Spontaneous etching of Si(100) by XeF2: Test case for a new beam surface experiment

Abstract

An ultrahigh vacuum multiple-beam setup has been designed to study surface reactions that are of importance in plasma etch processes. The setup consists mainly of five beams (reactive neutrals, ions, electrons, CFx radicals, and photons) all of which can be focused on the same sample area, and a quadrupole mass spectrometer detecting only molecules desorbing from this area. Data are reported on spontaneous etching for the Si(100)/XeF2 system. Both the incident flux of XeF2 on the sample and the desorbing fluxes of SiFx products and nonused XeF2 were measured quantitatively. The reaction of XeF2, the production of SiFx species and the accumulation of fluorine on the silicon surface were studied as a function of temperature (300–900 K) for XeF2 fluxes of 0.24 and 1.04 monolayers/s. For the reaction probability of XeF2 an exponential increase with temperature from 11% at 300 K up to 50% at 900 K was found. The main product was found to be SiF4 at 300 K gradually becoming replaced by SiF2 at temperatures higher than 600 K. A remarkable result is the accumulation of a rather thick SiFx reaction layer on the etched surface. From thermal desorption spectra the fluorine content was calculated to be 38 monolayers at 300 K decreasing rapidly for higher temperatures. This suggests that fluorine diffusion into the silicon bulk might be of importance in the etching process.

Surface roughness in XeF2 etching of a-Si∕c-Si(100)

“…3,4 In this section only the two modifications that have been made recently will be discussed. These are the addition of a sample exchange mechanism and the addition of an ellipsometer.…”

“…The present system has also been studied intensively by etch product analysis. [3][4][5][6][7][8][9][10][11][12] However, surface roughness caused by the etching has never been characterized in detail, although various authors 2,7,13 mention the importance in fully understanding the obtained models inspired by etch product analysis.…”

Ion-assisted Si/XeF2 etching: Temperature dependence in the range 100–1000 K

“…For a detailed description of the total multiple beam setup, we refer to a previous article. 18 The silicon sample ͓Si͑100͒, n type, phosphorus, 2-3 ⍀ cm͔ is clamped to a nickel sample holder, positioned in a UHV chamber (Ͻ10 Ϫ8 Torr͒ at the intersection of the XeF 2 and the Ar ϩ beam. Heating of sample and holder is achieved by means of a coaxial heating wire, and cooling by a liquid nitrogen vessel.…”

“…In the spontaneous etching period the inert nickel sample holder was occasionally shifted into the detection area for a few minutes to calibrate the detection signals. 18 All signals were corrected for background influences and the temperature dependence of the detection probability. 18 During this study several silicon samples were used and each sample was used for several experiments.…”

“…18 All signals were corrected for background influences and the temperature dependence of the detection probability. 18 During this study several silicon samples were used and each sample was used for several experiments.…”