The catalytic decomposition processes of PH 3 on heated tungsten surfaces were studied to clarify the mechanisms governing phosphorus doping into silicon substrates. Mass spectrometric measurements show that PH 3 can be decomposed by more than 50% over 2000 K. H, P, PH, and PH 2 radicals were identified by laser spectroscopic techniques. Absolute density measurements of these radical species, as well as their PH 3 flow rate dependence, show that the major products on the catalyst surfaces are P and H atoms, while PH and PH 2 are produced in secondary processes in the gas phase. In other words, catalytic decomposition, unlike plasma decomposition processes, can be a clean source of P atoms, which can be the only major dopant precursors. In the presence of an excess amount of H 2 , the apparent decomposition efficiency is small. This can be explained by rapid cyclic reactions including decomposition, deposition, and etching to reproduce PH 3 .

The production of PH 3 from red phosphorus and the atomic hydrogen formed in the catalytic decomposition of H 2 on heated W surfaces was confirmed. The absolute density of the PH 3 could be as high as 10 13 cm-3 ; the density was proportional to H-atom density in the absence of red phosphorus, although three H atoms must be involved in the production of one PH 3 molecule, but showed minor dependence on the amount of red phosphorus. These results suggest that the rate-determining step for the production of PH 3 is that which produces H atoms.

No abstract

The reaction of Zn(CH 3) 2 and activated H 2 O produced in a reaction of H 2 and O 2 on a Pt catalyst and effused from a nozzle was examined both experimentally and theoretically. This reaction has been shown to be effective in the preparation of high-quality ZnO films. Laser-induced fluorescence measurements showed that radical species, such as H atoms and OH radicals, do not play major roles. The rotational temperature of H 2 O, measured with a coherent anti-Stokes Raman scattering technique, was 250 K. This low rotational temperature suggests that H 2 O molecules must be accelerated along the beam axis and that the collisional energy between Zn(CH 3) 2 and H 2 O is as high as 43 kJ mol-1. This energy is higher than the potential barrier to produce HOZnCH 3 +CH 4 , 38 kJ mol-1 , obtained by ab initio calculations at the MP2/LANL2DZ level of theory. HOZnCH 3 thus produced can be the strongest candidate ZnO film precursor.