Improved near surface heavy impurity detection by a novel charged particle energy filter technique

K. Ishibashi, B. K. Patnaik, N. R. Parikh, H. Tateno, J. D. Hunn

Research output: Contribution to journalConference articlepeer-review

Abstract

As the typical feature size of silicon integrated circuits, such as in VLSI technology, has become smaller, the surface cleanliness of silicon wafers has become more important. Hence, detection of trace impurities introduced during the processing steps is essential. A novel technique, consisting of a 'Charged Particle Energy Filter (CPEF)' used in the path of the scattered helium ions in the conventional Rutherford Backscattering geometry, is proposed and its merits and limitations are discussed. In this technique, an electric field is applied across a pair of plates placed before the detector so that backscattered particles of only a selected energy range go through slits to strike the detector. This can be used to filter out particles from the lighter substrate atoms and thus reduce pulse pileup in the region of the impurity signal. The feasibility of this scheme was studied with silicon wafers implanted with 1 × 1014 and 1 × 1013 54Fe/cm2 at an energy of 35 keV, and a 0.5 MeV He+ analysis beam. It was found that the backscattered ion signals from the Si atoms can be reduced by more than three orders of magnitude. This suggests the detection limit for contaminants can be improved by at least two orders of magnitude compared to the conventional Rutherford Backscattering technique. This technique can be incorporated in 200-300 kV ion implanters for monitoring of surface contaminants in samples prior to implantation.

Original languageEnglish
Pages (from-to)399-404
Number of pages6
JournalMaterials Research Society Symposium - Proceedings
Volume354
StatePublished - 1995
Externally publishedYes
EventProceedings of the 1994 MRS Fall Meeting - Boston, MA, USA
Duration: Nov 28 1994Nov 30 1994

Fingerprint

Dive into the research topics of 'Improved near surface heavy impurity detection by a novel charged particle energy filter technique'. Together they form a unique fingerprint.

Cite this