JCK Web Exclusive: The Andesine Report

By Gary Roskin, G.G. FGA, Senior Editor

Posted on November 12, 2008

After years of difficult scientific experimentation by Dr. George Rossman, and more recently Dr. John Emmett, it has been proven that the diffusion of copper into andesine is possible. It is still uncertain, however, if we can positively determine whether or not a particular faceted andesine has been diffusion treated.


Slices of feldspar were diffusion treated by Dr. John Emmett, proving that copper diffusion is possible.

Here’s what we know
Rossman and his team at Caltech (California Institute of Technology, Pasadena, Calif.) have determined that the rough feldspar (andesine or labradorite) from each of the three known localities, Oregon, Mexico, and China, has an identifiable chemical signature. Because of that discovery, they can prove that the gems studied at Caltech, provided by Jewelry Television, Direct Selling Network, and Ande Gem (a Los Angeles–based AGTA member gem supplier), are from China. They are definitely not Mexican or Oregonian.


Slices of natural-yellow feldspar from China, Mexico, and two sources in Oregon—with diffusion-treated red andesine, faceted and rough—courtesy of Dr. John Emmett.

Rossman and his team determined a method to identify whether or not these gems were heated, not from the volcano in which they were created, but by a laboratory. He has proof that the red and green gems we see in the trade—other than the natural-color yellows from Oregon and Mexico and the natural-color reds and greens from Oregon—have indeed been heat treated.

In Brush Prairie, Wash., Emmett and his team at Crystal Chemistry have been working on the question of whether or not copper can be introduced into feldspar to create red and green andesine. (Emmett was responsible for proving beryllium diffusion treatment of corundum.) They have been successful with copper diffusion, creating red and green andesine/labradorite from yellow Mexican, Chinese, and Oregon feldspars, with no observable differences.

“Feldspars [from all localities] are layered and twinned and have a very complicated physical structure,” notes Emmett. “And the end result, whenever you have crystal interfaces like that, you have a whole series of dislocations around the interface, with the copper just streaming down these dislocations. Diffusion happens very rapidly down these diffusion pipes. You can deeply color to great depths in this material in a limited amount of time. Hundreds of hours. If it weren’t for these pipes, it would be tens of thousands of hours to do that.” And that’s what’s going on with this material.

Knowing this is possible is not enough to prove that the andesine being sold in the trade is diffusion treated. This is where identification becomes problematic.

Rossman notes that Mother Nature also diffuses copper into feldspar to make red and green Oregon sunstones. This brings to mind the problem of identifying the color origin of green diamonds. Both Mother Nature and laboratories use irradiation to cause the green color. “Right,” says Rossman. “So even if the andesine is [copper] diffused, the process seems to be doing something that in many ways appears to be the same as what has occurred naturally. It’s not like you’re putting cobalt into topaz to make it turn blue, which is not a natural process.”

That raises a question: Is it possible, and useful, to measure the amount of copper, the way the amount of beryllium in treated sapphires is measured?

So far, Rossman’s tests show that the copper content of both natural and diffusion-treated materials looks basically the same. “The amount of copper in both Oregon and China feldspar is quite variable,” notes Rossman. “The problem is that the range of copper contents we measured has significant overlap between the two localities.”

This may turn out to be a very difficult, if not impossible, identification.

The Gem Labs
Even with what we have learned so far, we still can’t identify a given stone as natural or laboratory diffused. Just to get to this point in the identification process, Rossman performed spectroscopy, chemistry, and radioactive isotope experiments, and he’s conducting diffusion experiments. Emmett and his partner Troy Douthit have been primarily focused on melting points, diffusion coefficients, and, in particular, diffusion experiments. Emmett also has been working in cooperation with Ken Scarratt’s GIA Gem Lab in Bangkok, Thailand, examining spectroscopy and chemistry.

“The major gem labs don’t have the facilities, equipment, or trained personnel for these types of experiments,” notes Emmett, “and in this world, that ought to be a major concern for us all. Gemology in the 21st century has to change from being just an observational science to one that also embraces experimental research.”

It will take a lot of people with special talents and equipment, working together, to fill in the whole picture.

Now that Rossman and Emmett have before-and-after samples, GIA’s gem lab is examining the gemological properties of known treated and known natural andesines and other feldspars to see if there’s a way to determine which is which. In fact, the GIA lab in Carlsbad, Calif., has been documenting hundreds of samples of feldspars collected from many sources with the goal of developing identification criteria.

The simple gemological tests like immersion, refractive index, and microscopic examination may not hold the answers.

For example, Mother Nature also uses pipe diffusion to color Oregon labradorite, so differentiating natural from laboratory-treated andesine/labradorite will be the challenge for the professional gem labs. It’s possible that because the treatment labs require only days (rather than centuries) to add color to feldspar, there may be some sign for gemologists to look for. But remember, even though this is called “diffusion,” it is very different from the diffusion seen in treated corundum. The vast majority of the samples GIA has examined have not had the same kinds of zoning seen in diffusion-treated corundum. Feldspar pipe diffusion is typically much more convoluted and irregular. 

Looking at natural inclusions, the amount of copper platelets found in the natural Oregonian material (not so much in andesines or labradorites from China or Mexico) might give gemologists clues to identification. But the labs will need time to find answers. And the trade will require patience.

Laboratories can report only on what is known at any given time. (To quote the adage, “There are things you don’t know you don’t know.”) There are only a few well-equipped gemological laboratories in the world, and they have limited personnel and resources. They are outnumbered by those seeking new ways to treat gemstones.

Moral of the Story
Even if you’ve never sold any andesine, you could end up in court for incorrect disclosure of any gemstone that you did not fully research. “I didn’t know,” is not an option, says Cecilia Gardner of the Jewelers Vigilance Committee. You cannot be relieved of your responsibility to your customer.

Citrine can be a heated amethyst or an irradiated quartz. Emerald can be filled with oil or resin. Irradiated blue topaz may be over the NRC limits for residual radioactivity. Yellow “topaz” might be citrine. Every gemstone in your case has such issues.

There is no practical way (cost-effective and nondestructive) to identify diffusion treated andesine—yet. Now that the gem community knows andesine can be diffusion treated, gemologists are trying to determine a way to prove that a gemstone has been copper diffused.

For now, absent proof that it came from a reliable source in Oregon, jewelers should assume red and green andesine has most likely been diffusion treated.