An affordable method for accurately determining the concentration of gold in solution.

[4metals]

Without investing in an Atomic Absorption or ICP.

I've got time on my hands now as I am waiting for the end of July for spinal decompression surgery. My spinal stenosis has me spending more time sitting than I am used to and also convinced me it is time to cut the BS and really retire. One of the things I've been toying with while sitting on my arse is AI. (Artificial intelligence, not Aluminum!). Being quite the nerd I started asking questions that most refining nerds have on the tip of their tongue like "what are some methods for the colorimetric determination of gold in solution". The reason this interests me is the fact that a good old fashioned colorimeter, like a Spec 20, is relatively inexpensive ($100-$200 used and working) as compared to an Atomic Absorption or ICP (Thousands to buy with annual maintenance on top of that) and those Spec 20's from my college days are literally bulletproof.

So this is what ChatGPT said to me;

"Yes, there are colorimetric methods commonly used to determine the gold concentration in a solution. Colorimetry is a quantitative analytical technique that relies on the measurement of the intensity of color produced by a chemical reaction. In the case of gold concentration determination, a colorimetric assay typically involves the formation of a colored complex between gold ions and a specific reagent.

One widely used colorimetric method to determine the gold concentration is utilizing the reaction between gold ions and a reducing agent such as stannous chloride (SnCl2) or sodium borohydride (NaBH4). The resulting gold nanoparticles exhibit a characteristic absorption spectrum, typically in the visible range, which can be measured using a spectrophotometer.

Another method involves the formation of a colored complex between gold ions and a ligand, such as thiourea or dimethylglyoxime. The intensity of the color produced is directly proportional to the gold concentration and can be measured using a colorimeter or spectrophotometer.

It's important to note that the choice of method may depend on factors such as the concentration range of gold being measured, the presence of interfering substances, and the required sensitivity and accuracy of the analysis. Additionally, it's always recommended to consult specific literature or consult an expert in analytical chemistry for detailed protocols and guidance when performing colorimetric assays for gold concentration determination."

That last paragraph is their legal disclaimer, they likely have as many lawyers as programmers!

Next I asked about cyanide solution specifically as it is an effective method to strip parts and it would be nice to. have colorimetric methods for that as well. Again ChatGPT came through.

"Colorimetric determination of gold in cyanide solution is a commonly used method in analytical chemistry. The process involves measuring the intensity of color developed in a solution containing gold when a specific reagent is added. The color intensity is directly proportional to the concentration of gold present.

One of the most commonly used colorimetric reagents for gold determination is potassium iodide (KI). When KI is added to a solution containing gold, it forms a brownish-colored complex known as the "iodide-gold complex."

The concentration of gold can be determined by measuring the absorbance of this complex using a spectrophotometer. The relationship between absorbance and concentration is established by constructing a calibration curve with known standards of gold concentrations.

To perform the colorimetric determination, a series of standard solutions with known concentrations of gold are prepared. These solutions are treated with the KI reagent, and their absorbance is measured at a specific wavelength using a spectrophotometer. The absorbance values are then plotted against the corresponding gold concentrations to create the calibration curve.

Next, the cyanide solution, which contains an unknown concentration of gold, is treated with the same reagent, and its absorbance is measured. By comparing the absorbance of the unknown solution to the calibration curve, the concentration of gold in the cyanide solution can be determined.

It's worth mentioning that precision and accuracy are crucial in this analysis, as slight variations in reagents, technique, or instrument calibration can affect the results. Therefore, it's essential to follow proper laboratory procedures and make necessary adjustments to ensure reliability in the determination of gold concentration."

Again the last paragraph is the lawyer speak.

So my question to our members is have any of you had experience using some prepared gold in stannous standards to analyze gold in acids or some prepared gold in iodine standards to analyze cyanide solutions?

I have always been lucky enough to have Atomic Absorptions or ICP's for this type of work so I do not have any first hand experience but I do have a Spec 20 stashed in the basement if this discussion gets interesting.

So....... opinions. Is ChatGPT blowing smoke or are these methods with potential for accurate analysis without breaking the bank?

 

[Lou]

Yeesh, sorry to hear about your spinal issues and I hope the surgery goes well. ChatGPT is probably mining the living crap out of this forum and repacking it for mass consumption. I find that interesting. Some of the questions you ask it are straight up lifted right off of what we've (as in you and I) have posted.

I don't see any reason whatsoever why you couldn't correlate an intensity chart of Au-SnCl2 nanoparticles to gold concentration.

Before I had a pot to pee in when I got into this as half hobby, half mere interest and had learned about the stannous test "Purple of Cassius" I had made some test solutions. I'm pretty sure I posted about this somewhere, but I was mostly using it to detect low levels not higher concentrations, because I didn't know when a gold drop was done or I wanted to see how a leach was going but I don't see why it couldn't be put into some sort of linear range. Probably works best at high dilution.

I think I made a 5 ppm, 25 ppm, 50 ppm, 100 ppm, 250 ppm and 500 ppm. I would use a q tip or a piece of filter paper against a white piece of copy paper to gauge the color relative to my known standards. I was using my (then considerably better) eyes but I think you can basically tell down to 1 ppm or less if there's gold in the solution with minimal base metals. Obviously, certain other things interfere and make different colored complexes or NPs.

 

[4metals]

It would be interesting to see if you could peak the wavelength and get minimal interference from base metals. Similar to what is done on an Atomic Absorption with minimal background correction. Then you could make a standard curve and pick your unknown samples value off of a graph. Stronger standards should last for some time as purple of cassius is colloidal and shouldn't settle.

And for cyanide leach processing the gold-iodide is soluble so the solutions should stay stable too. In the dark, for all of the standards.

I have clients without instrumentation who get in a pound sample of gold plated pins or headers and need to "assay" them to tell the customer. Stripping a weighed amount in a liter of cyanide solution and reading the sample on a colorimeter would be easier than dissolving a pound in aqua regia and looking for a few grams of gold after processing. (sad to say that is how a lot of them do it!)

 

[Lou]

You'll hate this, but I've always done quick and dirty analyses on pins by fuming in sulfuric and weighing the gold on a Gooch crucible. Not something to do without a fume hood and analytical balance!

 

[4metals]

I have done the same for a quick analysis. Even faster if you roll out the pin into a ribbon. But that is a result for 1 pin with the assumption that all are the same. I prefer to process a hand-full to get a better average, but that is too many for fuming in sulfuric.

 

[orvi]

I always had a luxury of having XRF on hand, so semi-assay of pins were just melting them down, account the weight loss, flatten the button, sand it and shot with XRF to determine gold content.

But I can imagine cutting sample of pins with bench shear to smaller pieces, and quickly strip base metal in strong nitric acid. This should be fairly quick and straightforward, leaving gold foils behind. When majority of interfering base metals is leached out, you can dissolve the gold with HCl/peroxide to obtain solution with known volume, which will be conveniently measured via colorimeter. I would avoid using nitric for the dissolution, as nitrogen species tend to be coloured very similarly to gold, and there is a possibility of skewing the analysis by interference.

Altough, this will work only for brass/bronze pins. Ferrous pins like kovar, these will be much more tough to deal with. Only dirty thing that comes to my mind right now is to dip them in molten lead or bismuth, repeat the dip with extra bismuth to strip all of the gold out of them and then cupell it to obtain metallic bead.