Next generation avalanche photodiodes: realising new potentials using nm wide avalanche regions

点击翻译按钮获取中文摘要

The internet data rate of Mb/s is currently available to UK homes thanks to installation of fibre network. Recently Fujitsu, a major telecom company, outlined their plan to lay Gb/s fibre network in UK, which can increase the data rate to 10 Gb/s and beyond. Therefore optical fibre will play an ever increasing importance in our life and hence there is a clear need to carry out research in ultrafast optical components such as photodiodes, used to convert optical signal to electrical signal. In photodiodes the energy from light is used to release an electron from an atom and a detectable current is generated when the electron is swept by an electric field. In a specially designed avalanche photodiode (APD) the electric field is increased such that a single electron generated by the photoelectric effect can produce an avalanche of electrons and holes. Consequently a much larger signal is produced, leading to a better signal to noise ratio. Unfortunately current commercial APD can only work up to 10 Gb/s and is therefore not future proof. In this proposal, we will develop extremely thin 10-50 nm semiconductor layer to achieve the avalanche effect at ps time scale such that our APDs can operate at bit rates of Tb/s. The new semiconductor materials that will be developed in this project are AlAsSb and AlGaPSb since they have great potential to withstand extremely high electric field while maintaining low dark current (essential to minimise errors in digital signal). Crucially since our materials are only nm thick, we can engineer the electric field in APD to impose some degree of coherence in the electron and hole behaviours so that the avalanche effect occurs with minimal noise. We believe our APDs can be designed to approach the performance of an ideal noiseless APD with high bandwidth for optical communications. We recently demonstrated that the avalanche effect in thin AlAsSb is relatively immune to temperature change. Therefore in addition to ultra high speed optical communication, our proposed nm scaled AlAsSb and AlGaPSb avalanche layers are envisaged to work as an ultra fast photon counter with high immunity to ambient temperature fluctuation. Since a photon is the basic unit of light, the "ultimate" light sensor is achieved by increasing the avalanche gain to approximately a million so that the APD works as a photon counter. Our thin avalanche layer has the potential to register a photon count in a few ps, which is at least an order of magnitude faster than current APD photon counters. If successful one of the major impacts of our photon counter will be to improve the data encryption technique called quantum key distribution in which the data is encrypted using a single photon. This is believed to be the most secure encryption technology. Any unauthorised detection of the photon will cause a significant error rate, and hence alerting the sender of the attempted hacking. Therefore the high thermal stability and fast response time of our APDs will enhance the robustness of future quantum cryptography systems. We also believe our new technology will bring significant improvement to medical X-ray imaging as the APD can improve the signal to noise ratio of X-ray detection system. Typically the avalanche effect increases the electrical signal, induced by the X-ray absorption, to above the electronic circuit noise and hence enhancing the image quality. Our recent work showed that having a thin avalanche layer is essential for high performance X-ray APD. Hence our work will enable a new generation of X-ray APDs for imaging applications. To achieve the goals discussed above we will carry out very systematic development of AlAsSb and AlGaPSb APDs via advanced growth of the semiconductor crystals and optimised chemical etching process as well as meticulous measurements to extract key material properties for design of high performance APDs utilising nm avalanche regions.