At EDAX the company I work for we build the top selling/designed XRF analyzers in the world along with our main competitor Oxford Instruments. Since I joined Ametek/EDAX I have corresponded with many collectors (~50 collectors/researchers) including Mike Diamond from Coin World on some U.S. Off-metal error XRF analyses. The recent article being that Brassy Lincoln Cent article which appeared in Collectors Clearinghouse in Coin World recently. To answer your question directly - The surface enrichment of silver copper alloys has been recognized for many years. In a typical regal (i.e., legitimate) example if made with silver say at ~ 90.0% as with a Morgan dollar the surface layer can achieve levels in the expected 90-94% range and within the core of the coin it's probably as low as 74-76%. With time, surface corrosion and the cooling down annealing process during the blank making process, silver tends to the surface of the blank and copper forms copper rich areas throughout the interior of the coin. A silver-copper alloy coin is NEVER homogenous in all its areas. The current research paper on this phenomenon is explain in L. Beck "Silver Surface Enrichment of Silver-Copper Alloys; A limitation for the analysis of ancient silver coins by surface techniques (June, 2004). So for Morgan dollars the silver levels will fluctuate proably in the 90-94% range (possibly higher?) if the base silver during production was 90% as is currently reported by the U.S. Mint. See my company at http://www.edax.com. The unit we use is the Orbis a micro-EDXRF unit which is much more powerful/sensitive than the simple hand gun types - but this person is correct - the silver values will fluctuate from Morgan to Morgan and Peace dollar to Peace dollar. Understand also we "never" expect the surface Ag readings to be below 90% - NEVER. If we see this and the weight tends to drop outside the expected weight range this suggest a counterfeit. Of course there a good and bad counterfeits ... but that's another story <BG>.
Edited by colonialjohn
01/31/2013 11:22 am
01/31/2013 11:22 am



















