New Mexico Mineral Symposium — Abstracts

The identification of new mineral species requires careful, detailed, and insightful scientific detective work. The two crucial defining characteristics of a mineral species are its chemistry and its crystal structure. In the absence of a crystal structure determination, the powder X-ray diffraction (PXRD) pattern, which depends upon the structure, is a strong indication of whether the mineral has the same structure as another. The powder pattern, coupled with at least semi-quantitative chemical analysis, generally allows one to determine whether an unknown phase is, in fact, a new mineral.

Unfortunately, the identification of minerals relies upon existing chemical and crystallographic data that, in some cases, are not as accurate or unambiguous as they should be. That is when the mineral researcher must become a sleuth, searching for clues in the literature and often examining the primary evidence—type specimens. Three recent mineralogical projects serve to illustrate mineralogical detective work in action.

More than 5 yrs ago, Bob Jenkins submitted a green tabular mineral from the Lomas Bayas mine, Antofagasta, Chile, for identification. Chemical and crystal structure analyses determined it to be a new Na-Cu molybdophosphate, essentially isostructural with betpakdalite. This led us to conduct a detailed study of other molybdophosphate and molybdoarsen ate minerals in an effort to elucidate their interrelationships. The other existing species examined were natrobetpakdalite, melkovite, mendozavilite, paramendozavilite, and obradovicite. For each species, type material was studied. Unfortunately, these minerals are typically fine-grained and, when well-formed crystals do occur, they are often complexly twinned. Nevertheless, we were able to obtain several structure refinements for species with the betpakdalite structure-type, to solve the structure of obradovicite, and to refine additional structures of new obradovicite analogues. Comparative PXRD proved effective in demonstrating structural relationships for those species for which single crystal studies could not be conducted. Electron microprobe analyses (EMPA) were performed for all species studied, including those for which analyses had been previously published. The final results of this study, which are yet to be published, include the establishment of a new mineral family, referred to as the heteropolymolybdates, the establishment of the betpakdalite, mendozavilite and obradovicite groups, the creation of a suffix-based naming scheme for the species in these groups, and the descriptions of a total (so far) of six new species in these groups. In addition, examination of the type specimen of paramendozavilite has revealed another new heteropolymolybdate mineral (but no paramendozavilite).

In recent years, Joe Marty's collecting efforts in the U-V mines of the Uravan mineral belt along the Colorado-Utah border, have so far resulted in the discovery of at least nine new vanadium minerals (some still under study), seven of which contain the decavanadate anion, (V10028)6- a distinctive cluster of 10 VO₆ octahedra. The decavanadate minerals are remarkable, not only because they contain the decavanadate anion, but because they readily form at room temperature from acidic mine waters and contain very high contents of water. During the course of our investigations, Joe Marty submitted well-formed orange crystals of a Na-Mg decavanadate. Their chemistry fit that of the existing species huemulite; however, our single-crystal and powder X-ray diffraction data differed significantly from those published in 1966 for the species. The only existing type specimen of huemulite in the Smithsonian Institution proved instead to be the recently described decavanadate mineral lasalite, rather than huemulite. We finally located the PXRD slides used in the original description in the collection at Yale University. Tests on this material, including recrystallization, and dehydration tests on the newly collected crystals, led us to the conclusion that the new material is huemulite and that the discrepancies between the sets of XRD data were the result of dehydration of the original material (and therefore modification in its crystal structure) prior to the recording of the original published XRD data.

We recently described two new hydrous aluminum phosphates: kobokoboite, Al₃(PO₄)₂(OH)₃•9H₂O, from the Kobokobo pegmatite in the Democratic Republic of Congo and afmite, Al₃(OH)₄(H₂O)₃(PO₄)(PO₃OH)•H₂O, from Fumade, France. Both of these minerals occur in clusters of thin flaky white crystals. During our study of afmite, based upon specimens provided by Georges Favreau, we encountered tiny colorless crystals of another phase that we thought might be new. It provided a PXRD pattern and single-crystal cell similar to those of matulaite, CaAl₁₈(PO₄)₁₂(OH)₂₀•28H₂O; however, EMPA and crystal structure analysis showed our mineral to have the ideal formula Fe³⁺Al₇(PO₄)₄(PO₃OH)₂(OH)₈(H₂O)₈•8H₂O. To clarify the chemistry of matulaite, we investigated type specimens deposited in the Smithsonian Institution and in The Natural History Museum, London, as well as additional material identified as matulaite from the type locality, the Bachmann iron mine, Hellertown, Pennsylvania. Interestingly, neither matulaite type specimen was found to contain any matulaite. Instead, they contained afmite, kobokoboite, and crandallite. Some non-type specimens examined from the Bachmann mine contained radial sprays of distinct blades, which fit the crystallographic characteristics of matulaite (and of the unknown from Fumade). These provided an excellent crystal structure refinement and EMPA consistent with the ideal formula Fe³⁺Al₇(PO₄)₄(PO₃OH)₂(OH)₈(H₂O)₈•8H₂O. Our conclusion is that the original published chemical composition of matulaite, which was determined using wet chemical techniques, was most likely conducted on a mixture of kobokoboite, afmite, and crandallite using material removed from the specimens that were deposited as the types for matulaite. The PXRD and single-crystal cell reported in the original description were clearly obtained from crystals that came from one of more different specimens that were not preserved as types.

I have had many collaborators in these projects. Besides Georges Favreau, Bob Jenkins, and Joe Marty, mentioned above, Stuart Mills, Fernando Colombo, John Hughes, and Mike Rumsey deserve particular recognition. Other collaborators include Ricardo Baggio, Maurizio Dini, Mickey Gunter, Barbara Nash, William Pinch, George Rossman, and Ian Steele.