Redistribution of boron during thermal oxidation of silicon studied by SIMS using Ar⁺ bombardment and the MISR method. Part I: Methodology

Abstract

Quantitative depth profiling of boron through SiO,/Si interfaces, when profiling with AT+, is complicated by the matrix effects. The presence of oxygen affects the sensitivity of boron by three orders of magnitude and accurate determination of the distribution coefficient of boron at the interface is only possible when the matrix effects can be eliminated or taken into account in the quantification. By using matrix-sensitive correlation factors, a sensitivity curve of boron at different primary ion current densities was established. This relation shows a good long-term reproducibility. Moreover, a physical interpretation of the shape of the sensitivity curve of boron is given by comparison with pure Si and SiO, . Tentative explanations for the hehaviour of different Si,O;+ species in conditions of varying primary ion current density and oxygen backfill are given. The influence of the oxygen backfill pressure on the sputter rate of silicon was investigated and problems arising in the profilation of boron in SiOJSi interfaces were addressed. Practical applications of the described methodology are discussed in Part I1 of this series.

In situ XPS and SIMS analysis of O2+ beam‐induced silicon oxidation

“…8. Figure 9 illustrates the contribution of the sub-oxidation states and the full oxidation state in the ion beam-induced oxide layer, as well as the calculated matrix ion species ratio (MISR) 13 values for 44 SiO C / 56 Si 2 C and 30 Si C / 56 Si 2 C at different incidence angles at steady-state sputtering. The Si 2p(IV)/Si 2p(ox) curve in Fig.…”

Atomic mass spectrometry

“…Quantification difficulties in SIMS continue to attract the research and development efforts of many groups. Application of the matrix ion species ratio (MISR) method was required for the accurate depth profile determination of B in silicon due to severe sensitivity effects caused by the presence of 0 (16). The study of relative sensitivity factors (RSF) and their transferability between instruments and with the same instrument at different analysis times has been the subject of attention in several investigations (17,18).…”