Mineralogical Association of Canada [MA]

 CC:713B  Tuesday  1630h

Tourmaline: An Ideal Indicator of Its Host Environment I

Presiding:  V van Hinsberg, McGill University; B Martin, McGill University; D Henry, Louisiana State University


Tourmaline Comes of Age

* London, D (dlondon@ou.edu), ConocoPhillips School of Geology & Geophysics, University of Oklahoma, Norman, OK 73069, United States

Minerals of the tourmaline group have made their way from juvenile beginnings as novel accessory minerals to promising petrologic indicators in all rock-forming environments. Advances in the utilization of tourmaline to solve petrologic problems began with analyses of natural tourmaline from diverse rock environments. As a result, the composition of detrital tourmaline grains in sediments has been effectively linked to their provenance. The evolution of tourmaline compositions within and among granites and pegmatites was found to mirror the overall chemical fractionation of these plutonic bodies. Crystallization of tourmaline within silicic plutons begins with a large dravite component at the contacts, but this decreases sharply inward to schorl- olenite-foitite (SOFTur) solid solutions. The Al content of the Y site increases with chemical fractionation, and if tourmaline persists through the entire magmatic event, then elbaite-rossmanite compositions predominate at the end. Idiosyncratic characteristics of a geologic setting may prevail: e.g., high Ca in some pegmatites of Madagascar culminates in the crystallization of liddicoatite rather than elbaite-rossmanite. Experimental investigations of tourmaline stability are complicated by the high chemical variance of tourmaline- forming systems. The common occurrence of tourmaline with albite and quartz means that activities of Na and of Si can normally be ignored. The chemical stability of tourmaline, therefore, hinges upon the activities of B, Al, H2O, F, and the multitude of occupants of the Y octahedral site. Despite its complexity, the stability field of common tourmaline (schorl-dravite-olenite-foitite) is now reasonably well defined in terms of composition, P, and T. The solubility of SOFTur in melt or in aqueous fluid varies with temperature according to an exponential (Arrhenian) relationship of CB2O3melt, vapor ~ 0.0032e0.0087T. Increasing aH2O or aF (independently) destabilize SOFTur. Fluorine does so by complexing with Al to reduce the aAl2O3 in melt or vapor. Increasing aH2O raises the solubility of excess Al (ASI) of melt, which is achieved by the enhanced dissolution of peraluminous minerals like tourmaline. Contrary to expectation, therefore, tourmaline may survive anatectic events when the aH2O attending melting is low. Tourmaline of other novel compositions (e.g., with 7 wt% Ag in the X site, or fully vacant X site) has been synthesized, but these are of limited petrologic value. Elbaite has been synthesized recently, which is significant for understanding pegmatite occurrences as well as for the possible influx of synthetic elbaite on the gem market. Tourmaline is potentially useful as a geothermometer in the same way as other AFM minerals: through experimental calibrations of elemental and stable isotope exchange equilibria. The fractionation of 11B/10B between dravite and aqueous vapor was calibrated experimentally, but applications of δ11B have been limited by the lack of a corresponding mineral for which δ11B is known. The systematics of 18O/16O, D/H, and 7Li/6Li exchange between tourmaline and other minerals are essentially unstudied. Experimental calibration of Fe-Mg exchange equilibria between tourmaline and biotite has just recently begun, and the preliminary results indicate that the complexities of coupled substitutions in relation to the multitude of site occupancies will make this a difficult, if not intractable effort. Nonetheless, if tourmaline is to grow up into a mature petrogenetic tool, then experimental calibrations of Fe- Mg exchange, stable isotope exchange, and a comprehensive and systematic data base on δ11B in natural tourmaline constitute the necessary next steps.


