Book of abstracts - Tourmaline 2017

10 downloads 12 Views 7MB Size Report
Jun 28, 2017 - Ms + topaz) close to a small quartz core. .... both cases, tourmaline preserves signatures of ...... detector with a Mercury (XIA) digital signal.

The authors are solely responsible for the content of the abstracts.

Table of contents

KT-Keynote Talk, T-Talk, P-Poster




Bačík Peter


The crystal-chemical autopsy of octahedral sites in tourmalines: site occupancy and bond-length relations


Ivan. A. Baksheev


Isomorphic substitutions and trace elements in hydrothermal tourmalines


Experimental studies on tourmaline


Authigenic tourmaline in Carboniferous and Permian siliciclastic rocks of western Poland


Tourmaline structure and crystal chemistry


Eleanor J. Berryman


Julita Biernacka


Ferdinando Bosi


Ferdinando Bosi et al.


Schorlitic and foititic tourmalines from the Arbus pluton, Sardinia, Italy: A first detailed crystal-chemical study


Igor Broska et al.


Inspecting tourmaline distribution in granites using 3D tomography


Zbyněk Buřival, Milan Novák


Hydrothermal replacement of amblygonite-montebrasite by blue tourmaline – examples from the Moldanubicum, Czech Republic


Ismahane Chaouche et al.


Synthetic study of the sources and origin of the gold deposit in the western Hoggar prospect (South of Algeria)


Marta S. Codeço et al.


Hydrothermal tourmaline from the Panasqueira W-Sn-Cu deposit, Portugal: chemical and B-isotope constraints on fluid source and evolution


Renata Čopjaková et al.


Simultaneous formation of fluor-schorl and fluor-elbaite in the Kracovice pegmatite, an effect of Fe and Mg on tourmaline stability


Barbara L. Dutrow, Darrell J. Henry


Tourmaline compositions: Reflecting the fluid phase


Andreas Ertl, Ekkehart Tillmanns


Tetrahedral substitutions in tourmaline and their applications in petrology



Mineralogical characterization of tourmalines from Mata Azul Pegmatitic Field, Central region of Brazil


Ti-rich schorl-bosiite from the Řečice elbaite -subytpe pegmatite


In situ X-ray diffraction studies on tourmalines at high pressures and temperatures


Twenty years of progress: Advances in tourmaline studies since the Tourmaline 1997 Conference


Simone Ferreira da Silva et al. Tomáš Flégr et al.

Klaus-Dieter Grevel et al. Darrell J. Henry, Barbara L. Dutrow



Monika Kowal-Linka


Detrital tourmalines from the Triassic and Cretaceous sedimentary rocks from the NE foreland of the Bohemian Massif (S Poland)


Lukáš Krmíček et al.


Boron isotopic variations in tourmaline from metacarbonates and associated calc-silicate rocks of the eastern part of the Bohemian Massif


James S. Lambert-Smith et al.


B-isotopes in tourmaline indicate common fluid sources in Orogenic Gold and Fe Skarn deposits in West Africa


Elizabeth Levy et al.


Geochemical Analysis of Tourmaline: Coupling EPMA and XANES





Mahjoubi El Mahjoub et al.


Tourmaline as a recorder of hydrothermal processes and a guide to mineral exploration in Achmmach tin deposit, Moroccan Central Hercynian Massif


Katharina Marger et al.


Tourmaline reference materials for oxygen isotope analysis by SIMS


Horst R. Marschall Nancy McMillan et al.

Federico Pezzotta



Tourmaline isotope geochemistry


Lithologic determination of tourmaline based on Laser-Induced Breakdown Spectroscopy: An alternative approach to provenance studies


Tourmaline crystal growth in gem-bearing pegmatites


Adam Pieczka, Mateusz Sęk


Bosiite or not bosiite? That is the question…


Jiří Prokop et al.


Hydrothermal alteration of tourmaline from pegmatites enclosed in serpentinites; examples from the Moldanubian Zone (Czech Republic)


Charlotte Redler et al.


Major and trace element composition of tourmaline in aplites and granites as indicator for magmatic and hydrothermal development (Elba, Italy)


Marjorie Sciuba et al.


