Towards the ultimate limits of depth resolution in sputter profiling: Beam‐induced chemical changes and the importance of sample quality

Abstract

Recent progress in materials characterization by sputter profiling is discussed, with particular emphasis on a critical evaluation of the parameters that determine the depth resolution. The net effect of beam‐induced material transport and sputtering can be studied in a straightforward manner using abrupt doping distributions embedded in the matrix of interest, parallel to its flat surface. It is shown that the quality of the data that can be obtained with modern instruments is sufficient to identify lack of sample quality in terms of either surface flatness or abruptness of the doping distributions. Except for the presence of special artefacts, the depth resolution is always improved by using as low a bombardment energy as possible. The optimum angle of beam incidence depends on the crystalline structure of the sample as well as on the chemical identity of the primary ions. With oxygen primary ions a rather complex situation is encountered. Compared to inert gas ions, an improvement or a degradation in depth resolution may be achieved, depending on the impurity–matrix system under study. The merits of sample rotation are also discussed.

SIMS Analyses of III-V Semiconductor Quantum-well and Superlattice Heterostructures

“…They do not actually coincide due to energy-and angledependent changes in the SIMS resolution function. 2,4 In agreement with many other works in the literature regarding SIMS depth proÐling of delta-doped distributions,2h8,19 the leading and trailing edges of the measured indium signal have an exponential behaviour characterized by a leading edge decay length j u (upslope) and a trailing edge decay length (downslope). j d The inset in Fig.…”

“…4 Nevertheless, several authors in latter years have shown that by using appropriate experimental conditions and particular technical and mathematical procedures, SIMS can reach high levels of performances in the analysis of very sharp elemental distributions.4h8…”

Concentration-depth calibration and bombardment-induced impurity relocation in SIMS depth profiling of shallow through-oxide implantation distributions: a procedure for eliminating the matrix effect

“…32 The results of Fig. 18 O 2à gree with the previous Ðndings in that the 0.5 keV proÐles measured at 0¡ and 75¡ exhibit essentially the same exponential tails. Hence, in terms of bombardmentinduced mixing, measurements with oxygen primary ions at rather oblique beam incidence (75¡) do not provide an advantage compared to 0¡.…”

“…This lack of an improvement can be attributed, at least in part, to the favourable oxygen "bu †erÏ e †ect achieved at normal incidence. 32,56 In this context it is also worth mentioning that improvements in depth resolution, to be achieved by operating at oblique impact angles h, are often overestimated in the literature. It has been shown recently,57 for 0.6È2 keV bombardment at 44È80¡, that the h O 2 dependence of j(Ge in Si) has a form that may be written as with j(h) \ j 0 (sin2h s ] cos2h)0.5, sinh s \ 0.36.…”

“…the same as the decay length that has been reported for B delta distributions. 32 This Ðnding implies that, at least in the Si substrate, the proÐle shape is fully determined by bombardment-induced redistribution e †ects and does not reÑect the original implantation distribution any more. In order to identify the true proÐle shape for B implanted at 0.5 keV, one has to perform depth proÐling at much lower beam energy, preferably not exceeding half the implantation energy.…”