| Pages:
1
2 |
garage chemist
chemical wizard
Posts: 1803
Registered: 16-8-2004
Location: Germany
Member Is Offline
Mood: No Mood
|
|
Recovery of mercury and silver from dental amalgams
Some time ago, I got 200g of old dental amalgams (leftovers, I bought them from someone who had a shitload of those).
They consist of 53% mercury and 12,5% silver, the rest being copper and tin.
Now, the obvious question: how can I get out the mercury and silver?
Dissolution of the amalgams in nitric acid would yield an insoluble residue of SnO2 (correct? Would this be hard to filter or something?) and get
Hg2+, Ag+ and Cu2+ into solution.
Diluting the solution and adding HCl would precipitate Ag as AgCl.
If I use too little HNO3, some Hg2Cl2 will precipitate too. How can I be sure that all Hg is oxidised to Hg2+? Will using enough HNO3 suffice, or is
there a convenient oxidising agent that can oxidise all (Hg2)2+ to Hg2+ after dissolution?
Now I will have a solution of Cu2+ and Hg2+. This is my problem. How do I get pure mercury from that?
Adding iron filings (after neutralisation) is a viable method to get metallic mercury from its solutions, but copper would precipitate, too, leaving
me with copper amalgam.
I have no idea for selective precipitation of Hg2+ from Cu2+ or vice versa. Any ideas?
Another- and perhaps much better- approach would be to distill off the mercury from the amalgam as the first step.
I would do this in a bent test tube, in the same way as I often distilled white phosphorus from red P.
The stopper will carry a small glass pipe filled with glass wool and sulfur, to absorb any Hg vapors that would escape due to thermal expansion of the
air inside.
Apart from the toxicity (I'll do that either in my fume hood or outside, and if I do it outside I'll have to get a gas mask rated for Hg fumes) I can
see two problems:
1. Not all mercury will be possible to distill off. The amalgam composition has been specially designed to bind the mercury as tight as possible, and
the metals could hold back the mercury in the same way as e.g. sulfuric acid holds back water during distillation. Thus, high temperatures will be
needed, and the residue after distillation will contain mercury residues which will require special handling in subsequent isolation of the silver.
2. The mercury will not condense as droplets, but as a mirror. I have seen it doing so in the classic experiment "thermolysis of mercuric oxide".
This would be very inconvenient, as it would make simple pouring ot of the mercury puddle impossible. I would have to use some kind of scraper to
scrape out the mercury, and it will be a mess. Are there any methods to make mercury stop forming a mirror on glass and flow together?
The mercury wil of course be redistilled to purify it.
The metal residues will be dissolved in HNO3 and worked up for silver. The Hg2+ in the solution will be precipitated as HgS after precipitation of
AgCl.
Now if there are methods to selectively dissolve CuS and leave behing HgS it would be convenient. I think heating with dilute HCl will dissolve the
CuS. I'm going to try this out with pure CuS.
Any advice?
|
|
|
Eclectic
National Hazard
Posts: 899
Registered: 14-11-2004
Member Is Offline
Mood: Obsessive
|
|
Is mercury oxide soluble in aqueous ammonia? Copper hydroxide is...
|
|
|
garage chemist
chemical wizard
Posts: 1803
Registered: 16-8-2004
Location: Germany
Member Is Offline
Mood: No Mood
|
|
Hmm, good idea.
I'll look that up or try it out.
|
|
|
The_Davster
A pnictogen
Posts: 2861
Registered: 18-11-2003
Member Is Offline
Mood: .
|
|
What form are the amalgams in? Are you able to cast electrodes from it? Electrolysis of a HCl solution with the alloy anode and copper cathode has
allowed me to separate numerous alloys. What I have separated before are just binary alloys, with 4 metals it could go weird and require multiple
electrolysisis(what is the plural of electrolysis anyway?). I can give more info on this, if you are interested, in a few days, I am in the middle of
some midterms again, and could write pages on my alloy separations(currently still working on a semi-continous process for separating Sb and Sn from
solder).
