Tom's Hardware Recovers Gold And Silver From CPUs

[micropage7]

In our first step, the processors are dropped into a bath of concentrated nitric acid. They’re going to spend a while there, as we'll see. And the experience won't be particularly pleasant for them, either.



During this stage, the strong acid reacts with certain metals, particularly copper and silver. Nitric acid has no effect on gold, however.

Interestingly, our K5 doesn’t seem to react to the forced bath...yet.



Let’s throw in a few more processors. On the left, our CPUs have just been tossed into a concentrated nitric acid solution. On the right, after a few minutes, the solution is already taking on a bluish color, characteristic of the presence of copper(II) nitrate, and thus of the Cu2+ ion. It's also important to note that the bath is now giving off a brown gas (nitrogen dioxide, or NO2).

Cu + 4 HNO3 -> Cu(NO3)2 + 2 NO2 + 2 H2O


The reaction can be more rapid (and violent) at times, but it’s always dangerous.


After a few weeks, the solution is very blue, due to the strong presence of Cu2+ ions. The copper and silver (among other things) are dissolved, but not the gold


Now we remove the processors from the bath.Needless to say, they wouldn’t perform quite as well now...


The next step is to filter. Silver is in the solution as silver nitrate (AgNO3), whereas the gold and a few other metals that weren't attacked by the nitric acid, plus the impurities, remain in the filter. We’ll set the filter aside for later and concentrate on the solution.


We add a little sodium chloride (ordinary table salt) to the solution.


It’s magic (well, almost)! The silver in the silver nitrate precipitates and forms silver chloride.

AgNO3(aq) + NaCl(Aq) -> AgCl(s) + NaNO3(aq)


Now we add a little hydrochloric acid. Then, we immerse a piece of zinc in the same solution. Attention: This is an exothermic reaction!


After a few hours, metallic silver becomes noticeable. We also see that the silver chloride precipitate has begun to blacken in the light. The zinc has reacted with the hydrochloric acid to form dihydrogen (and zinc chloride).

Zn + 2 HCl -> H2 + ZnCl2

The dihydrogen will then reduce the silver chloride;


We filter the solution and rinse everything. The result is metallic silver! Let’s let the powder dry...


Now, all we need to do is melt the silver in a crucible with an oxy-butane torch.


And voilà: a pretty little silver coin! Of course, it’s not 100% pure, which is impossible with this kind of do-it-yourself process, but it’s reasonably close. We suspect the presence of platinum, palladium, and rhodium. Let's continue.


We haven't finished with our CPUs yet, though. We still need to get them to give up their gold!

After a short rinsing with distilled water, we put them in a bath of hydrochloric acid and 3% hydrogen peroxide, in a proportion of two to one


You can see the very fine gold wires around the processor die separating. We’ll leave the processors in the bath for a few hours...


After a few hours, we filter the solution to recover the gold flakes and various waste substances.


What’s left behind in the filter is a mixture of various metals and impurities. We put everything in a mixture of 35% hydrochloric acid and 5% chlorine bleach, in a proportion of two to one. The reaction is highly exothermic and produces extremely dangerous chlorine gas (Cl2)!

2 HCl + NaClO -> Cl2 + NaCl + H2O


We filter everything one more time. The filter retains all the impurities, and nothing is left in the solution except gold(III) chloride.


To recover the metallic gold, we need to get the gold that’s in solution to precipitate. For that, we use powdered sodium metabisulfite. In the presence of water, the sodium metabisulfite produces sodium bisulfite.

Na2S2O5 + H2O -> 2 NaHSO3


The sodium bisulfite is what will allow the gold to precipitate.

3 NaHSO3 + 2 AuCl3 + 3 H2O -> 3 NaHSO4 + 6 HCl + 2 Au

Now we need to recover the powder by filtering the solution, then allow it to dry.


It still doesn’t look like gold, and yet…


After drying the gold powder, we need to melt it in a crucible. Our oxy-butane torch can do the job, since the melting point of gold is around 1064°C


The result is a pretty gold BB! As with the silver, our homemade process can’t extract perfectly pure gold, which is one reason why industrial operations use other, more dangerous processes. But that’s another story...

http://www.tomshardware.com/pictures...ilver-cpu.html

[animal007uk]

Watched a program called bang goes the theory and they did the same thing as this, Was very intresting, The guy went round the scrap yards getting all the old pc parts.

[Millennium]

Pretty cool, I wonder was all that gold from a single processor or all the processors they show? Probably the latter I guess.

[animal007uk]

Originally Posted by Millennium

Pretty cool, I wonder was all that gold from a single processor or all the processors they show? Probably the latter I guess.

From all of them

[micropage7]

i remember several years ago theres a rumor that these coins contain gold

I hope no one decides to try this at home, although I should really know better than to think that. Aside from being pretty dangerous, I'm sure the cost of the chemicals would outweigh the value of the relatively impure metals you extract.

[Damn_Smooth]

Originally Posted by twilyth

I hope no one decides to try this at home, although I should really know better than to think that. Aside from being pretty dangerous, I'm sure the cost of the chemicals would outweigh the value of the relatively impure metals you extract.

There you go fucking up my fun. I was just getting ready to blow some shit up.

[95Viper]

Thanks, micropage7...

Nice find on this.
Not for me, though, I had my fill of chemistry years ago.

[Super XP]

Todays PC parts have a lot more gold than in them