Electrically driven reverse energy transfer process in Er-doped SiO2 layers containing Ge nanocrystals

Optically active Er3+ ions dispersed in a SiO2 layer are often used for amplifying signals at the wavelength of ~1.5 m that coincides with the maximum transparency window of silica-based optical fibers. In order to enhance the 1.5 m Er luminescence, a sufficient number of Er ions must be in the excited state, and it is only possible by employing a high-power pumped laser. Si nanocrystals (NCs) embedded in an Er-doped SiO2 layer show a new avenue to excite Er3+ ions indirectly, where Si-NC behaves like a sensitizer [1]. The basic mechanism of the above mentioned system relies on three facts: (i) optical excitation of Si-NC, (ii) nonradiative transfer of energy from Si-NC to the neighbouring Er3+, and (iii) subsequent relaxation from the first excited state to the ground state of Er3+ by emitting light at 1.5 m. However, to integrate these photonic systems into current microelectronics, it is essential to fabricate metal-oxide-semiconductor based light-emitting devices. Based on electroluminescence (EL) experiments, some groups have shown the importance of Si-NCs in enhancing the 1.5 µm Er luminescence. Here we use an experimental approach for evaluating the influence of Ge NCs on the Er EL in Er-doped SiO2 layers. In particular, we demonstrate an increase in intensity of the 400 nm band, characteristic of the Ge-NC related oxygen-deficiency centres (GeODC), at the expense of the Er-related signals with maxima at 522, 550, 660 and 1532 nm during electrical pumping at room-temperature  indicating an energy transfer process from the excited Er ions to the confined carriers in the GeODC, which is just opposite to the concept commonly used for the Er-doped SiO2 layers containing Si-NCs. Formation of Ge-NCs was verified by transmission electron microscopy. [1] Polman A, Nature Materials 2002;1:10.