The properties of silicon nitride, oxynitride, and oxide films formed by the pyrolysis of various mixtures of SiH4 , NH3 , and NO are presented. The variation in physical, optical, and electrical properties of this oxynitride false(SixOyNzfalse) series is examined. The electrical and passivation properties of these films on Si are examined and compared with oxides. These electrical data describe the general characteristics of nitride and oxynitride on top of Si and over thin (∼300Aå) and thick (∼1000Aå) thermal oxide films.

No abstract

750~in Fig. 7. As is evident, the deposition rate is practically independent of the total flow rate. This is also indicative of a kinetically controlled process. This experiment was repeated for the B type orientations with similar results.Flow rate independence is a necessary but insufficient condition for elimination of diffusion controlled processes since it could be due to peculiar flow conditions such as poor mixing at high flows (7). However, all available evidence is consistent with the concept of a kinetically controlled deposition process in the region where the deposition rate increases with increasing substrate temperature. This evidence may be summarized as (i) a strong dependence of the deposition rate on the substrate orientation, (ii) rapidly increasing deposition rate with increasing substrate temperature, and (i~i) insensitivity of the deposition rate to the total gas flow rate.The rate behavior in high temperature region beyond the maximum may indicate a close approach to equilibrium. Within experimental error the high temperature regions of the curves approximate the shape of the equilibrium curve (Fig. 6) in this region. However, the decreasing rate at high temperatures could also be due to an activated etching process whose rate increases with increasing temperature and becomes the dominant process at temperatures above 8O0~ The fact that the {lll}B rate was found to be low and relatively constant over the entire temperature range remains unexplained. One possibility is that an impurity is selectively adsorbed at the critical growth sites on the {lll}B surface. Perhaps the answer may be found by a study of the effects of gas composition on epitaxial growth. This study is now in progress.The variation of the deposition rates as a function of the substrate orientation is interesting. The high {lll}A deposition rates attest to the ease of arsenic addition to the growing surface. For an ideal {lll}A surface, an incoming arsenic makes fewer surface bonds than with any other orientation. However, this same {lll}A surface has the greatest deposition rate for the reactant gas composition used in this study. The decrease in polarity effects is also obvious in this series as the surface deviates by greater amounts from a {111} toward the nearest {10O}. Thus the differences between the A and B surfaces at 750~ were {111} = 85 ~/hr, {112} = 43 ~/hr, and {113} ----3 /~/hr. This trend confirms the earlier results which indicated no significant polarity effects for the {115} orientation. The increase in deposition rate in the vicinity of the {l13}B orientation on the curve shown in Fig. 5 offers support for Sangster's theoretical predictions (3). Thus {113} is a favorable orientation for crystal growth since it tends to combine the good nucleation characteristics of a {1OO} with the favorable surface packing and stoichiometry of the {111}. Further clarification of the effects of orientation and surface polarity must await completion of the study concerning the effects of vapor composition on the deposition...

No abstract

A qualitative study of the masking properties of thin (≤1500Aå) silicon nitride and silicon oxynitride films on Si is presented. A range of diffusion conditions was studied for doping sources including B, P, Ga, and As. Silicon nitride was not found to be a diffusion mask for all conditions. Conditions under which it can be expected to mask are specified.