Emily Scribner et al.


Uvite and feruvite from lamprophyre dikes and metasediments near the O’Grady batholith in the Northwest Territories, Canada


Tatiana Setkova et al.


The problems of tourmaline single crystals growth


Ryuichi Shinjo et al.


Development of boron isotopic analysis of tourmaline by LA-MC-ICP-MS and its application to natural samples


Lenka Skřápková et al.


Fractionation in Li-tourmaline from the Lhenice lepidolite pegmatite - "darrellhenryite loop" in F-saturation


Tamás Spránitz et al.


Petrographic study of tourmaline-rich metamorphic rocks in the Sopron area, Eastern Alps (W-Hungary)


Vladimír Šrein et al.


Tourmaline in graphitic material as a potential provenience indicator in archaeology


Jiří Toman et al.


Tourmaline from Věžná I pegmatite; an indicator of external contamination and internal fractionation


Tourmaline applications to ore systems


Robert B. Trumbull


Composition of tourmaline in orogenic gold deposits


Pavel Uher et al.


Tourmalines of the siderite-quartz-sulphide hydrothermal veins, Gemeric Unit, Western Carpathians, Slovakia: crystal chemistry and genetic aspects


Scarlett Urbanová, Jan Cempírek


Tourmaline vs. rare-element mineralization in the Velká Bíteš pegmatite field, Czech Republic


Trace elements in tourmaline


Vincent J. van Hinsberg et al.


Oleg S. Vereshchagin et al.


Structural adjustments, stability and mineral species relations of 3delements bearing tourmalines


Oleg S. Vereshchagin et al.


Provenance interpretation of tourmalines and garnets from Mesozoic rocks of northern part of Siberian craton


The crystal-chemical autopsy of octahedral sites in tourmalines: site occupancy and bond-length relations 1


Peter Bačík#

Comenius University in Bratislava, Faculty of Natural Sciences, Department of Mineralogy and Petrology, Ilkovičova 6, 842 15 Bratislava, Slovak Republic, # [email protected]

Key words: tourmaline, octahedron, site occupancy, bond length

INTRODUCTION The structure of tourmaline-supergroup minerals with the general formula XY3Z6(T6O18)(BO3)3V3W include two types of octahedral sites (Henry et al. 2011). The ZO6 octahedron is smaller than the YO6 octahedron and is more distorted (Henry and Dutrow 1996). Although the Z site is predominantly occupied by trivalent cations, usually Al, some tourmaline group end-members have Cr, V, or Fe3+ as the dominant Z-site cation accompanied by Mg. Moreover, Ca tourmalines have Z-sites occupied by both Mg and Al (Henry et al. 2011) and Mg occupancy also can increase because of Al-Mg disorder (Hawthorne et al. 1993). The Y site can accommodate a larger range of valences in accessing elements such as monovalent Li, divalent Fe, Mg, Mn, trivalent Fe, Al, Cr, V and tetravalent Ti (Henry et al. 2011). Although the site occupancy normally follows Goldschmidt rules because the ZO6 octahedron is smaller and more distorted than YO6 octahedron, local structural and bondvalence requirements may result in various cation disorder. Along with the Al-Mg disorder (Hawthorne et al. 1993), the Cr and V disorder between Y and Z sites was documented (Bosi et al. 2004, Cempírek et al. 2013). The order-disorder reactions were studied in detail by structural-refinement methods. More general statistical approach was applied on Mg, Al and Fe ordering at octahedral sites in schorl-dravite series (Bačík 2015). Here, the focus is broadened including other Na- and Aldominant tsilaisitic-elbaitic, olenitic and buergeritic (including those artificially prepared by heating experiments) tourmalines.