Based on experiance, SnO2 from tin+nitric is easy to filter.
Electrolysis of the Hg2+ and Cu2+ would give Hg metal initially, and likely a minor copper impurity especially if you kept the electrolysis going too
long. A cell design which would not allow the Hg to puddle below the cathode but rather flow away would reduce copper contamination.
Copper does not amalgam so easy with Hg from what I have read, Pt electrodes are connected to copper wire in a lot of old electrochemical procedures
with a little puddle of Hg in a glass tube.
[Edited on 12-2-2006 by rogue chemist]
|
|
|
Nerro
National Hazard
Posts: 596
Registered: 29-9-2004
Location: Netherlands
Member Is Offline
Mood: Whatever...
|
|
Electrolyses = plural
If you have persulfate you could oxidize everything and then selectively precipitate the metals.
|
|
|
neutrino
International Hazard
Posts: 1583
Registered: 20-8-2004
Location: USA
Member Is Offline
Mood: oscillating
|
|
A prospector's trick I have heard is to stick a copper wire in a solution of mercury ions to recover the mercury. The Hg simply collects at the bottom
in a puddle, no annoying solid Cu-Hg alloys are formed. If you're redistilling the Hg anyway, a small copper contamination will be no problem.
I don't know of any tricks for separating the Ag first, though. This site may be of some help.
|
|
|
12AX7
Post Harlot
Posts: 4803
Registered: 8-3-2005
Location: oscillating
Member Is Offline
Mood: informative
|
|
I vote for Hg > Cu cementation.
Alternately, you could do the reduce/dilute trick and precipitate Cu2Cl2, but, Hg2Cl2 would do the same thing.
To break a metal film on glass, I bet you can add a flux like rosin, or even oil may do it. Rosin (paste flux from RadioShank) has a low melting
point. My container is solid in this room, in Wisconsin winter, but has a consistency like honey at body temperature. I'd go for boiling water temp
just to be sure.
Tim
|
|
|
garage chemist
chemical wizard
Posts: 1803
Registered: 16-8-2004
Location: Germany
Member Is Offline
Mood: No Mood
|
|
I can't cast anodes out of the amalgam (small irregular lumps), because it is hard and heating to soften it would liberate Hg fumes.
I read up about aqueous chemistry of mercury and developed a wet process which would theoretically work.
Hg2+ compounds often behave abnormal in solutions because of very low dissociation.
Example: HgCl2.
When HgO is treated with a solution of NaCl, the HgO dissolves and the liquid becomes alkaline. NaOH and HgCl2 are formed in the solution, and no HgO
precipitates because HgCl2 is very sparingly ionized. Only a large excess of NaOH would make the reaction go backwards.
Therefore, after precipitation of silver as AgCl from the solution in HNO3 I add an excess of NaCl solution and then slowly add NaOH solution while
hot, which will selectively precipitate CuO and leave HgCl2 (further stabilized as the complex Na2HgCl4, and as the basic chloride) in solution.
The CuO is filtered and washed with NaCl solution, which redissolves any HgO that might have precipitated.
Now I only have HgCl2 in solution, which I can precipitate as the metal by adding coarse iron filings.
Or I'll just stick a copper wire into the solution (after precipitation of silver of course) and redistill the mercury/copper amalgam that forms.
That's probably the easiest way.
My only worry is that (Hg2)2+ might be formed during dissolution in HNO3 and precipitate along with the silver.
I can test for this by the calomel reaction (ammonia added to Hg2Cl2 gives a black color, while pure AgCl only dissolves) but I don't have enough
persulfate to oxidise this. Would H2O2 also work?
[Edited on 13-2-2006 by garage chemist]
|
|
|
garage chemist
chemical wizard
Posts: 1803
Registered: 16-8-2004
Location: Germany
Member Is Offline
Mood: No Mood
|
|
I just did a preliminary test with pure mercury in order to evaluate its behavior and find out the conditions needed in order to produce divalent
mercury in solution.