Oscillatory Zoned Liddicoatite from Central Madagascar: Crystal Chemistry, Crystal Structure, and Compositional Variation

* Lussier, A J (umlussi0@cc.umanitoba.ca), University of Manitoba, Department of Geological Sciences Wallace Building, Fort Garry Campus, Winnipeg, MB R3T 2N2, Canada
Hawthorne, F C (frank_hawthorne@umanitoba.ca), University of Manitoba, Department of Geological Sciences Wallace Building, Fort Garry Campus, Winnipeg, MB R3T 2N2, Canada
Aguiar, P M (pedrom.aguiar@gmail.com), University of Manitoba, Department of Chemistry Parker Building, Fort Garry Campus, Winnipeg, MB R3T 2N2, Canada
Michaelis, V (vladimirkm@gmail.com), University of Manitoba, Department of Chemistry Parker Building, Fort Garry Campus, Winnipeg, MB R3T 2N2, Canada
Kroeker, S (kroekers@cc.umanitoba.ca), University of Manitoba, Department of Chemistry Parker Building, Fort Garry Campus, Winnipeg, MB R3T 2N2, Canada

A slice oriented parallel to (001) of a large, single crystal of liddicoatite from the Anjanabonoina Pegmatite in Madagascar showing pronounced, oscillatory zones was examined. 11B and 27Al Magic Angle Spinning Nuclear Magnetic Resonance spectroscopy (MAS NMR) gives no evidence of tetrahedrally- coordinated B or Al at the T-site throughout the entire crystal and hence Si = 6 apfu. Site-scattering refinements of 26 sub-samples (1.5 < R1 < 2.9% ) and <T-O> values are in agreement with full Si-occupancy. Structure refinement data also confirm that YAl = 6 apfu through the entire crystal, and hence only X- and Y-site occupants are involved in the observed zoning patterns. The crystal shows two types of colour zoning. First, from crystal core to edge, the average colour of the crystal varies through purple, light green, dark green, to greenish-black, where each of these zones is on the order of centimeters. These broad colour variations correlate with decreasing and increasing average Mn and Fe contents, respectively. Second, throughout the entire crystal there are approximately 50 oscillatory zones. In the central region of the crystal, zones are oriented at approximately 45 degrees to the (001) plane and they range in width from <1 to 8 mm; toward the edge, they are parallel to the {110} prism and are consistently << 1 mm in width. Each of the inclined zones is marked by a sharp, dark discontinuity that fades in colour intensity with increasing distance from the core. A comprehensive suite of electron microprobe data shows that Fe, Mg, Li, Ca, and Na contents vary in accord with the oscillatory zoning, whereas Mn and Al contents do not. The dark edge towards the inner margin of each oscillatory zone correlates with a spike in (Fe + Mg) of approximately 0.05 - 0.10 apfu, followed by an exponential decay to (Fe + Mg) → 0. This pattern is repeated at the onset of each new zone. However, despite the conspicuous nature of the oscillatory zoning, nowhere is the crystal observed to have an (Fe+ Mg) exceeding 0.2 apfu. This is sharply contrasted by the monotonic decrease in Mn from ∼ 0.65 apfu in the crystal center, to < 0.02 apfu close to the crystal edge, and the nearly constant YAl content (∼ 1.1 apfu) observed throughout the bulk of the crystal. These preliminary observations of element-selective zoning patterns are consistent with the operation of coupled-feedback mechanisms that relate f(c,t) with f(c,x) via a series of nonlinear expressions for growth-velocity, element partitioning, and nutrient diffusion. This dynamical systems approach to crystal growth suggests that internal, crystallographic controls on compositional variability are dominant during crystallization, as opposed to external, rhythmic variations in composition and/or pressure resulting from melt evolution in either a magmatic or hydrothermal regime.


A Tourmaline is Forever -- Sort of

* Henry, D (glhenr@lsu.edu), Department of Geology and Geophysics, Louisiana State University, Baton Rouge, LA 70803, United States
Dutrow, B (dutrow@lsu.edu), Department of Geology and Geophysics, Louisiana State University, Baton Rouge, LA 70803, United States

