A correlation of Auger electron spectroscopy, x‐ray photoelectron spectroscopy, and Rutherford backscattering spectrometry measurements on sputter‐deposited titanium nitride thin films

Abstract

Articles you may be interested inTungsten silicide composition analysis by Rutherford backscattering spectroscopy, Auger electron spectroscopy, and x-ray photoelectron spectroscopy Analysis of iridium-aluminum thin films by xray photoelectron spectroscopy and Rutherford backscattering spectroscopy J. Vac. Sci. Technol. A 8, 2251 (1990); 10.1116/1.576745The characterization of titanium nitride by xray photoelectron spectroscopy and Rutherford backscattering J. Vac. Sci. Technol. A 8, 99 (1990); 10.1116/1.576995Auger electron and xray photoelectron spectroscopy of sputter deposited aluminum nitride Auger and electron spectroscopy for chemical analysis (ESCA) data ofTiN x were analyzed as a function of film composition as established by Rutherford backscattering spectrometry (RBS). The overlap of the N(KVV) and Ti(LMM) Auger transitions necessitated the assessment of two methods previously proposed for the derivative spectra. These results were compared with peak height and peak area measurements (after background subtraction) of the well-separated N ( Is) and Ti (2p) ESCA photoemissions. Neither the Auger nor the ESCA N/Ti intensity ratios scaled linearly with the N/Ti compositional ratios determined by RBS, especially for low nitrogen content. This behavior most likely results from ion-bombardment-induced losses of nitrogen in those phases with dissolved nitrogen rather than from an increased satellite emission in the Ti (2p) spectra from the near-stoichiometric nitride. In terms of precision and analysis speed, the Auger peak-to-peak quantification methods are preferred over ESCA quantification. In the nearstoichiometric phase (N/Tiz 1), RBS analysis shows higher sensitivity to nitrogen compositional changes than either ESCA or Auger.

Influence of the plasma nitriding conditions on the chemical and morphological characteristics of TiN coatings deposited on silicon

“…The A5 sample does not present either N1s nor Ti 2p spectra. The most prominent one for all samples (except A4) has a binding energy in the range of 396.4 -396.9 eV, and is associated to TiN [37][38][39][40][41] . The component at lower binding energy (395.5 -395.9 eV) can be attributed to N-C 41 ; the one at 398.2 -398.8 eV, to N-O (in TiO x N y ) 40,41 ; and the one that appears at higher binding energy for some samples could also be related to N-O bonds in an oxynitride compound 41 .…”

“…14). The component at 455.3 -455.5 eV is related to TiN [37][38][39][40][41] ; the one at 456.6 -456.8 eV, to TiO x N y ; the one at 458.5 -458.6 eV, to TiO 2 40,41 .…”

Determination of the Composition of TiNx and TiBxNy Films by AES

“…Much work has of been undertaken to establish a simple and accurate method for the determination of the composition of TiNx Films. Many techniques, principally Auger Elec tron Spectroscopy (AES), X-ray Photoelectron Spec troscopy (XPS), Rutherford Backscattering Spectros copy (RBS) and Energy Dispersive X-ray Analysis (EDX) have been considered [1][2][3][4][5][6][7][8]. None of the techReprint requests to Dr. J. Haupt.…”

“…EDX also has a large escape depth and this, combined with the overlap of the Ti LI and N Ka lines and the low probability of the N Ka emission causes difficulties with quantification. The more surface sensitive electron spectroscopic techniques are normally prefered for thin film analysis, but for TiNx problems of peak overlap and peak fit ting in AES and XPS, respectively, have resulted in many different approaches [1,2,5,6,8]. For XPS, the unusual structure in the Ti 2p region between the Ti 2p1/2 and Ti 2p3/2 lines has been ascribed to satel lite structure or the presence of oxygen which varies as a function of N concentration, causing uncertainties in the peak area measurements [1,[6][7][8].…”

Native target chemistry during reactive dc magnetron sputtering studied by ex-situ x-ray photoelectron spectroscopy

“…2(b)) represented with Voigt functions. While the origin of these features is under discussion 36,37,38,39 appealing evidence points towards shake up effects due to intraband transitions between filled and partially filled valence electron energy levels near the Fermi energy, 40 and a decrease in the screening ability of the conduction electrons leading to "well-screened" and "poorly-screened" XPS peaks. 32 In order to extract physically meaningful information, the line shapes for pairs of Ti, TiN, and TiN-sat peaks, 2p3/2-2p1/2 energy splitting and the 2p3/2/2p1/2 area ratios are first obtained by fitting the reference Ti and TiN spectra, and then used as input to construct more complex peak models like the one shown in Fig.…”