Sciencemadness Discussion Board

Adulteration of gold with tungsten - continued

JohnWW - 27-12-2009 at 13:34

A few weeks ago, I contributed to a thread about gold refining, which, however, seems to have completely disappeared from this site; no trace of it could be found by Google searches of the site with the keywords used in it.

One of my posts on the thread noted a recent news item about claims that some of the gold bars stored in Fort Knox, Tn., and elsewhere are adulterated by tungsten, possibly as a powder due to tungsten being much harder and having a much higher melting-point than gold. This is made undetectable by simple density measurement using Archimedes' principle, and indeed by any non-destructive testing method, by the fact that gold and tungsten have the same density, 19.3; unless flakes of tungsten were on the surface of the bars, in which case arc emission spectroscopy, neutron activation analysis, and X-ray fluorescence spectroscopy may detect it.

I have now found the original news item about this claim; here it is:
http://www.businessinsider.com/the-tungsten-filled-gold-bar-...

hissingnoise - 27-12-2009 at 13:45

Gold-plated tungsten?
It sounds hilarious enough to be true, and on top of that, they seem to have a Russian tank sitting on the dais!


JohnWW - 27-12-2009 at 15:40

By the way, a year or two ago I heard a story about some gold bars that were shipped from Ethiopia's central bank to South Africa by air to pay for something. It was found that the "gold bars" were in fact made of cast iron plated with gold, having been made using a mold used for casting gold bars, and even carrying the assayer's logo and "analysis". Of course, the fraud was quickly detected when the bars arrived in South Africa, due to iron having less than half the density of gold. It is not known whether the Ethiopian central bank was aware of the fraud.

[Edited on 28-12-09 by JohnWW]

bbartlog - 27-12-2009 at 15:50

There are other non-intrusive ways of detecting such adulteration. Electrical conductivity would be different, probably magnetic properties as well. Whether these are part of a standard assay I don't know, but it wouldn't be hard to add such tests if it seemed necessary.

Fleaker - 28-12-2009 at 07:20

If tungsten were interspersed with gold, the detection would be quite simple. Anyone who has ever held one such bar of gold (or any bar of commercially pure gold) would be able to tell the difference. One would have to saw the bar in half, hollow it out, and then insert a tungsten insert that is machined precisely.

Recall also that most sintered tungsten is significantly below its theoretical 19,3 g/cc density by the very virtue of its manufacturing process. Good luck with melting it in any quantity like such needed for duping good delivery bars.

A standard assay is a fire assay, which may or may not turn up the adulteration (probably will, if random drillings are taken). X-ray fluorescence is another popular method, but it's really a surface technique. ICP would easily show tungsten present in the bar, if the tungsten were added to cooling or still molten gold.

Rosco Bodine - 28-12-2009 at 11:18

Maybe all the bullion should be recast with authentication markings to restore public confidence. What I was thinking was maybe a relief type engraving
displaying a bust of Barney Frank sporting a big smile, and beside it the
inscription Fannie Mae and Freddie Mac certified fine gold , guaranteed tungsten free, accept only genuine, take no queer ! There should be a hologram surface effect when viewed obliquely, revealing fairies in flight using their wands to
sprinkle fairy dust upon marijuana leaves by the light of a full moon.

12AX7 - 29-12-2009 at 04:05

I'm thinking metal-metal composite, tungsten powder dispersed in gold. Fine particles (micrometer sized) won't affect the fluidity of the metal much, so it can still be cast. They will be fairly disperse due to brownian motion, though there may still be a higher concentration at the top of the bar.

Hardness test would be very simple. Adulterated material will be considerably harder than pure gold. Sensitivity should be a few percent; although gold is managed to the parts-per-thousand scale, it wouldn't really make sense to cut it less than a few percent, so that should be sufficient. Although a hardness test isn't non-contact, it does not remove material, and the dents made can be arbitrarily small.

(Incidentially, because tungsten has such a high melting point, it is only "soldered" (metallurgically speaking) by ordinary nonferrous metals like copper, silver and gold. Such composites are popular for robust connections, like motor brushes, where high abrasion resistance, electrical and thermal conductivity are required.)

Tim