SIMS Measurement of Na

Research output: Conference Papers (RGC: 31A, 31B, 32, 33)32_Refereed conference paper (no ISBN/ISSN)peer-review

View graph of relations

Author(s)

Detail(s)

Original languageEnglish
Publication statusPublished - Jun 1986
Externally publishedYes

Conference

TitleSILI-CON 1986 Conference
PlaceUnited States
CityPortland
Period30 June - 2 July 1986

Abstract

Sodium ions located in the gate oxide of an MOS device are known to negatively affect the device performance depending upon the Na concentration and mobility in the oxide. Recently quantitative measurement techniques have been developed to measure surface Na on silicon or in silicon dioxide by analyzing surface material removed by etching. These techniques include: flameless atomic absorption [1]; and instrumental neutron activation analysis [2]. Both techniques are reported to have excellent detection limited (below 1x1010/cm2), but one must assume that all the Na is removed from the surface in order to know the true amount of Na present.

A complementary technique, Secondary Ion Mass Spectrometry (SIMS), can be used to unambiguously measure all the Na on the surface of silicon or in the silicon dioxide. SIMS can be a very powerful analytical technique, particularly when matrix effects can be avoided, or accounted for, and when appropriate standards can be developed for quantitation. This work focuses on these issues for the particular measurement of Na on the surface of a silicon wafer which has a thin native oxide or Na in silicon dioxide. In this case there can be a matrix change from Si0x to silicon during the measurement.

The difference in ion yield between a silicon matrix and a silicon matrix with oxygen has been empirically determined for an 0 2 primary ion beam on a CAMECA IMS-3f Microanalyzer using Na ion implants into silicon dioxide, into silicon, and into silicon co-implanted with a high dose of oxygen. The Na implant fluence was varied from 1x1011/cm2 up to 3x1013/cm2, and the implant energies were selected to insure all the Na was located in the silicon dioxide for the oxide sample and deep in the silicon for the silicon samples. The SIMS profiles and calculations of relative sensitivity factors (RSF) from the profiles show that the ion yield of Na is independent of the matrix.

A scheme can be developed then to quantitatively measure surface Na levels using Na implanted into silicon dioxide or into silicon for calibration. In the case where the silicon dioxide is thicker than the depth at which a reactive ion beam (e.g. 02+) stabilizes its ion yield, the natural drift if the Na ions under the electric field created by the ion beam charging effects during the sputtering process to the silicon dioxide/silicon interface becomes an advantage for the signal-to-noise ratio. In the case where the silicon dioxide is thinner than the stabilization depth for the reactive ion beam, the addition of an oxygen leak or jet is required to provide a constant ion yield. If a non-reactive primary ion beam (Ar or Xe) is used, the oxygen jet or leak is not required. Data will also be presented showing a correlation between relative Na levels on silicon wafers using a Xe primary ion beam on a PHI 560 with a SIMS II probe and ESCA quantitative measurements for intentionally contaminated wafers.

Bibliographic Note

Information for this record is supplemented by the author(s) concerned.

Citation Format(s)

SIMS Measurement of Na. / Hockett, R. S.; Chu, P. K.; Bleiler, R. J.

1986. Paper presented at SILI-CON 1986 Conference, Portland, United States.

Research output: Conference Papers (RGC: 31A, 31B, 32, 33)32_Refereed conference paper (no ISBN/ISSN)peer-review