ASTM F1894-98(2003)

1.1 This test method covers the quantitative determination of tungsten and silicon concentrations in tungsten/silicon (WSIx), semiconductor process films using Rutherford Backscattering Spectrometry (RBS). (1) This test method also covers the detection and quantification of impurities in the mass range from phosphorus A (31 atomic mass units (amu) to antimony (122 amu).

1.2 This test method can be used for tungsten silicide films prepared by any deposition or annealing processes, or both. The film must be a uniform film with an areal coverage greater than the incident ion beam (~2.5 mm).

1.3 This test method accurately measures he following film properties: silicon/tungsten ratio and variations with depth, tungsten depth profile throughout film, WSIx, film thickness, argon concentrations (if present), presence of oxide on surface of WSIx films, and transition metal impurities to detection limits of 1 x 10 14 atoms/cm2.

1.4 This test method can detect absolute differences in silicon and tungsten concentrations of +/- 3 and +/- 1 atomic percent, respectively, measured from different samples in separate analyses. relative variations in the tungsten concentration in depth can be detected to +/- 0.2 atomic percent with a depth resolution of +/- 70A.

1.5 This test method supports and assists in qualifying WSIx films by electrical resistivity techniques.

1.6 This test method can be performed for WSIx films deposited on conducting or insulating substrates.

1.7 This test method is useful for WSIx films between 20 and 400 mm with an areal coverage of greater than 1 by 1 mm.

1.8 This test method is non-destructive to the film to the extent of sputtering.

1.9 A statistical process control (SPC) of WSIx films has been monitored since 1993 with reproducibility to +/- 4%.

1.10 This test method produces accurate film thicknesses by modeling the film density of the WSIx film as WSI2 (hexagonal) plus excess elemental SI2. The measured film thickness is a lower limit to the actual film thickness with an accuracy less than 10% compared to SEM cross-section measurements (see 13.4)

1.11 This test method can be used to analyze films on whole wafers up to 300 mm without breaking the wafers. The sites that can be analyzed may be restricted to concentric rings near the wafer edges for 200-mm and 300-mm wafers, depending on system capabilities.

1.12 This standard does not purport to address all of the safety concerns, if any, associated with its use. It is the responsibility of the user of this standard to establish appropriate safety and health practices and determine the applicability of regulatory limitations prior to use.