The problem of fake gold bars (2012)
blogs.reuters.com
blogs.reuters.com
I wonder if the acoustical properties of a fake bar would be easy to distinguish from solid gold.
Actually, I'm agreed that you could measure the density, but somebody could fiddle with the density too by adding just a touch of this or that to bring everything up to the right weight.
If you're willing to melt them down though, you probably don't have to deal with mass spec as some property will have to give (melting point, density, etc).
It should be straightforward to make an acoustic time of flight go/no go meter you could place on a gold bar. Piezo excitation transducer/fixed distance/piezo receiver/process signal propagation velocity -> light up red or green light.
I'm not sure this is easy to achieve! A best case commercial speaker/microphone will have a ~100khz bandwidth, which gives ~10 microsecond resolution. If you're aiming for high accuracy you also need to account for thermal variations in speed of sound, which would require a some precise thermometers.
I'm afraid that either myself or the author doesn't fully understand fluid displacement.
Given two gold bars, one adulterated with Tungsten, and one not, either the weight will differ or the volume will differ. So, given equal weights (i.e. standard 400oz bars), you can check the displacement to establish purity.
If you use a container that is filled up to a spout, and you measure the displaced water by collecting it in a measuring cup, you can just directly calculate density, and then compare it to gold.
Tungsten is less dense than gold. Not as much a difference as actual twinkie filling, but either way you're stuck with a similar issue.
EDIT: tungsten + platinum looks viable as sibling comment mentioned.
19.3 * 100 = 19.25 * x + 21.4(100-x)
gold is left, mix is right (a sum of the tungsten and platinum parts). I solve for 100 grams because voila, it's now a percentage. That solves to x = 97.7273 (you get to use 97.7273 grams of tungsten) or you just need 2.2727% platinum. (which recall costs about the same as gold.) so if this were a great test people could just use an alloy, which apparently exist:
https://www.google.com/search?q=tungsten+platinum+alloy
and target the exact same density as gold. tungsten alone seems close enough for them, at least, according to the author.
The article also says that a micro-ohm meter can detect such replacement.
As a glass flameworker, I and my colleagues buy gold, cut it up, stick it on rods of glass, and put it on the flame to deposit gold vapor onto glass. We would notice fake gold right away, but I haven't heard of that happening.
(Though, the typical flameworker doesn't buy whole 1kg gold bars!)
There are certification authorities that do exactly that. Here in Spain you can hire them, and they will melt the gold in front of you and use expensive instrumental to know how much gold a gold bar has.
We also have the concept of "notary" that is a person whose job is to certify something by law. They are trustworthy.
It is individuals who should be afraid of being tricked. But you have at your disposal cheap microbalances, micro ohm meters(I made myself one for like USD30), 3d scanners(to calculate the volume as using water is tricky because surface tension of water), quartz transducers to measure speed of sound.
Which may be because the Fed didn't actually have all that gold available, nicely separate in a vault labelled "Germany", but even if that is true, the case is rather more complex than just "The Fed embezzled the gold that Germany gave them for safekeeping". Because Germany never gave that gold - they never actually had it. It's in fact the German export surpluses from the time before the USA went off the gold standard.