Atomic Layer Etching of Porous Silicon

  • PDF / 304,932 Bytes
  • 6 Pages / 414.72 x 648 pts Page_size
  • 71 Downloads / 234 Views

DOWNLOAD

REPORT


ABSTRACT In this work we describe the controlled shifting of the PL peak of p4 (10 m~cm) porous silicon (PoSi) by means of atomic layer etching (ALEP). We hereby study the cluster-size dependence of the PL of this material. By this technique of repeated oxidation by H20 2 and stripping of the oxidized surface layer, we reduced the size of the crystallites layer by layer. In all previous reports the PoSi PL appeared to have a natural lower energy limit of- 1.4 eV. We report for the first time a continuous shift of the PoSi PL peak between 1.01 and 1.20 eV. This observation allows us to draw conclusions for the luminescence mechanism: it proves that geometrical quantum confinement in Si crystallites is responsible for the efficient room-temperature PL in PoSi near the indirect bandgap of c-Si. Together with observations of size-independent PL peaks around 1.6 eV in thermally oxidized samples this result indicates that the PoSi PL cannot be described by one origin alone. Both the existence of molecular centers and the geometrical quantum confinement are valid in their specific range of etching and post-anodic treatment parameters.

CONCEPT AND REALIZATION The ALEP technique is the repeated oxidation by H20 2 and stripping of the oxidized surface layer. The starting material for this process is as-prepared, H-terminated p4 PoSi [1]. Immersing the sample in H 20 2 (aqueous 30 %) for 5 minutes leads to a typical oxidation of the PoSi layer [2]. The peak at 2250 cme' observed in FTIR measurements is assigned to the O3-Si-Hx stretching mode. Dipping the sample afterwards in ethanoic HF (mixture 1:1 vol. of HF (aqueous 49 %) and ethanol) for 5 minutes produces the typical FTIR spectrum of H-terminated PoSi [3]. This change in surface termination can be explained by the reactivity of HF with the backbonded oxygen. This results in the removal of the outermost Si layer bonded to the attacked oxygen layer [2]. The repeated application of the oxidation and stripping of the oxidized layer leads to a controlled size reduction, layer by layer. After a certain number of steps, ALEP was no longer a useful method, because of the internal instability of the sponge-like structure, caused by the reduced crystallite size.

PL SPECTRA The light-emitting properties of the ALEP samples are studied as a function of the crystallite size. Fig. I shows the PL spectra of several H-terminated samples under 1.96 eV excitation at a temperature of 15 K collected by a 22 cm Spex monochromator and detected by a Hamamatsu PMT. 511

Mat. Res. Soc. Symp. Proc. Vol. 452 0 1997 Materials Research Society

1.2

E

1.0

Etching Steps

0.8

Substrate

E.., =1.96 eV 300 mW/cm

I,,

1100>, p-(B),

0.6

Data Loading...