A drop of mercury was put into a test tube, followed by ca. 2ml water. Then it was heated and 65% HNO3 added drop by drop until a reaction started to
become evident (gas evolution). About 1-2ml were needed.
It was then boiled until the mercury had dissolved, which took about 15 minutes.
It was left to cool down, and NaCl solution was added dropwise.
A white precipitate which looked like AgCl immediately appeared, which could only be Hg2Cl2.
Obviously the Hg had not been oxidised to Hg2+, but only to (Hg2)2+ . 
I added more NaCl solution in order to precipitate the whole of the present (Hg2)2+.
The solution was then heated to boiling, which caused the Hg2Cl2 to agglomerate into a white mass, with clear solution above it (AgCl does that too,
I've done this often with AgCl in order to be able to wash it).
An oxidiser was needed to oxidise the mercury. I didn't want to use H2O2, since heavy metals catalyse its decomposition, possibly creating a hazardous
situation.
My choice was sodium chlorate.
A small amount of NaClO3 was added, dissolved and the solution boiled. The white precipitate dissolved in a matter of seconds. Perfect! I now had a
solution of Hg2+.
It was somewhat yellowish, due to present chlorine dioxide from the chlorate and acid. A bit of sodium disulfite removed the color and made the
solution clear, without reducing the divalent mercury (no precipitate occured).
Now it was time to test a method of producing elemental mercury from this solution.
A piece of 1,5mm copper wire was added to the solution.
After a few seconds, some spots on it became grey, and those
grey spots grew until the entire wire was grey.
A slow gas evolution was observed.
However, there are no signs of accumulating mercury- the wire is as grey as before, and nothing is dropping down.
If this "trick" actually works, it does so very slowly.
I'm leaving the experiment over night in order to see if something happens.
Maybe the solution must be neutralised before the mercury can be reduced with copper? I don't know.
I'm going to try iron as a reducing agent next, if the copper works too slowly. This will definately require neutralisation, as iron rapidly
dissolves in acids, in contrast to copper.
EDIT: @ neutrino: That site is very interesting, especially the production of sodium amalgam- although the guy thinks that the white stuff he gets by
adding salt to his mercury solutions is silver chloride. Maybe I should tell him?
[Edited on 13-2-2006 by garage chemist]
|
|
|
garage chemist
chemical wizard
Posts: 1803
Registered: 16-8-2004
Location: Germany
Member Is Offline
Mood: No Mood
|
|
I added Na2CO3 to the solution until it stopped fizzing, to neutralise the acidity. I have the feeling that the excess nitric just dissolves the
mercury as it starts forming.
Let's see what this does now. The solution is already green/blue.
BTW:
I read up on solubility of metals in mercury. Copper is soluble in mercury to the extent of 0,002% at 20°C.
This means that a saturated solution of copper in mercury is mercury of 99,998% purity.
That means of course that no copper particles are floating around in the mercury, meaning that it must be filtered through a very fine filter.
|
|
|
Eclectic
National Hazard
Posts: 899
Registered: 14-11-2004
Member Is Offline
Mood: Obsessive
|
|
With an excess of HNO3, you should get Hg+2.
|
|
|
Eclectic
National Hazard
Posts: 899
Registered: 14-11-2004
Member Is Offline
Mood: Obsessive
|
|
With an excess of HNO3, you should get Hg+2 (and lots of NO2).
|
|
|
garage chemist
chemical wizard
Posts: 1803
Registered: 16-8-2004
Location: Germany
Member Is Offline
Mood: No Mood
|
|
Yes, but that's a waste of nitric since a large excess is needed. Hence my choice of NaClO3 as a more efficient oxidant which works in dilute
solution.
I've looked for my precipitation of mercury with copper wire.
It didn't look good at all! Apparently I used too much Na2CO3 yesterday, since some flocculent greenish stuff has precipitated (likely some copper
carbonate), along with black crud.
I added some HCl, which slowly dissolved the fluccolent material.
Then I heated the solution, to drive out the CO2 that slowly evolved.
