Original paper
Chemical variation in eudialyte
Johnsen, O.; Gault, R. A.

Neues Jahrbuch für Mineralogie - Abhandlungen Band 171 Heft 3 (1997), p. 215 - 237
41 references
published: May 15, 1997
DOI: 10.1127/njma/171/1997/215
ArtNo. ESP154017103000, Price: 29.00 €
Abstract
A large suite of eudialyte samples has been investigated by electron microprobe analysis, out of which 36 are presented. All analyses are transformed into formula units (pfu) based on 78 anions (Z = 3), what is proposed to be the best procedure at present. Regression analyses of the elemental variation are performed and correlations between elements and groups of elements are discussed in the light of present structural knowledge from single crystal data. Selected eudialytes have been further studied by high resolution X-ray diffractometry and examples of the variation of X-ray data with chemistry are presented. In addition, a few eudialyte samples have been investigated by TEM in order to look for crystal defects causing non-stoichiometry of which none were found. An overview of the eudialyte structure is given using the site notation from Johnsen et al. (in press) comprising [6]M1, [4-5]M2, [6]M3, [4]M4 and Nal through Na5 in 6-9 coordination in addition to the framework, composed of three-membered and nine-membered rings of SiO4 tetrahedra linked together by Zr octahedra. A special property of eudialyte is the existence of oblong cages along the triad axes carrying additional bonded and possibly also non-bonded components. Our data show that the framework is stable and uniform in composition, while considerable variation takes place in other parts of the structure; Na, though, is always present in large amounts (13.95-17.44 pfu). The Na ⟷ Ca substitution often proposed for eudialyte is not supported by our data. Ca, generally assigned to the M1 site, varies from 1.92 to 7.82 pfu with an average of 5.45 pfu; Mn and REE are the main substitutional elements. One eudialyte from Mt. St.-Hilaire has 9.78 wt % REE2O3 (Y2O3 not included), the highest content recorded so far. Fe, assigned to M2 in four-fold coordination in an almost planar arrangement, varies from 0.68 to 3.33 pfu, average only 1.73 pfu, which is just slightly more than Mn (av. 1.39 pfu). Mn also substitutes for Fe, but is five-fold coordinated because of the incorporation of an extra oxygen (O19) close to the triad axes. The average content of Si is 25.89 pfu, higher than the 24 pfu necessary for the formation of the framework. This confirms the presence of one or two Si positions on the triad axes (M4) thus transforming the nine-membered rings into [Si10O28]16-platforms. The amount of Si is negatively correlated to that of Nb, not because these elements substitute in a classical sense, but because they are competitors in certain micro regions in the cages. Nb is accomodated in M3 sharing O19 with M2. Combining significant correlations leads to a eudialyte series between two end-members, one relatively rich in Si, Ca, Fe and Cl and another relatively rich in Nb, REE, Mn and F. Plots of X-ray data vs. chemistry show that the ΣNb, REE, Mn, F component is negatively correlated to the α parameter, positively to Dx calc., and the intensity of some reflections, especially reflection (003) proves to be a good indicator of the ΣNb, REE, Mn, F/ΣSi, Ca, Fe, Cl-ratio. Finally, a procedure for calculating the formula of an analyzed eudialyte is proposed.
Keywords
Eudialyte • kentbrooksite • microprobe analysis • X-ray diffraction • crystal structure