The dependence of selected octahedral bond lengths, bond-length distortion and the site occupancy is studied. YO6 OCTAHEDRON Bond-length and site-occupancy analysis in YO6 octahedra revealed some intricate dependencies. Total Fe displays positive correlations with all bond lengths except Y-O2 which is without correlation in schorl-dravite series. In olenitic tourmalines, similar behaviour is observed but there is positive trend indicated between Fe and Y-O2. It is similar to elbaitic-tsilaisitic tourmalines, which, however, have weaker to absent positive correlations of Fe to other bond lengths. Buergerites are totally out of the general trend due to the dominance of Fe3+ over Fe2+. In contrast, Mg displays negative trends with bond lengths except Y-O2 with a positive correlation in schorl-dravites. In other tourmalines, correlations are limited due to low Mg content. Interestingly, Mn does not display any correlation with bond lengths except Y-O6. This different behaviour to chemically similar Fe likely results from its affinity to Li-bearing tourmalines in which increased YLi and YAl content has opposite effect on expansion induced by the presence of YMn in olenites and elbaites-tsilaisites. It is well documented by the negative correlations between Li and bond lengths in YO6 octahedra. However, there is a difference between olenites and elbaitestsilaisites; olenites displays larger reduction of bond lengths.


There is a difference between schorls-dravites and other tourmalines in dependence of Y-O bond lengths and Al. Schorlitic and dravitic tourmalines and also buergerites display negative correlation only between Al and O2, which was described as a result of bondvalence requirements in Al-rich tourmalines with preferential ZAl occupancy resulting in a weakened Y-O6 bond and consequent Y-O2 bond shortening (Bačík 2015). In contrast, Alrich and Mg-poor olenites, elbaites and tsilaisites display negative correlation of Al to all bond lengths. However, in case of Y-O2 and Al, all tourmalines fall into the one unique trend, which suggest that this is general property of (at least) Al-dominant tourmalines.

correlation between Al and Z-O3, Z-O7, Z-O7', Z-O8 and Z-O8'. There is, however, difference in dependence of Al and Z-O6 bond length between schorls and dravites on the one side and elbaites, tsilaisites and olenites on the other side. The first group displays negative correlation, while the second correlates positively. Negative correlation in tourmalines with lower Al can be explained by the presence of ZMg substituting for ZAl; with decreasing Z Mg and increasing ZAl, the bond length shortens. However, due to low Mg content and no other possible elements substituting for Al, this mechanism cannot be applied to explain positive correlation in Al-rich tourmalines. However, if we consider that O6 is shared between Z and neighbouring Y site, increase in Y Al content results in observed shortening of YO6 bonds and consequent expansion of Z-O6.

ZO6 OCTAHEDRON Similar analysis in ZO6 octahedra also revealed differences between compositional types of tourmalines; olenites, elbaites and tsilaisites formed different trends compared to schorls and dravites. Moreover, differences were observed in behaviour of selected cations. Li and Mn display no significant correlations and only small variability in bond lengths. It suggests that their Z site is almost completely occupied by Al and variations in bond lengths result from either the small proportion of ZMg (Mg-rich olenites) or variable occupancy of neighbouring sites, mainly Y site with Al, Mn and Li. Moreover, Z-O3 bond-length shortening may also result from the deprotonization of O3 (V) site. This assumption is supported by the significant ZO3 shortening in VO-dominant buergerites. Positive correlation of Mg and Z-O bond lengths documents its partial presence at the Z site in all Mg-bearing tourmalines. It is also the only cation other than Al, which presence at the Z site is indicated by this analysis. The relations between Al and Z-O bond lengths are the most interesting because it is in abundance in all studied samples. The Z-O3 bond length correlates negatively with Al in all tourmalines except buergerites, which deviate from the general trend due to VO dominance. This Z-O3 shortening induces expansion of other bonds, which is the most pronounced in Z-O7' and Z-O8' bonds. In other tourmalines, there is one general trend of the negative

This work was supported by the Slovak Research and Development Agency under the contract No. APVV-0375-12 and and the Ministry of Education of Slovak Republic grant agency under the contracts VEGA1/0079/15 and VEGA-1/0499/16. Acknowledgment:

