snoman701 said:
g_axelsson said:
snoman701 said:
g_axelsson said:
Only quick field test I know of for small amounts of gold in alloys is using an XRF-gun. An expensive tool unless you handle lots of scrap metal.
Göran
I wouldn't trust a gun on gold filled scrap in a well prepared sample. A gun on a few ppm on dirty field sample wouldn't even begin to be reliable.
Fire assay only.
We are not discussing gold filled, we are discussing old copper from pre-electrolysis times.
XRF is used by mining companies in the field, both when prospecting and while mining to track gold ore.
Göran
My point about gold filled is that it's low grade material.
I have a pretty good understanding of the limitations of XRF technology, and it's use in the field. In the field it's used with correlative data and known associations, not the actual measurement of gold. You track elements that are known to be associated with gold, and compare this to a series of fire assays you have performed.
Have you used an XRF gun much? You can see considerable resolution problems when discussing 10^-2 resolution if you aren't skilled. In order to accurately measure in the 10^-3 range, you need to have stationary sample and long sample times. Even then, the numbers on the screen are only suggestive and you are looking for the pattern of minerals. With Wavelength dispersive technology, you can get 10^-4 with long sample times and a really good analyst comparing results to gravimetric data. But even when discussing 3 oz / ton, very good blister, you are talking about 10^-5. You would now be trying to make use of a number that is smaller than your uncertainty.
You are correct, I have no personal experience of using XRF, the equipment I have is part of an electron microscope and not brought back to a running state yet. I got the detector (a SiLi diffused detector that needs liquid nitrogen to run) and the electronics associated to it, but after the last move the SEM ended up in a corner of my storage and hasn't been set up yet.
My experience of XRF in the field is only discussions with the geologists on a nearby gold mine and I never asked them if they were looking for tellurium or gold when they analyzed the samples or the walls in the mine. At the last visit I found an interesting specimen, they didn't have the XRF at the mine at that time, but they kept a sample and I got the result from the test after a week. The minerals we found showed high number of lead and tellurium with traces of gold. How high numbers I don't know.
Going online and doing a quick googling there are for example claims of detection limits of 20 ppm Au and 10 ppm Ag ( https://www.911metallurgist.com/equipment/handheld-xrf-analyzer/ ), that would put the gold in GSP's example at undetectable levels while the silver would be detectable. There are other contaminants like zinc, lead, iron... that would also show up. Blister copper contains between 98.5-99.5% copper so there are 0.5-1.5% contamination that can be detected with an XRF to show if you are dealing with blister, a copper alloy or electrolytic copper.
I found this article about refining blister copper, http://eprints.nmlindia.org/3800/1/54-58.PDF
In it they give numbers of gold and silver up to 100 (0.01%) respectively 1000 (0.1%) grams per ton.
These numbers would be detectable if he above XRF data is true.
Every method have it's limitations but I would still prefer XRF for sorting copper in the field over fire assay.
stella polaris, there are no streak test for copper that I know of and I suspect that just as for gold when you get close to 95% or more the streak test doesn't work any longer.
I was trying to come up with a simple sorting method but as there are copper alloys with similar amounts of other metal or silicon I can't see any easy test.
The most I could come up with would be to dissolve the copper in nitric acid, filter off the solids and then add a drop of aqua regia (just add a drop of HCl) and test with stannous.
Another test would be to check for presence of silver. Clean a piece down to pure copper surface to avoid false positives from silver plated scrap. Dissolve a piece in nitric acid, filter off any solids so you have a totally clear solution. Then add salt or HCl and look for the white precipitate (check for turbidity with a light source).
First test would be to use a file, pure copper is soft. But you would need a number of samples to test that method to see if it is a working method.
Göran