Ultrahigh depth resolution secondary ion mass spectrometry with sub-keV grazing O2+ beams

2005

It is known that by lowering the impact energy the sputter rate and surface transient width in SIMS will be reduced. However, few studies have been done at ultralow energies over a wide range of impact angles. This study examines the dependence of sputter rate and transient width as a function of O 2 + primary ion energy (E p = 250 eV, 500 eV and 1 keV) and incidence angles of 0-70• . The instrument used is the Atomika 4500 SIMS depth profiler and the sample was Si with 10 delta-layers of Si 0.7 Ge 0.3 . We observed that the lowest transient width of 0.7 nm is obtainable at normal and near-normal incidence with E p ∼ 250 eV and E p ∼ 500 eV. There is no significant improvement in transient width going down in energy from E p ∼ 500 to ∼250 eV. The onset of roughening is also not obvious at E p ∼ 250 eV over the whole angular range studied. Although the sputter rate during the surface transient is normally different from that at steady state, only at E p ∼ 250 eV was it observed that the sputter rate remained fairly independent of depth. We conclude that the best working ranges to achieve a narrow transient width and accurate depth calibration are at E p ∼ 250 eV/0 • < q < 20• and 500 eV/0 • < q < 10• .

Section: Introductionmentioning

confidence: 93%

Surface transient effects in ultralow‐energy O₂⁺ sputtering of silicon

Chanbasha

2005

“…The sample, which consists of sixteen boron delta layers (henceforth referred to as 'Bdelta'), was provided by SEMATECH for a round-robin (3,6,9,12) were not calculated owing to the fact that these values will be influenced by the surface transient. study on ultrashallow junction measurement; details are given elsewhere.…”

Section: Change In Erosion Rate Due To Interfacementioning

confidence: 99%

“…The doses were calculated by summing the calibrated concentration (in atoms cm 3 ) over the depth sputtered. The precleaning procedure has no influence on the measured boron doses in the non-implanted samples (samples 13 and 14), the doses being the background boron dopant concentration in these samples.…”

Section: Surface Yield Enhancement: Nature Of the Surface Transientmentioning

confidence: 99%

“…It has been shown repeatedly that using low-energy primary ions reduces this surface transient to a few nanometres, but further reduction is impeded by the presence of the thin native oxide. 2,3 Moreover, surface roughening comes into play and complicates the measurement when the primary energy is within the sub-keV regime. 4 Oxygen flooding, on the other hand, helps to stabilize the ion yield and suppresses surface roughening by furnishing an oxygen-rich environment that favours rapid and complete oxidization of the sputtered surface.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Evaluation of the silicon capping technique in SIMS

Huan

et al. 2002

The effectiveness of the silicon capping technique in ultrashallow secondary ion mass spectrometry using 1 keV O 2 + at 56• incidence angle is investigated. It is shown that a capping layer of at least 20 nm is needed to ensure a steady-state erosion rate when profiling the doped region. The matrix intensity is very sensitive to the presence of native oxide, and enhancement is evident for oxides as thin as one monolayer. The oxygen enhancement is further influenced by the high dopant concentration on the sputter surface. The present results favour the model that relates the surface spike present during depth profiling to the sputtering of native oxide.

“…Most work on surface transient has been done using O 2 C beams. To reduce the transient width for O 2 C beams, many have attempted reducing the impact energy (E p ) and/or increasing the incident angle (Â) 14,15 with respect to surface normal, while others have found normal and near-normal incidence to be advantageous. 1,16,17 Work on the surface transient for ultralow-energy Cs C beam has been limited.…”

Section: Introductionmentioning

confidence: 99%

Surface transient effects in ultralow‐energy Cs⁺ sputtering of Si

Chanbasha

2007

This study examines the dependence of the sputter rate and the transient width (z tr ) as a function of Cs + primary ion energy (impact energy .E p / = 320 eV, 500 eV and 1 keV) and incident angles between 0 and 70°. The instrument used was the ATOMIKA 4500 SIMS depth profiler and the sample was Si with ten delta layers of Si 0.7 Ge 0.3 . We observed the narrowest transient widths of between 1.4 and 2.0 nm apparent depth. This was achieved at incident angles (q) of 30-50°. An extended transient effect was observed when profiled at q > 50°. Below this incident angle, the transient width is less than twice the penetration depth (z tr < 2R norm ). At minimum z tr , z tr ≈ R norm . The detection sensitivity is best achieved at q ≈ 30°for all energies investigated. The sputter rate is lowest at normal incidence, rising gradually to a maximum at q ≈ 50-60°. This is similar to that observed with ultralow-energy O 2 + primary ion beams. 1 At ultralow energies, reducing E p does not have a significant effect in reducing z tr . We conclude that for E p < 1 keV, the optimum condition to achieve minimum z tr while maintaining good sensitivity and high sputter rate is at q ≈ 30°.