REFERENCES Bačík, P. (2015): Cation ordering in octahedral sites in tourmalines of schorl-dravite series. Canad. Mineral., 53:71-–90. Bosi, F., et al. (2004): Crystal chemistry of the dravite-chromdravite series. Eur. J. Mineral., 16:345–352. Cempírek, J., et al. (2013): Crystal structure and compositional evolution of vanadiumrich oxy dravite from graphite quartzite at Bítovánky, Czech Republic. J. Geosci., 58: 149–162. Hawthorne, F.C., et al. (1993): Reassignment of cation site-occupancies in tourmaline: AlMg disorder in the crystal structure of dravite. Am. Mineral., 78:265–270. Henry, D.J. and Dutrow, B.L. (1996): Metamorphic tourmaline and its petrologic applications. Rev. Mineral., 33:503–557. Henry, D.J., et al. (2011): Nomenclature of the tourmaline-supergroup minerals. Am. Mineral., 96:895–913.


Isomorphic substitutions and trace elements in hydrothermal tourmalines 1


Ivan. A. Baksheev#

Geology Department, Lomonosov Moscow State Univeristy, #[email protected]

Key words: tourmaline species, chemical substitutions, trace elements, hydorthermal deposits INTRODUCTION


Tourmaline is stable over a wide range of conditions from low temperature hydrothermal alteration to high grade and high pressure metamorphic environments, and magmatic conditions. It occurs in caprocks and salt domes, in granite and granite pegmatite, in the volcanic hosted massive sulphide deposits, in the porphyry-copper deposits, in metamorphosed stratabound Pb-Zn deposits, in mesothermal gold and tin deposits, in greisen tin and tungsten deposits, and schist-hosted emerald deposits. In this paper chemical substitutions, trace element compositions, and Fe3+/Fetot values, which allow distinguishing tourmalines from different hydrothermal deposits are reported.

The early dravite–oxy-dravite from low-Fe greisen replacing ultramafic rocks (emeraldbearing glimmerite at the Emerald Mines in the Urals, Russia) is characterized by the uvite type and MgFe-1 chemical substitutions; Fe3+ incorporates at site Y according to the burgerite scheme. The late tourmaline tourmaline belongs to the dravite–magnesio-foitite solid solution with the (NaMg)(Al)-1 and MgFe-1 substitutions. Change in the chemical substitution could indicates decreasing pH of mineralizing fluid. The high fF2 results in the formation of fluor-dravite. Tourmaline from greisen replacing granitic rocks belongs to the dravite–oxy-dravite– schorl–oxy-schorl–fluor-schorl system (Kolyvansky, Krasnokamensk, Russia; Akchatau, Kazakhstan; Yunlong, China). A combination of the low Ca content (up to 0.1 apfu), negative correlation between the F content and proportion of X-site vacancy, and the Mg(Fe)-1 substitutuion is characteristic of this tourmaline The high fF2 results in the formation of fluor-schorl. The low Fe3+/Fetot ratio (up to 10 %) in tourmalines both from ultramafic and granitic greisens testifies to low fO2 of mineralizing fluid. The Ca content in the tourmalines depends on host rocks: tourmalines in granitic greisen are depleted in this element, whereas those in sedimentary and ultramafic greisens are enriched in it. Tourmalines from greisen and accompanying quartz veins are the rihest in Li among all hydrothemal tourmalines; the Li content ranges from 50 to 500 ppm. Sn content in greisen tourmalines indicates type of deposit: Sn or Mo. Tourmaline from Sn deposits contains a few hundered ppm, whereas that from Mo deposits contains a few ppm.

MAGNESIAN SKARN Tourmaline from magnesian skarn belongs to the dravite-uvite solid solution. If magnesian skarn replaces magnesite marble with low Fe content (Kuhki Lal spinel deposit, Tajikistan) then the major chemical substitutions are (CaMg)(NaAl)-1 (uvite type) and (Mg(OH))(AlO)-1 (deprotonation). Isomorphic substitutions in dravite-uvite from Fe-bearing magnesian skarn (Tayezhny iron deposit, Russia) are different: (CaMg2OH-)(Na-1Al-2 O-12-) and (Mg(OH)(Fe3+O2-)-1. The high fF2 in the both cases results in the formation of fluoruvite. The high Fe3+/Fetot value (48%, Korovushkin et al. 1979) in dravite–uvite solid solution of magnesian skarn refelects high fO2 during the formation of magnesian skarn formation.



