Secondary ion yield changes in Si and GaAs due to topography changes during O+2 or Cs+ ion bombardment

Abstract

Changes in secondary ion yields of matrix and dopant species have been correlated with changes in surface topography during O+2 bombardment of Si and GaAs. In Si, profiles were measured in (100) wafers at 6- and 8-keV impact energy. At 6 keV, a yield increase of about 70% occurred for Si+ over a depth range of 2.5 to 3.5 μm, with changes in other species ranging from a decrease of ∼20% for Si+3 to an increase of more than 25% for O+. The development of a rippled surface topography was observed in scanning electron micrographs over the same depth range. Similar effects occurred over a 3–5 μm depth range for 8-keV ions, and in (111) silicon at a depth of 3 to 4 μm for 6-keV ions. No differences were noted between p- and n-type silicon, or implanted and unimplanted silicon. In GaAs, profiles were measured in (100) wafers at 2.5-, 5.5-, and 8-keV impact energies. At 8 keV, a yield increase of about 70% was found for GaO+ in the range 0.6–1.0 μm, with smaller changes for other matrix species. At 5.5 keV, similar effects were observed, but over a depth interval of 0.3 to 0.7 μm. No yield changes were detected at 2.5-keV impact energy. The yield changes at the higher energies were again correlated with the onset of changes in topography. No change in ion yield or surface topography was noted for Cs+ bombardment of Si or GaAs. The topography and ion yield changes are affected by the angle of incidence and, for Si, the oxygen coverage. The results show that the practice of normalizing secondary ion mass spectrometry dopant profiles to a matrix signal must be modified for situations where matrix yield changes occur.

Ultra-high depth resolution SIMS for the interface analysis of complex low-dimensional structures

“…More serious roughening effects were demonstrated [12] where ion yields and erosion rates having apparently been stable for depths of lms could suddenly change with an accompanying loss of depth resolution. It was shown that for non-normal incidence of the primary beam the relatively smooth crater bottom collapsed into a pronounced ripple structure.…”

Ripple wavelength dependence on ion bombardment energy

“…Most of the ripple experiments done fall in this energy regime. In Figure 4, the Biersack-Ziegler nuclear stopping power (Biersack & Ziegler, 1982) is plotted as a function of the reduced energy E. From this figure it can be seen that S, scales in various regions approximately as Ar+ at 60°and 10" ions/em"; Stevie et al (1988) 0; at 39°to 42°; Vajo et al (1996) 0; at 40°; Smirnov et al (1999) In similar fashion as Bradley and Harper extended equation (1), it was also extended to include more topographical mechanisms, resulting in the noisy Kuramoto-Sivashinsky partial-differential equation, which can be written in generalized form as (Carter, 1999):…”

Development of B‐doped Si multiple delta‐layer reference materials for SIMS profiling

“…There was no degradation of the depth resolution parameters with the increase of sputter depth because there was no significant development of surface topography in this shallow range. 21 The average values of L , T , and FWHM are 0.21, 3.57, 1.70 and 6.75 nm and the relative standard deviations are 4.0, 1.6, 1.3 and 1.3% respectively. Depth resolution parameters measured at the second delta-layer are plotted as a function of ion energy in Fig.…”