The authoritative public forum
  for Metal Finishing since 1989
  

  -----

Quantitative Analysis of Au in spent plating baths

I've been assisting a jeweller to recover Au from his spent plating baths (alkaline CN) and would like to begin working with his floor sweepings. I do not have an Au lamp for AA spectroscopy ⇦ this on eBay or Amazon [affil links] to determine how much Au is present in the samples to calculate the efficiency of the recovery process. I would like to know if there is a colorometric method (SnCl2 ?) that is published and free from interferences from Ag, Cu, Ni and other trace contaminants.

Thanking you in advance for your comments.

Richard Michael Grazul
Ph. D. Chemistry - Juiz de Fora, MG Brasil
2005

---

2005

You refer to a method where the gold is reduced to colloidal gold (they used to call this "Purple of Cassius") by stannous ion. I have tried this method, and have had no luck with it at all. That doesn't mean that you might not, with perseverance. My understanding is, however, that this and related methods are interference ridden horrors.

I used to do Au electrogravimetrically. To a sample that contained 25 - 400 mg of gold, I'd pH adjust to somewhere in the 9 - 10 range, add 500 mg sodium cyanide, adjust the volume to 75 ml, and plate the gold out onto a platinum clad mesh for 1 hour. Weighing it before and after got me to the answer.

Dave Wichern
Consultant - The Bronx, New York

---

2005

Dear Mr. Wichern,

Could you please cite the applied voltage (potential) for selective electrodeposition of Au in the presence of Ag and other base metals? Furthermore, have you validated this method to ensure that it is quantitative as opposed to "efficient"?

Thank you for your prompt reply to my inquiry.

Richard Michael Grazul
Ph. D. Chemistry - Juiz de Fora, MG Brasil

---

I used a current of 0.50 A. The mesh used as the cathode was 7 x 20 cm. Since it was an expanded mesh, the actual surface area was somewhat greater than 140 cm2. How much greater, I am not sure.

I did not face a selectivity problem. My samples were gold plating baths, and the only precious metal present was Au. If I did face this problem, I'd try to use chemical means to separate out the other metals.

No, I did not do a rigorous validation of this test method. My estimates of the limits on the amount of Au to be used in the sample came from tests I did where I dried and weighed the cathode, then put it back in the cell and continued plating onto it to see if there was any further weight gain. Also, samples of the bath that were sent to an outside lab yielded results that were very close to our in house ones. Lastly, the advice I gave based on the results from this method worked.

Dave Wichern
Consultant - The Bronx, New York
2005

---

2005

Mr. Wichern,

Thank you for the clarifications. My current line of thinking for a quick field test Au plating baths is as follows:

In a small beaker [beakers on eBay or Amazon [affil links] add small aliquot (0.5 mL) bath test soln.
verify pH >9 w/ pH paper
Add small amt aqua regia heat w/alcohol lamp to volatilise HCN and oxidize Au (I) to HAuCl4 (Cassius Purple apparently doesn't work on [Au(CN)2]-)
add additional conc. HCl to chase off XS HNO3 w/ heat.
(same as we do for H2PtCl6) and eveporate to near dryness
dilute w/ a little water
spot test for Au(III) on filter paper impregnated with SnCl2/HCl

I hope that procedure and the intensity (or absence) of color will help identify the alkaline CN baths, as well as any acid Au baths or silver

I think this could also be adapted to a quantitive method by preparing a set of standard HAuCl4 solutions and make a calibration curve at the absorbance maximum in the UV or Vis region. Pretreatment of unknown sample w/ aq. regia as described above followed by dilution and reading absorbance at lambda max for HAuCl4 should work on any Au solution. I think Ag interference can be removed by filtering AgCl prior to completing to volume. Hopefully Cu, Ni etc. won't absorb at the same wavelegth or interfere.

Will communicate results (or lack thereof) when available
best regards

Richard Michael Grazul
- Juiz de Fora, MG Brasil