Tourmaline is known to have a very wide thermal and baric stability range i.e. from ~100-150° C and 0.1- 0.2 MPa to >900° C and >6-7 GPa. It exhibits extensive compositional variability that responds to the local environment in which it forms. Because it has very limited volume diffusion, even at high temperatures, tourmaline retains this chemistry. Further, tourmaline, together with zircon and rutile, is considered one of the most mechanically stable minerals in clastic sediments. Even though tourmaline appears to be an everlasting repository of geochemical information, it is not necessarily so -- there can be modifications to tourmaline in igneous and metamorphic rocks as the petrologic environment evolves. A systematic overview of tourmaline found in the well-characterized metapelitic rocks of western Maine (USA) illustrates the manner in which tourmaline textures and chemistry evolves. An extensive collection (>600 samples) of metapelites covers the spectrum of metamorphic grades (chlorite through second sillimanite zone) under roughly isobaric conditions (∼ 400 MPa) during the dominant regional-contact metamorphic overprint (M3) related to the intrusion of ∼ 380 Ma sill-like granitoids. Prior to M3 the metamorphic history of these metapelites includes an initial greenschist facies (M1) metamorphism of Ordovician to Devonian basinal sediments at > 400 Ma followed by D1 deformation. This was followed by regional M2 metamorphism that was initiated by intrusion of ∼ 400 Ma granitoid plutons at 300 MPa which outlasted contemporaneous D2 late slip cleavage and resulted in the formation of characteristic andalusite-staurolite- biotite assemblages. Throughout the complex history (clastic sedimentation to subsequent polymetamorphism), tourmaline responds by development of temperature-dependent overgrowths on detrital tourmaline grains as a consequence of B release associated with continuous and discontinuous reactions. At higher grades pre-existing tourmaline grains become partially resorbed reflecting the presence of reactive fluids. Further, a completely new generation of tourmaline can form as a consequence of the advection of B- enriched fluids derived from local plutons. The whole of the metamorphic history and prehistory are contained under multiple guises in the tourmaline found in these rocks.


Order-Disorder Mechanisms in the Tourmaline-Group Minerals

* Bosi, F (ferdinando.bosi@uniroma1.it), Department of Earth Sciences, Sapienza University of Rome, P.le Aldo Moro, 5, Rome, 00185, Italy
Andreozzi, G B (gianni.andreozzi@uniroma1.it), Department of Earth Sciences, Sapienza University of Rome, P.le Aldo Moro, 5, Rome, 00185, Italy

Tourmaline-group minerals are complex borocyclosilicates in which the distribution of cations over the Y and Z non-equivalent octahedral sites is often associated with order-disorder processes. Structural refinement and stereochemical investigations showed a significant disorder of Al, Mg, Fe3+ and Fe2+, expressed in Mg and/or Fe2+ occurring at the Z site. In accordance with bond valence theory, Hawthorne (1996, Can Min, 34, 123-132) showed that disordering of Mg and Al over Y and Z sites may occur via the following substitution: 2YMg + ZAl + WOH = 2YAl + ZMg + WO2-. Similar arguments also apply to their Fe2+-Fe3+ analogues: 2YFe2+ + ZFe3+ + WOH = 2YFe3+ + ZFe2+ + WO2-. Noteworthy, the ratio ZFe2+/YFe3+ = 1/2 is consistent with the Fe distribution observed in several tourmalines. From a steric point of view, the above mentioned disorder is made by the incorporation of smaller cations (R3+ = Al, Fe3+) into Y and larger cations (R2+ = Mg, Fe2+) into Z. As a result, <Y-O> mean bond length decreases and <Z-O> increases. To explain the Mg-Al and the Fe2+-Fe3+ disordering process, as well as the occurring of B at the T site in tourmalines, by geometrical fitting of T6O18 and YO6 polyhedra, analogies have been drawn between tourmaline and lizardite structures, both characterized by six-membered tetrahedral ring T6O18 and three-equivalent octahedra YO6. In tourmaline, the disordering of Mg and Al occurs to relieve the strain due to the misfit of the three YMgO6 octahedra within the Si6O18 ring. Such an analogy for relieving the strain also applies to the Al-rich tourmaline containing B excess. Tourmaline crystal chemistry appears to be mainly controlled by structural rather than chemical factors. The lack of ordered dravite and schorl in nature may be ascribed to structural constraints rather than to a hiatus in petrologic conditions.


