Band Edge Modified Rare Earths - A route to the mid infrared in Silicon

摘要

We describe a new technology, band edge modification (BEM) of rare earth (RE) optical transitions in silicon. BEM refutes previous assumptions on the interaction of REs with semiconductors and other hosts (M. A. Louren?o et al, Adv. Funct. Mater., 26, 1986-1994, 2016). This approach opens up a route to efficient, fully-silicon-based, optoelectronic devices across the near and mid-IR. Intrinsic RE transitions are internal to the RE and do not contribute directly to carrier conduction in the bands of the host. Consequently, this makes them of limited use for optical detectors. The band edge modified RE levels exampled here interact directly with the silicon bands and so offer the possibility of extrinsic photovoltaic or photoconductive detectors. Silicon detectors and cameras currently completely dominate the ultra violet, visible and very near-IR regions - however they do not work well beyond 1.1 mm, the silicon band gap. We have used ion implantation to introduce europium, ytterbium and cerium into silicon photodiodes, also formed by ion implantation. We show that BEM enables efficient silicon detectivity to be extended from 1.1 mm out to the mid-IR region. The responsivites and detectivities of these new silicon detectors offer a real challenge to existing detector materials and devices in the 2 to 6 mm range - currently dominated by more challenging, and expensive materials such as mercury cadmium telluride, indium antimonide and the arsenides. Replacing these materials with silicon would offer enormous benefits in cost, reliability and also integration with the silicon microelectronics for detection and imaging. An additional benefit is using much less toxic materials and production processes - a major concern with current technologies. Low leakage currents achievable in silicon based photodiodes mean that further development of this new mid-IR silicon technology could lead to thermoelectrically cooled or even room temperature detectors. Current commercial detectors in this area have to be cooled to liquid nitrogen temperatures (77 K) to achieve the performance needed for most applications. Higher operating temperature (HOT) detectors are an industry aim and, particularly if implemented in silicon, would be a major breakthrough. We acknowledge the Royal Society UK for the award of the 2015 Brian Mercer Award for Innovation.