Structural differences between W, WC, and W2C?

[Sputki]

In my research group, I'm tasked with determining the purity of a tungsten sample sent from a supplier. Using FIB, we cut a 20 by 20 by 20 micron well into the sample and tested a wall of the well about 10 microns down from the surface.

After taking a spectrogram, there seemed to be a large amount of carbon, 66% W with 33% C by atomic percentage (tungsten carbide/tungsten semicarbide?)

Although we used 2 small strips of carbon tape to mount the sample (cylinder 2 cm tall 0.5 cm diameter), we don't think there should be that much carbon on the sample. I plan to do a simple density calculation to roughly determine purity tomorrow.

What is the next step in determining purity? Is there an obvious simple test I can perform (preferably without using expensive machinery first)? Is tungsten carbide a common contaminant of tungsten? Could this just be a small layer of WC on pure W?

This link is to the SEM and EDS images themselves http://imgur.com/a/qIwIL
(You're looking at a close up of the wall of the well)

 

[Sputki]

As a follow up, the tungsten cylinder was measured (rather crudely with a caliper but with a precise scale) to have a density of 22 g/cm^3, which (imho) is far from the supplier's guarantee that the sample was 99.9% pure W and as such would have an i(deal) density of 19.3 g/cm^3 at room temperature (which was calculated to be 19.5 g/cm^3 at 25 C).

Could the FIB work have this big effect on the sample (which had 1.2 cm hight with 0.3 cm diameter)? Could WC have formed on the surface?

Should I bring up the discrepancy with my PI? (again, actual FIB SEM work is in an imgur album in the original post).

 

[DrDu]

Given that the density of tungsten cabide is lower than that of tungsten, and a density as high as 22 g/cm^3 is only reached by osmium and iridium, I have some doubts concerning the precision of you density determination.