electrodes. (b) In our technique, the electrodes and the sensing elements are made in a single step. The sensing elements consist of nanowires that bridge the gap between the electrodes without coming into contact with the substrate. Au: Gold. CVD: Chemical vapor deposition.
We have made photodetectors by creating bridging nanowires between electrodes in a single-step chemical vapor deposition process: see Figure 2(b). The photodetector consists of thick nanowire layers that serve as electrodes and additional nanowires that bridge the gap between the layers and act as sensing elements3–5 (see Figure 3, photos). To achieve selective growth of the nanowire layers, we patterned very thin (∼2nm) gold layers in the shape of the electrodes on a quartz substrate. The bridging nanowires are formed across the gap between the electrodes and have no contact with the substrate.
Figure 3. Left: Scanning electron microscopy images of nanowires bridging the gap between electrodes. Right: UVC photoresponse and spectral selectivity of the gallium oxide bridging nanowires.
The bridged structure gives these photodetectors several advantages over their conventional nanowire counterparts: efficient and cost-effective fabrication, contamination-free surfaces (no post-treatment required following nanowire growth), and special properties that derive from their substrate independence. Figure 3 shows the performance of bridged structures having nanowires made of gallium oxide. Note that there is no sensitivity to light at wavelengths greater than 280nm, which indicates a solar blind.5 We also constructed nanowires made of zinc oxide for detecting UVA light (insensitive above 380nm, i.e., ‘visible blind’) in applications such as chemical and biological sensing and solar-energy monitoring. All our devices show high photoresponse, good spectral selectivity, low photocurrent noise, and fast time response.
In summary, we have described metal oxide nanowire-based photodetectors that show promise as a cost-effective alternative to epitaxially grown thin-film-based sensors. Our goal now is to monitor electron flow in the electrodes and nanowires to better understand their capabilities and to improve our designs. To that end, we will be investigating the semiconductor material used in the nanowires to identify defects that can degrade the conversion of light into an electrical signal. Next, we will fine-tune the growth conditions to reduce those defects. In addition, we will test the ability of our photodetectors to operate in harsh environments, such as extreme heat, to expand their usefulness. Finally, we will seek to integrate the devices into highly accurate UVC sensor arrays for a variety of applications.
Jean-Jacques Delaunay, Yanbo Li, Takero Tokizono
University of Tokyo
Meiyong Liao, Yasuo Koide
National Institute for Materials Sciences
1. M. Razeghi, Short-wavelength solar-blind detectors—status, prospects, and markets, Proc. IEEE 90, pp. 1006-1014, 2002.
2. C. Soci, A. Zhang, B. Xiang, S. A. Dayeh, D. P. R. Aplin, J. Park, X. Y. Bao, Y. H. Lo, D. Wang, ZnO nanowire UV photodetectors with high internal gain, Nano Lett. 7, pp. 1003-1009, 2007.
3. Y. B. Li, F. Della Valle, M. Simonnet, I. Yamada, J.-J. Delaunay, High-performance UV detector made of ultra-long ZnO bridging nanowires, Nanotechnology 20, pp. 045501, 2009.
4. Y. B. Li, A. Paulsen, I. Yamada, Y. Koide, J.-J. Delaunay, Bascule nanobridges self-assembled with ZnO nanowires as double Schottky barrier UV switches, Nanotechnology 21, pp. 295502, 2010.
5. Y. B. Li, T. Tokizono, M. Liao, M. Zhong, Y. Koide, I. Yamada, J.-J. Delaunay, Efficient assembly of bridged β-Ga2O3 nanowires for solar-blind photodetection, Adv. Funct. Mater. 20, pp. 3972-3978, 2010.