III. This facts testifies to increasing fO2 during hydrothermal process. Two types of tourmaline were identified in the studied porphyry Sn deposits (Mramorny tin cluster and Pridorozhnoe, Russia): (1) early pre-ore tourmaline and (2) ore-stage tourmaline. The early-stage tourmaline is an intermediate member of the schorl–fluorschorl–foitite and oxy-schorl–“oxy-foitite” series with the vacancy type substitution. The ore-stage tourmaline is schorl-oxy-schorl solid solution with the Fe3+Al-1 substitution identical to that in tourmalines from porphyry copper deposits. The Fe3+/Fetot value for the pre-ore and ore-stage tourmalines ranges from 19 to 25% and from 11 to 23%, respectively that is slightly higher than that of mesothermal Sn deposits. This fact could testify to the higher fO2 during the formation of tourmaline from porphyry deposits as compared to mesothermal deposits. Tourmaline from both Sn porphyry and mesothermal deposits is distinguished by the high Sn content reaching 8000 ppm. The Li content in Sn mesothermal and porphyry tourmalines reaches only 40 ppm.

Tourmalines from porphyry coppermolybdenum gold deposists belong to oxydravite–povondraite and magnesio-foitite– dravite solid solutions. Isomorphic substitutions and contents of species-forming elements reflect increasing fCO2 and fS2 and decreasing fO2 and pH of mineral foming fluids in time. The Fe3+Al-1 substitution in tourmaline from pre-ore propylite is followed by AlO2(Fe2+(OH-))-1 in tourmaline of ore-bearing quartz-sericite altered rock, which changes to Al(NaFe2+)-1 in argyllic tourmaline. Tourmaline of carbonate-base metal (subepithermal) assemblage contains a few hunndreths apfu Ca, whereas in tourmalines from propylite and quartz-sericite altered rock Ca reaches a few tenths apfu. Decreasing fO2 is supported by the Mössbauer data. The Fe3+/Fetot value in tourmalines decreases from propylite (60-80%) through ore-bearing quartz sericite rock (40-50%) to carbonate-base metal assemblage (25-30%). The Mg content in tourmalines of all stages is about 2 apfu that could be explained by the stability of the tourmaline crystal structure. The Li content in the porphyry tourmalines does not exceed 20 ppm.

GRANITOID-RELATED GOLD DEPOSITS Early alteration at granitoid-related gold deposits is propylite, for example at the Berezovsky and Zolotaya Gora deposits in the Urals, Russia. Tourmalines from these rocks belong to the schorl–dravite–oxy-dravite– magnesio-foitite system. The vacancy type substitution and deprotonation characteristic of these tourmalines differ them from propylitic tourmalines at porphyry Cu-Au deposits. However, Fe3+/Fetot value and Li content are similar ranging from 50 to 60% and not exceeding 20 ppm, respectively. The high Fe3+/Fetot value testifies to the high fO2 during the propylite formation. Tourmaline is rare in gold-bearing quartz-sericite-carbonate-pyrite alteration. This tourmaline belongs to the same series with the similar isomorphic substitutions as propylitic tourmaline. However, the Fe3+/Fetot value is much lower ranging from 5 to 20% indicating lower fO2 during alteration.

MESOTHERMAL AND PORPHYRY Sn DEPOSITS Tourmalines of mesothermal Sn deposits were studied in the case of the Russian Far East deposits. The early pre-ore tourmaline I is dravite; tourmaline II of the second Mo stage belong to the dravite–fluor-dravite–schorlfluor-schorl system; tourmaline III associated with cassiterite is schorl. All tourmalines are characterized by the FeMg-1 substitution. However, tourmalines are different in some other substitutions. The vacancy type substitution (R3+(NaR2+)-1, where R3+ = Al, Fe3+, R2+ = Mg, Fe2+) is chararacteristic of tourmaline I. The uvite-type substitution is common in tourmaline II. The fourth generation of tourmaline III is featured by a combination of vacancy and uvite types, whereas the second and third generations are charactwerized only uvite type. Thus, isomorphic substitutions show oscillatory behavior that indicates oscillatory changing Na and Ca concentration in mineral-forming fluid. The Fe3+/Fetot value is much lower than that in the porphyry and skarn tourmalines and close to that in greisen tourmalines. It increases from 3-9% in tourmaline II to 12-16% in tourmaline

REFERENCES Korovushkin, V.V., et al. (1979): Mössbauer studies of structural features in tourmaline of various genesis. Phys. Chem. Minerals, 4:209–220.


