I am not a metallurgist but I would be extremely nervous purchasing gold bullion bars from a non-certified / traceable source. Maybe I am missing some basic science but how would I, as an amateur, design a fake bullion bar ? Gold is almost unique as it has a very high density of 19.30 g per cm cubed. A 100g Metalor bar for example measures 45 x 25 x 5mm Forget the difference between 22 carat and 24 carat and the fact that a bar will have rounded corners etc. The dimensions and weight can be accurately measured. An acid or electronic gold tester of the surface will show the material is 22ct or higher gold. This bar will be valued at around USD $4,500 How therefore do we cheat ? Tungsten metal has a density of 19.25 which is only 0.25% less than gold. A tungsten bar of the same dimensions as a Metalor gold bar would weigh 99.75 g. If the tungsten bar was only 0.01mm thicker the weight would be 100g. Use a Dremel to engrave a few marks of "authenticity" with a carbide or diamond tip which is harder than pure tungsten. Colour, hardness ( gold is 2.5 ( soft) and Tungsten 7.5 ( hard ) ) and of course the surface scratch tests - we don't want to scratch too much because we don't want to spoil the bar. The cost of this 100 gram bar of tungsten is ONLY about USD $4 ( yes FOUR US DOLLARS ) So we can afford to add some real gold to give the bar its proper shiny colour and to prove that it is genuine gold. Add a coating of gold to the tungsten bar - now I don't know how to do thick plate or maybe sputtering. Perhaps the tungsten needs a thin coating of Nickel or another metal to give a strong bond to the gold layer. Perhaps as tungsten's melting point is 3422 C compared to gold 1065 C it can be dipped in molten gold in a crucible or coated like chocolate bars, albeit in a furnace ? I am sure this would be relatively easy to find out. By sacrificing a gold sovereign ( value $320 ) weighing about 8g the tungsten bar would get a nice thick coating of gold about 0.14mm. That should pass the standard gold tests but add a half sovereign more gold and the coating will be almost 0.2mm thick. Do we scratch that deep ? Maybe seal the bar in thermal shrink-wrap plastic so the buyer doesn't want to test it thinking it will loose value if opened. Estimated cost of materials - less than $500 Sell on an internet auction site to rookies for about $4700 so a nice profit of $4000 per bar after costs. This to me seems very scary and I would never consider buying a bullion gold bar from a non-certified dealer. Perhaps the science doesn't work but I have a horrible feeling that this is not too much of a challenge. Other than drill holes through the bar and take core samples how would you ever spot such a fake ? Of course this would be fraud / counterfeit but how could you sleep at night after you've congratulated yourself as the highest bidder on the web auction sites from someone not directly in the precious metals business ?
yeah that is I suppose how the amateurs do it... I wonder how the governments do it. Specific gravity test should expose? How different would the results between the two (fake and real) bar be?
Which you should still get tested. Sovereigns and old US gold in particular are susceptible to jewelers copies, which might contain less gold than they should.
The XRF wont test the middle of the bar......... but in saying that when i test bars i test all the way around and get an average.... Since gold is very dense to have silver in the middle and a very thick layer of gold on the outer would be not for the novice person to achieve. Correct moulds, crucibles and pouring is very crucial.
It would require a fair degree of skill. The tungsten core and gold casing would have to be just so. Pouring the gold around without a seam is hard. Only way I can imagine is if the tungsten is suspended with thin gold wire or a few small pieces of shot and gold poured around encompassing it to a few millimetres. 100g? technically challenging. 10oz bars, definitely more profitable for the same effort.
Ah, for some more free time.. Since getting into silver I've been considering developing a hypersonic spectrum analyser for exactly this purpose - to quickly examine the SG of the core of a coin or bar, and if there are any "layers". Seismic on a minute scale if you will... Technically feasible, but more R&D hours than I have at the moment... Oh well, scales and water for now..
The Perth Mint encapsulated 1 oz bars would seem pretty safe to me. Although, I guess you would have to take them out of their packet to test them. So selling fakes would be easier as people mightn't want to open them to check.
Buy an unsealed gold cast bar instead. Use a cheap ultrasonic tester on the cast bar. I got one for less than usd200. It's pretty accurate if you disregard the few mm of error.
The latest XRF's that are fitted with the new software can tend to pick up a bit of background noise when testing in some capsules (especially thicker old square Perth mint capsules.) due to the wider range of elements they test for. Hence the need for training in order to become familiar with the technology. It ain't as point and shoot as some seam to believe.
Fake sovereigns are pretty easy to spot if you know what to look for. but, Only buy from reputable dealers, buy in 1 oz coin form / sovereigns/qtr oz coins and always keep the receipt !. And for me.. I wouldn't buy anything from anyone other than a reputable dealer/mint. Nor anything that comes in sealed packaging (like a 1oz buffalo coin - someone should unwrap and test one )
Key thing when using XRF testing through plastic is that there is no Fe or Cu or even. The plastic shows up a bunch of trace elements, including Palladium, Rhodium and similar rare metals. Never seen iron or copper (or nickel).