Now I saw that the copper wire was completely covered with a shimmering smooth silvery amalgam layer. It was absolutely beautiful!
I'll try to make a photo later, when I take out the wire. I hope that the amalgam layer doesn't stop the cementation...
At the bottom there was still some black stuff, which slowly dissolved after prolonged boiling.
At the bottom of the copper wire, a mercury drop is visible, although smaller than the one I used. Some mercury clings to the surface of the copper,
as the amalgam layer shows. This will have to be removed by distillation.
[Edited on 14-2-2006 by garage chemist]
|
|
|
garage chemist
chemical wizard
Posts: 1803
Registered: 16-8-2004
Location: Germany
Member Is Offline
Mood: No Mood
|
|
I've made some pics of the wire.
You can find them here.
|
|
|
vulture
Forum Gatekeeper
Posts: 3330
Registered: 25-5-2002
Location: France
Member Is Offline
Mood: No Mood
|
|
There is an analytical process with very high sensitivity to determine mercury, which uses a mixture of nitric and sulfuric acids to oxidize all
mercury to Hg2+, after which it is reduced to the metal again with SnCl2. The rest of the process isn't relevant, but the sample preparation is
obviously quantitative, given the very good results obtained with this method.
One shouldn't accept or resort to the mutilation of science to appease the mentally impaired.
|
|
|
garage chemist
chemical wizard
Posts: 1803
Registered: 16-8-2004
Location: Germany
Member Is Offline
Mood: No Mood
|
|
Thanks for that info, if the other processes don't work well I could still use SnCl2. But I'd first have to get some tin...
And wouldn't SnO2 be formed during this reduction and stop the finely divided mercury that forms from flowing together into a single blob?
If someone knows any other reducing agents (except metals, of course) that reduce Hg2+ to the metal while leaving Cu2+ untouched, please tell me.
I know that NaBH4 reduces Hg2+ to the metal, too, but copper would also be reduced.
|
|
|
The_Davster
A pnictogen
Posts: 2861
Registered: 18-11-2003
Member Is Offline
Mood: .
|
|
SnCl4.xH2O would be formed(forgot the value of x), I don't think it would hydrolyse to SnO2 that easy.
Just based redox potentials on the table in front of me, sulfurous acid could work. Fe2+ could work as well, as well as peroxide.
According to the potentials, SnCl2 would also reduce Cu2+ to copper metal, however the potentials are rather close so it must be negligible.
|
|
|
garage chemist
chemical wizard
Posts: 1803
Registered: 16-8-2004
Location: Germany
Member Is Offline
Mood: No Mood
|
|
Yes, I just thought of Fe2+ too. If it works it would be perfect.
Copper wouldn't fall out of solution, only the Hg, and directly in the form of metal (silver would have to be removed first though).
Let's hope that it doesn't get reduced to only Hg2Cl2 which falls out of solution... maybe if enough Fe2+ is used the Hg2Cl2 gets reduced to the metal
too.
This will be my next experiment, for sure.
Sulfite would reduce copper to CuCl before doing anything to the mercury, therefore not suitable here.
Peroxide is interesting, though heavy metals catalyse its decomposition. Experiment will have show whether this is a problem.
|
|
|
unionised
International Hazard
Posts: 5128
Registered: 1-11-2003
Location: UK
Member Is Offline
Mood: No Mood
|
|
It would be nice if you could get some of the SnCl2 by leaching the amalgam with HCl first- this would save on HNO3 too.
I once had a similar problem, I wanted some pure Hg but didn't want to distill it.
I disolved the stuff with I2 and KI then added NaOH.
(Many other metals won't disolve under these conditions).
Then I recovered the Hg with NaBH4 (because I had some). I think SnCl2 or formalin would also redce the [HgI4]2- complex.
|
|
|
garage chemist
chemical wizard
Posts: 1803
Registered: 16-8-2004
Location: Germany
Member Is Offline
Mood: No Mood
|
|
I tried to distill the Hg out of the amalgams today.