Crystallographic Preferred Orientation of Tourmaline Driven by Dissolution-Precipitation Creep

* Spiess, R (richard.spiess@unipd.it), Department of Geoscience - University of Padua, Via Matteotti 30, Padua, 35137, Italy
Bavila, M, Department of Geoscience - University of Padua, Via Matteotti 30, Padua, 35137, Italy
Dibona, R, Department of Geoscience - University of Padua, Via Matteotti 30, Padua, 35137, Italy

Grain size reduction and development of crystallographic preferred orientations (CPOs) are most easily explained for minerals deforming in the dislocation creep regime. For tourmaline the activation of slip systems is unknown, yet within schists of the European Alps tourmaline crystals grown within veins become strongly grain size reduced and develop a strong CPO during deformation. Grain size reduction is accommodated by a cataclastic process leading to the development of discrete fine-grained shear zones that envelope larger fractured clasts. X-ray mapping shows that the boundaries of these shear zones coincide with a sudden change in tourmaline chemical composition. EBSD analysis reveals that the crystallographic orientation of the tourmaline crystals becomes strongly modified in correspondence to the shear zone boundaries, with the prism planes progressively being aligned parallel to the shear zone boundaries, and the strength of the CPOs growing with increasing strain. All analytical data are consistent with an hypothesis that the measured CPO developed in a regime that has driven dissolution and precipitation of tourmaline in sites of high localized deformation. Vein parallel fluid flow reasonably enabled this deformation mechanism. The development of a CPO during dissolution-precipitation creep has been theoretically modeled by Tullis (1989) and Bons and den Brok (2000) for minerals whose rates of dissolution and growth is crystallographically controlled. Our study shows that tourmaline appears to be a good candidate for such a behavior, and it may be worth testing if tourmaline develops CPOs that vary systematically as a function of changing environmental conditions.


There and Back Again, a Complete History of Subduction and Uplift Recorded by a Tauern Window Tourmaline

* van Hinsberg, V (V.J.vanHinsberg@gmx.net), Hydrothermal Geochemistry Laboratory - McGill University, 3450 University Street, Montreal, QC H3A 2A7, Canada
Franz, G (gerhard.franz@tu-berlin.de), Fachgebiet Mineralogie - Technische Universitaet Berlin, Ackerstrasse 76, Berlin, 13355, Germany

Minerals are our best source of information on conditions and processes in the Earth's interior. The presence or absence of a mineral as well as its chemical and isotopic composition can act as precise indicators of the pressure, temperature and chemical conditions at depth. Tourmaline is especially suited for this, due to its singularly large stability range in P-T and bulk rock composition. It is further characterised by highly variable trace element chemistry, which, in combination with negligible diffusion rates, allows it to capture and preserve a chemical signature of its environment. Here we present the results of a detailed investigation of a tourmaline grain from the eclogite zone of the Tauern tectonic window in the Austrian Alps. Using tourmaline inter-sector thermometry [1] we can show that this crystal has experienced the full subduction to uplift history of its host rock and preserved this in subsequent growth zones. The chemical composition of these growth zones further provides information on the associated pressures, recording a progression from subduction to transfer to the overriding plate to denudation uplift. It further allows reconstruction of the mobility of F, Ca and Na in its host environment. These results indicate that tourmaline is a comprehensive monitor of the conditions in its host rock, with wide applicability. [1] van Hinsberg, V.J. and Schumacher, J.C. (2007) Intersector element partitioning in tourmaline; a powerful single crystal thermometer. Contributions to Mineralogy and Petrology 153, 289-301.


The B-Site in Tourmaline as a Key to Cation Ordering

* Clark, C M (christine.clark@emich.edu), Department of Geography and Geology, Eastern Michigan University, 205 Strong Hall, Ypsilanti, MI 48197, United States
Sochalski, L M (lsochals@vt.edu), Department of Geosciences, Virginia Tech, 4044 Derring Hall (0420), Blacksburg, VA 24061, United States

One of the more challenging puzzles with tourmaline is determining the nature of cation occupancies at the Y- and Z-octahedral sites. Traditionally aluminum and other trivalent cations have been preferentially assigned to the Z-site, although there is abundant evidence for more complicated ordering. The large number of cations commonly present in these two sites is the main cause for this difficulty in site assignment. Of all the cation sites in tourmaline, the B-site is the most constrained, as it is commonly presumed to be completely occupied by boron. Stereochemically, the BO3 triangle is not static however, with coupled length variations occurring between the B-O2 and B-O8 as a direct consequence of changes in occupancy of the adjacent cation sites. Statistical analysis has shown that the Z'-O8 bond has the largest effect on the B-site geometry. We are currently studying this relationship to try to ascertain why this particular bond has the greatest effect, and what that can tell us about the occupancy of the Z-site.