Investigation of the Composition of the Si/SiO₂Interface in Oxide Precipitates and Oxide Layers on Silicon by STEM/EELS

Abstract

We investigated thermal oxide layers of different thickness on (100) and (111) silicon substrates by STEM/EELS to determine the stoichiometry profiles and compared these with stoichiometry profiles of plate-like and octahedral oxide precipitates in silicon. It was found that the stoichiometry of SiO x (x = 2) cannot be reached if the oxide layer thickness is lower than 10 nm for thermal oxides grown at 900 • C. This is due to an interface layer of equal maximum slope of x for all oxide layers. The slope of x is the change in stoichiometry with position and was obtained from fitting by sigmoid functions. Similar results were found for the oxide precipitates in silicon. However, there are arguments which question the slope determined via the low loss EEL spectra and the maximum x value could be closer to 2 in reality. On a sample with an oxide layer of 13.9 nm thickness we compared stoichiometry profiles obtained from the plasmon region and the Si-K 2,3 and O-K ionization edges. The width of the interface measured on stoichiometry profiles decreases with increasing energy loss and is lowest for the O-K ionization edge with a width of 1. In a previous work, we found that both the interface between an oxide precipitate and the surrounding silicon matrix and the interface between a silicon substrate and a thermal oxide layer are of the same nature.1 In both cases, between SiO 2 existing in the center of the precipitate and in the oxide layer and Si of the matrix and the substrate a suboxide region of 2-3 nm was found. These results were obtained by electron energy loss spectrometry (EELS) carried out by scanning transmission electron microscopy (STEM). In the low loss region, it is possible to distinguish between Si, SiO, and SiO 2 which all exhibit different maxima of the plasmon loss energy. By deconvolution, the local composition of the phase can be determined with the help of reference spectra of the three components.The stoichiometry of the oxide precipitates (SiO x ) was debated for a long time and application of different methods on different samples containing oxygen precipitates resulted in different values for x ranging from 1 to 2. Recently, it was demonstrated by applying several direct and indirect methods that the oxide precipitates consist of SiO 2 .2 Inspired by the results in Ref. 1 and 2, a layer model was proposed explaining the different experimental values for x for oxide precipitates of different geometry.3 Especially, for plate-like precipitates with a small aspect ratio the lower x values could be explained assuming that they consist of SiO 2 surrounded by a 2 nm SiO shell. However, neither is it known until now if all oxide precipitates consist of SiO 2 in the center nor is it known how thick the suboxide region for oxide precipitates grown at different temperature and time is. It needs to be mentioned that the suboxide region is not a layer of SiO composition but a profile of x increasing from the silicon matrix to the SiO 2 core. The impression of a layer just stems from the analysis of ele...

Radiation-induced defect accumulation and annealing in Si-implanted gallium oxide

“…The spectra revealed a change in the Si L2,3-edge lineshape in one of the dark contrast regions (area 6) in the ADF image from the annealed sample (Figure 7c)). In the EELS spectrum from area 6, the L-edge lineshape was consistent with previously reported results from silicon dioxide 26,28,29 . Hence, EELS results indicate that the Si-accumulation observed in EDS was likely due to the formation of SiO2 particles during annealing.…”

Compositional gradients at the nanoscale in substoichiometric thin films deposited by magnetron sputtering at oblique angles: A case study on SiOx thin films

“…In addition, spectral features in the form of shoulders can be detected in the spectra of intermediate zones where the edge is not sharp but adopts an increasing relatively low slope profile. According to the literature, this feature can be attributed to the presence of Si + and Si 3+ oxidation states, [ 45,48,49 ] which supports the existence of a certain progression in the oxygen content along the observed profile. In this way, a transition can be seen from an O‐rich to Si‐rich region along the first tens of nanometers from the left, while a subsequent smooth transition along the remaining part of the profile from an Si‐rich to O‐rich region is detected at the right side.…”

Energy-Loss Near-Edge Structures and Low-Loss Structures of Polymers in a Solid Electrolyte Interface Formed from Fluoroethylene Carbonate on a Si Anode Clarified by DFT Calculations

“…Comparing the spectrum acquired at point β 2 (Figure b) with the reported spectra of Si and Li 7 Si 3 reveals that the SiSnTi anode material is completely delithiated. The spectrum acquired at point β 1 exhibits a peak at 18.9 eV that corresponds to SiO (Figure b), indicating that the thicker O layer compared with the C layer on the SiSnTi surface, as observed in the C and O K-edge maps (Figures b,c), is attributable to the contribution of O in SiO. The appearances of both surface SiO (Figure b) and Li 2 O (Figures b,c) confirm that Li can react with SiO, thus reducing this surface oxide according to the following reaction: SiO + 2Li → Si + Li 2 O .…”

“…(a) O K-edge mapping of the SiSnTi–SEI interface and (b) low-loss spectra acquired at points β 1 and β 2 in panel (a) in comparison with the reported spectra of SiO, Si, and Li 7 Si 3 . (c) Loss functions for cross-linked poly­(FEC), poly­(VC)–(CH 2 CH 2 O) n branched at a C­(OR) 2 binding site, poly­(FEC), and LEDC in comparison with experimental low-loss spectra of the outer and inner SEI (regions γ 1 and γ 2 in panel (a), respectively).…”