As I had heated enough for the amalgams to become soft, mercury droplets appear on their surface and slowly inflate like dough because of the mercury
vapor, my test tube shattered. The bottom fell out. 
Of course I did this in my fume hood, but I still immediately opened all windows and doors and left my lab to let it air out.
I think it broke because of thermal shock. The inflating amalgams most likely touched very hot sections of the glass and caused it to shatter.
It seems like one must not distill mercury or amalgams in a glass aparatus.
My father suggested to do it like the goldminers in third world countries: put the amalgam into a steel can, cover it with a wet cotton cloth and heat
from below. The mercury will condense as droplets in the wet cloth and can be recovered by squeezing it out.
Sounds simple. What do you think?
|
|
|
Eclectic
National Hazard
Posts: 899
Registered: 14-11-2004
Member Is Offline
Mood: Obsessive
|
|
I think you should make a retort from black iron pipe fittings and pipe.
|
|
|
garage chemist
chemical wizard
Posts: 1803
Registered: 16-8-2004
Location: Germany
Member Is Offline
Mood: No Mood
|
|
For ca. 8ml of mercury in total it will have to be small, otherwise losses will occur. But this would be the preferred method, no doubt.
However, some of the Hg will be held back, meaning that I can't remove all of the Hg by distillation. Some of it will have to be separated when the
metal residue is dissolved in HNO3.
Or I'll just throw the thing into HNO3 and apply chemical means of separation. Those dangerous experiments with distillation of mercury can't be good
for my health.
|
|
|
garage chemist
chemical wizard
Posts: 1803
Registered: 16-8-2004
Location: Germany
Member Is Offline
Mood: No Mood
|
|
I did it: I dissolved the 20g of amalgam in HNO3 (40ml, 53%).
I now have a blue solution over a white precipitate (SnO2).
I tried to filter it, and guess what: the SnO2 completely clogged the filter plate (glass filter, porosity 3). With strong vacuum, about 1 drop every
minute came through.
I had to pour the solution back into the flask and scrape the SnO2 sludge from the filter to wash it out.
As I had to wear gloves and be really careful since a few drops of this solution would be lethal, this was not the most pleasant of operations.
I now use a filter paper and gravity filtration, this works somewhat better. However the solution that comes through is still turbid. I'll have to
purify the AgCl that I'll get from this solution by redissolving in ammonia and filtering or better decanting again.
Does anyone know how I can clean the filter plate from the SnO2? It must be a chemical method to dissolve the SnO2.
|
|
|
Eclectic
National Hazard
Posts: 899
Registered: 14-11-2004
Member Is Offline
Mood: Obsessive
|
|
Doesn't hydrochloric acid dissolve SnO2? For difficult filtrations you can lay down a bed of celite (www.worldminerals.com/CeliteIndex.asp) by making a slurry in water and pouring onto filter before filtering your product. Check pool maintenance
supply stores.
|
|
|
Lambda
National Hazard
Posts: 566
Registered: 15-4-2005
Location: Netherlands
Member Is Offline
Mood: Euforic Online
|
|
Stannic Oxide is soluble in only two common Electrolytes; Hydrofluoric Acid and Strong Caustic, and both attack and dissolve Glassware.
SnO2 + 2NaOH ----> Na2SnO3 + H2O
Unfortunately, Mercury and Silver will now become insoluble. They do however dissolve in HNO3, after which the above mentioned process may be
repeated.
After the Patient has died, the Specialists give the best advice:
1 - You can also try your Mercury and Silver recovery procedure with a Chelating agent in order to keep the Tin in solution (Procedure used in
Electrochemical Stripping etc.).
2 - Decant the Mercury and Silver nitrate off the Sludge.
3 - Dissolve the Sludge in hot NaOH solution, and filter off or decant the Na2SnO3 from the insoluble Mercury and Silver Salt.
I emphasize and weep with you Garage Chemist. 
[Edited on 18-2-2006 by Lambda]
|
|
|
| Pages:
1
2 |
![[*]](images/xpblue/default_icon.gif){kind=icon}
