White Electroluminescence from a Gadolinium-Doped Si-Nanocluster-Enriched SiO[sub 2]–SiON Interface Region

Abstract

SiO2 implanted with carbon and silicon. In this article, we present bright white light emitters based on the silicon-rich SiO2‐SiON interface in a metal‐SiON‐SiO2‐Si MONOS structure coimplanted with gadolinium Gd. At a certain level of the constant injection current excitation, the MONOS structure containing Si-nc and Gd works as a white light emitter. At a constant current excitation of 100 A, the EL spectrum covers a wide spectral region from 360 to 700 nm with a broad flat region of about 200 nm. The increase in the constant current excitation switches the color of the light emitters observed by the naked eye from green to blue. For constant current excitation above 1 mA, the energy transfer from the Gd 3+ ions to the neutral oxygen vacancy defects and the silicon nanoclusters was observed. The photoluminescence PL measurements taken from only silicon implanted structures show two broad-bands with peaks at 640 and 780 nm. In Gd coimplanted Si-rich MONOS structures, the PL spectrum consists of one peak at 650 nm related to a nonbridging oxygen hole center NBOHC and two peaks at 760 and 830 nm associated with silicon nanoclusters with sizes of 1.7 and 1.8 nm, respectively. EL devices, metal‐SiON‐SiO2‐Si light emitting device MONOSLED, were prepared by a standard silicon MOS technology on 4 inch, n-type silicon 100 wafers with resistivities of 2‐5 cm. Thermally grown 100 nm thick SiO2 layers with a 1000 nm field oxide were processed using the local oxidation of silicon technology and were implanted by silicon and coimplanted with Gd at energies of 20 and 140 keV with atomic concentrations of 10 and 2%, respectively. Samples were annealed at 1100°C for 30 min after silicon implantation to form Si-nc and at 900°C for the same time after Gd coimplantation to reduce the radiation defects and to rebuild Si-nc. To protect the oxide layer against breakdown, a 200 nm SiON layer was deposited on top of it O:N =1 :1 by plasmaenhanced chemical vapor deposition. The atomic composition of the SiON layers and the concentration of implanted elements were confirmed by scanning Auger electron spectroscopy Microlab 310F. The gate and bottom electrodes consist of a 100 nm thick indium tin oxide ITO layer deposited by radio frequency sputtering and a 300 nm thick evaporated Al layer, respectively. Cross-sectional transmission electron microscopy images were taken by means of an FEI Titan 80-300 scanning transmission electron microscope operating at 300 keV. The EL measurements were performed using electron injection mainly from the ITO electrode. The EL spectra and the quenching of the EL intensity were measured at room temperature on MONOS structures with a circular ITO electrode of 300 m diameter at a constant current supplied by a source meter Keithley 2410. The EL signal was recorded with a monochromator Jobin Yvon Triax 320 and a photomultiplier Hamamatsu H7732-10 .P L experiments were conducted at room temperature with an excitation wavelength of 514.5 nm from an argon laser with a power density of 0.2 mW/mm 2 . The PL signal was recorded with an SPM-2 monochromator and a photomultiplier Hamamatsu R943-02. The photograph was taken by a standard digital camera combined with a microscope. Figure 1 shows the PL spectra taken from Si-nc’s embedded in the SiO2 layer and from the SiO2 layer containing Si-nc’s coimplanted with Gd. In the silicon implanted oxide layer with subsequent annealing at 1100°C, the PL signal shows a broad spectrum with two main bands centered at around 650 and 800 nm. The Gaussian peak deconvolution inset of Fig. 1 enables us to show