Blue and Green Electroluminescence from Porous Materials
- PDF / 432,613 Bytes
- 6 Pages / 414.72 x 648 pts Page_size
- 57 Downloads / 218 Views
because the inner surface of PS is covered with hydrogen. The Si surface saturated by hydrogen is inert against further attack of fluoride ions if there are no holes at the Si electrode. Therefore, for n-type substrates, the light illumination is indispensable to forming PS because of the generation of holes. According to the PS formation mechanism proposed by V. Lehmann et al. [14], as the etching process proceeds, the band gap of the PS layer increases due to the quantum size effect, depending on the decrease in the size of Si crystallites. Therefore, for n-type substrates, when the band gap becomes larger than the energy of the illuminated light, the etching process automatically stops because the light cannot continue to generate holes. That is, the size of the Si crystallites depends on the energy of the illuminated light. However, although Steiner et al. have already used a similar method with UV light, neither green nor blue PL was observed [6]. In our opinion, however, this is probably due to the surface band gap of the Si nanocrystallites being smaller than that of the Si core. The theoretical calculation achieved by Takeda reveals that the band gap of two dimensional (2D) polysilane (SiH2)n drastically decreases (from 2.6 eV to 1.7 eV) when it is oxidized and changed to 2D siloxene (Si2H2-xOHx)n [15]. Therefore, the PS exposed to air shows only red PL regardless of its crystalline size [11]. In our experiments, to suppress oxidization as much as possible, PL was measured while the samples were in a C2H5OH solution. The substrates were n-type crystalline Si (c-Si) wafers, and their resistivities were 1 - 5 iacm. Al was used as an ohmic contact material. The PS layers were formed for 5 minutes by electrochemical anodization in a HF-C2H5OH solution (HF:H20:C2H5OH = 1:1:2) using a constant anodic current of 10 mA/cm 2 . The thickness of the PS layers was about 2-3 gm. During anodization, n-type c-Si wafers were illuminated with various light settings produced by filtering with a 500W halogen lamp. Samples #1, #2 and #3 were anodized using a light with a wavelength longer than 490 nm, 440 nm and 390 nm, respectively. Figure 1 shows PL spectra for each sample. The excitation light source was a HeCd laser (325 nm). The sample prepared with the light having shorter wavelength components shows a PL with a shorter peak. That is, the dependence of the PL on the Si crystalline size is clearly observed. Furthermore, the line width of the PL was found to be narrower than that of a conventional PS. We think this is another evidence that the origin of the PL in Fig. 1 (quantum confinement state) is difference from that of a conventional PL (surface effect). However. further experiments are necessary to confirm it. Using the above results to fabricate green EL devices, we electrically anodized c-Si (n type, I5 Qcm) for 5 minutes with the UV light (wavelength, from 240 to 420 nm) produced by filtering the light with a 150 W Hg-Xe lamp. On the PS anodized with UV light (UV-PS), we deposited ITO (thickness, 600 A) as a tran
Data Loading...