(Invited) Rare Earth Doped Metal-Oxide-Semiconductor Structures: A Promising Material System or a Dead End of Optoelectronic Evolution?

Abstract

Integrated photonics is a key technology of the 21st century, and the electrically driven, integrated light emitter is an important building block, but difficult to realize. Thus, an enormous variety of different materials and material systems have been investigated in the past, ranging from the various approaches to integrate III-V semiconductors to the different types of Si-based light emission. Within the latter group rare earth (RE) implanted MOS structures feature a high conformity with standard CMOS processes combined with the excellent optical properties of RE elements.After more than 20 years of intense research in the field of Si-based light emission the results are mixed. With respect to important key parameters, most of all power efficiency, Si-based light emitters are not yet able to compete with their counterparts based on compound or organic semiconductors. The present contribution discusses the problems of Si-based light emitters at the example of RE-doped MOS structures, and compares various light emitter designs and their potential to overcome these problems.In detail, the power efficiency, the operation lifetime and the operation voltage of Tb- and Er-doped MOS structures are investigated. The main electroluminescence excitation mechanism is impact excitation of hot electron which plays an ambivalent role: efficient excitation is often related with efficient defect creation. In addition, a dark zone close to the injecting interface limits the scalability towards low voltages. The excitation mechanism und thus the performance of the light emitter is affected by the structure and composition of the dielectric stack of the MOS structure. Within this study, several host materials for the RE ions, namely stoichiometric and Si-rich silicon oxide or silicon nitride; different fabrication methods, namely plasma enhanced chemical vapour deposition, ion implantation and atomic layer deposition; and the use of additional buffer or injection layers are investigated. Finally, a short perspective to potential applications is given.