A gemologist colleague pulled out a parcel of various colored rough spinel from his pocket and asked if one of the odd ones could be checked with Raman. No wonder he asked since the crystal was completely colorless. Spinel may exist about in any color but this colorless pebble just didn't fit in the picture. After a minute the analysis was ready; While Raman fingerprint zone clearly indicated topaz, photoluminescence part of the spectrum exhibited familiar unambiguous pattern of spinel 'organ pipes'.
Before announcing breaking gem lab news it's good to take a short timeout and start to think clearly what may cause this kind of mixed spectrum. Enter the microscope. A doublet? No. Is spinel among known inclusions in topaz? No. Did I just found a new diffusion treatment done by notorious far east heat treaters? Well, umh... sorry about that. Couldn't resist since we have witnessed just too many hilarious whatever-fusion lab alerts coming out lately. Have I left a small spinel inside the sample chamber? No, but that would explain it... "Hey mate, did you just say it is spinel parcel you have there", YES!
The topaz had rock n' rolled with spinels who knows for how long in the parcel. It was pulled out and only two seconds later found itself lying on the sample stage, waiting for the almighty laser beam. Plain and simple, there was enough spinel dust on the surface of the topaz to get mixed results. Spinel PL reaction is thousands of times stronger than topaz Raman peaks. One wipe with dry cloth was good enough to get rid of spinel PL peaks.
I have introduced Raman to quite many folks and seen how easy it is to adapt this bad habit of tossing the sample on the stage without thinking. The tool is just too easy to use once one get's familiar... point and shoot! Next one, please...bang... aaaaand next one! Of course this is not how the work is done in serious gemological laboratory. There has to be established procedures and process flow including appropriate cleaning steps before the examination. For now on we will change our instructions from "no sample preparation" to "minimal sample preparation".