Experimental studies on tourmaline 1


Eleanor J. Berryman#

Department of Geosciences, Princeton University, NJ, USA, # [email protected]

Keywords: tourmaline, mineral synthesis, mineral stability, experimental petrology


2011) between tourmaline and other phases; as well as the partitioning of B isotopes between tourmaline and fluid (Palmer et al. 1992; Meyer et al. 2008; Marschall et al. 2009).

Experimental studies on tourmaline are critical for unravelling this supergroup mineral’s complex crystal chemistry. As summarized in the reviews of Werding and Schreyer (2002) and London (2011), they are the major tool to unlocking tourmaline’s established petrogenetic potential. Experiments allow us to restrict the extensive tourmaline system to a controlled number of endmembers; establish the key physicochemical controls on its composition; and systematically quantify their relative effects. Concurrently, synthetic tourmaline crystals allow us to probe variation in tourmaline’s short- and long-range crystal structure with changing composition. This allows us to better understand the nature of solid solution in tourmaline. It will also be crucial to understanding the role of inter-site crystallographic effects on tourmaline composition, a poorly understood, but potentially important control as suggested by the rock record. The first reported synthesis of tourmaline was in 1947 by Frondel et al., followed by Smith (1949). Since then, experimental studies on tourmaline have investigated its stability range, in terms of pressure, temperature, and composition (Frondel and Collette 1957; Robbins and Yoder 1962; Werding and Schreyer 1984; Vorbach 1989; Krosse 1995; von Goerne et al 1999a; Ota et al. 2008) and fO2 (Fuchs et al. 1998); the accommodation of elements in tourmaline’s flexible crystal structure (Tomisaka 1968; London et al. 2006; Setkova et al. 2009; Berryman et al. 2014, 2015; Wunder et al. 2015); the partitioning of major (von Goerne et al. 2011; Berryman et al. 2016a) and trace elements (van Hinsberg

TECHNICAL ADVANCES Tourmaline has been synthesized at conditions ranging from 0.1 GPa (Rosenberg and Foit 1979) to 5 GPa (Krosse 1995) and 300°C (Vorbach 1989) to 952°C (Krosse 1995), reaching tourmaline’s maximum, but not its minimum, pressure and temperature stability limits. The most common approach is the hydrothermal synthesis of tourmaline from its oxide components, with or without seed crystals, in the presence of excess H2O, but growth from natural-mineral starting materials (e.g., Ota et al. 2008) or from a silicate melt (e.g., Wolf and London 1997) have also been done. Since the early synthesis experiments, it has been known that tourmaline is stable in moderately acidic solution (Smith 1949; Frondel and Collette 1957), benefits from double the amount of stoichiometric B (Taylor and Terrell 1967), as well as excess H2O and SiO2. Depending on the target conditions, various apparatuses have been used: standard cold-seal pressure vessel, internally heated pressure vessel, piston-cylinder press, and multi-anvil press. To produce larger crystals, von Goerne et al. (1999b) developed the chamber method, which imposes a chemical gradient in the experimental capsule to inhibit nucleation and encourage crystal growth. Crystals longer than 0.5 mm have been synthesized with this method (e.g., Berryman et al. 2016b). Imposing a temperature gradient is expected to produce a similar effect (Taylor and Terrell 1967).


