recent selected publications from the HyMet project

Publication single view

Title:

Quantitative depth-profile analysis of transition metal nitride materials with combined grazing-incidence X-ray fluorescence and X-ray reflectometry analysis
Author(s): Simona Torrengo, Diane Eichert, Y. Mazel, M. Bernard, Yves Ménesguen, Marie-Christine Lépy and Emmanuel Nolot
Journal: Spectrochimica Acta Part B: Atomic Spectroscopy
Year: 2020
Month: September
Day: 1
Volume: 171
Pages: 105926
DOI: 10.1016/j.sab.2020.105926
ISSN: 0584-8547
File URL: http://www.sciencedirect.com/science/article/pii/S0584854719305920
Web URL: https://doi.org/10.1016/j.sab.2020.105926
Abstract: Refractory transition metal nitrides are attracting attention in many microelectronics research and development fields. Their different applications and desired performances require a precise control of the film thickness and elemental depth profile. This information can be non-destructively obtained combining Grazing-Incidence X-ray Fluorescence (GIXRF) and X-ray Reflectometry (XRR) analysis. GIXRF-XRR joint analysis can be performed following a reference-free approach, which requires the knowledge of the whole experimental system. However, such a rigorous metrological approach cannot be easily applied to in-lab or in-fab tools due to the difficulty to access the various experimental set-up parameters, therefore another approach is needed. In this paper, we used a reference-based method consisting in the use of known standard samples, close in composition and structure with respect to the samples to analyse, to deduce the key parameters of the instrumental setup. These are then used to perform quantitative GIXRF-XRR combined analysis of the samples of interest. We applied this method to titanium tungsten nitride thin films elaborated by Physical Vapour Deposition (PVD). Our results show that the reference-based method, using carefully determined instrumental functions, can meet the requirement for both qualitative and quantitative depth-profiling analysis. The reference-based approach result easier to implement into labs and fabs than the reference-free method.

Back to the list view