New Approach for Improvement of Secondary Ion Mass Spectrometry Profile Analysis

Abstract

In this paper, we describe the improvement of secondary ion mass spectrometry (SIMS) profile analysis by a new approach based on partial deconvolution combined with scale-frequency shrinkage. The SIMS profiles are obtained by analysis of the delta layers of boron doped silicon in a silicon matrix, analyzed using Cameca-Ims6f at oblique incidence. These profiles can be approximated closely by exponential-like tail distributions with decay length, which characterizes the collisional mixing effect. The partial deconvolution removes the residual ion mixing effect. The contributions of high-frequency noise are removed by shrinkage to a great extent of the profiles. It is shown that this approach leads to a marked improvement in depth resolution without producing artifacts and aberrations caused principally by noise. Furthermore, it is shown that the asymmetry of the delta layers, caused by the collisional mixing effect, is completely removed, the decay length is decreased by a factor of 4 compared with that before deconvolution.

Preparation and characterization of Zn18O/Zn16O isotope heterostructure thin films

“…25 In the present analysis condition, the exponential decay length should be 5 nm, almost the same as the experimental decay length of the first component near the surface of approximately 6 nm. The shape of the depth profile near the surface is related to ion mixing by the primary ion beam during analysis.…”

Tikhonov-Miller regularization with a denoisy and deconvolved signal as model of solution for improvement of depth resolution in SIMS analysis

“…the input signal and the measured one) with a sufficient regularization. (6) Where H * is the conjugate of H and C represents the constraint operator which must be applied in the time domain after an inverse Fourier transformation. The accuracy of the solution is referred to the accuracy of the model which suggests a reasonable formulation.…”

“…Considering the well-known SIMS limitations, the measured profiles are broadened than the actual profiles, especially at the decaying parts, which are governed mostly by the collisional mixing [1][2][3][4][5][6]. Furthermore, the SIMS profiles exhibit an energy-dependent shift.…”

“…The origin of these oscillations is the presence of strong local concentrations of high frequencies in the signal which belong to noise. For this reason, it is important to eliminate noise components from the signal [6]. Denoising with the sole purpose of extracting desired information from measured data has proven to be a crucial preliminarily step in any analytical method.…”

Multi-Scale Deconvolution of Mass Spectrometry Signals

“…This DRF can be described by three parameters: λ u (the rising exponential decay), σ g (the standard deviation of the Gaussian function), and λ d (the falling exponential decay). The latter characterizes the residual mixing effect, which is considered to be the main mechanism responsible for the degradation of the depth resolution (Boulakroune et al, 2007;Yang et al, 2006). For any possible values of these parameters, the DRF is normalized to unity.…”