Due to the low intra-volume diffusion of elements within tourmaline, synthetic crystals are often significantly zoned, with their cores being in clear disequilibrium with the crystallizing medium (fluid or melt). This zoning can result from various phenomena during the experiments, including: early nucleation of tourmaline during heating/ pressurization; Rayleigh fractionation during growth; or the formation of precursor phases (Kutzschbach et al. 2017). Despite creating difficulties in the characterization of synthetic tourmaline, the zoning itself can also provide useful insight into solid solution mechanisms in tourmaline.

tourmaline and fluid. Contrib. Mineral. Petrol., 171, 31. Berryman, E.J., et al. (2016b): Influence of the X-site composition on tourmaline’s crystal structure: investigation of synthetic Kdravite, dravite, oxy-uvite, and magnesiofoitite using SREF and Raman spectroscopy. Phys. Chem. Minerals, 43:83–102. Frondel, C., et al. (1947): Synthesis of tourmaline. Am. Mineral., 32:680-681. Frondel, C. and Collette, R.C. (1957): Synthesis of tourmaline by reaction of mineral grains with NaCl-H3BO3 solution, and its implications on rock metamorphism. Am. Mineral., 42:754–758. Fuchs, Y., et al. (1998): Fe-tourmaline synthesis under different T and fO2 conditions. Am. Mineral. 83:525–534. Krosse, S. (1995): Hochdrucksynthese, Stabilität und Eigenschaften der Borsilikate Dravit und Kornerupin, sowie Darstellung und Stabilitätsverhalten eines neuen Mg-Al borates. Thesis, Ruhr-Universität Bochum, Germany. Kutzschbach, M., et al. (2017): Jeremejevite as a precursor for olenitic tourmaline: consequences of non-classical crystallization pathways for composition, textures and B isotope patterns of tourmaline. Eur. J. Mineral., 29:1–17. London, D., et al. (2006): Synthetic Ag-rich tourmaline: structure and chemistry. Am. Mineral., 91:680–684. London, D. (2011): Experimental synthesis and stability of tourmaline: a historical overview. Canad. Mineral., 54:117–136. Marschall, H.R, et al. (2009): Experimental boron isotope fractionation between tourmaline and fluid: confirmation from in situ analyses by secondary ion mass spectrometry and from Rayleigh fractionation modelling. Contrib. Mineral. Petrol., 158: 675–681. Meyer, C., et al. (2008): Boron-isotope fractionation between tourmaline and fluid: an experimental re-investigation. Contrib. Mineral. Petrol., 156:259–267. Ota, T., et al. (2008): Tourmaline breakdown in a pelitic system; implications for boron cycling through subduction zones. Contrib. Mineral. Petrol., 155:19–32. Palmer, M.R., et al. (1992): Experimental determination of fractionation of 11B/10B between tourmaline and aqueous vapour: a temperature- and pressure-dependent isotopic system. Chem. Geol., 101:123–129.

OUTLOOK Despite the amount of progress made over the past 7 decades of experiments on tourmaline, the assertion made by Smith (1949) remains valid today: it will take many years to study the full tourmaline system. The complexity of tourmaline’s crystal structure gives it great potential as a recorder of its formation environment and provides experimental petrologists with an exciting challenge, but a lot of work remains to be done. It is important to emphasize that both external (i.e., P-T-X) and intra-crystalline effects are expected to be significant in influencing a given tourmaline’s composition, the latter remaining largely unexplored. As a result, current experimental findings are strictly constrained to conditions similar to those of the experiment. The challenges in conducting equilibrium experiments need to be overcome, particularly in regards to the difficulty of buffering the system. Further progress on the number and nature of experimental studies on tourmaline will allow us to make even greater strides in understanding this wonderfully complicated mineral. REFERENCES Berryman, E., et al. (2014): Synthesis of Kdominant tourmaline. Am. Mineral., 99:539– 542. Berryman, E.J., et al. (2015): An experimental study on K and Na incorporation in dravitic tourmaline and insight into the origin of diamondiferous tourmaline from the Kokchetav Massif, Kazakhstan. Contrib. Mineral. Petrol., 169, 28. Berryman, E.J., et al. (2016a): P-T-X controls on Ca and Na distribution between Mg-Al


Robbins, C.R. and Yoder, H.S. (1962): Stability relations of dravite, a tourmaline. Carnegie Inst. Wash. Yearbook. 61, 106-108. Rosenberg, P.E. and Foit, F.F. Jr. (1979): Synthesis and characterization of alkali-free tourmaline. Am. Mineral., 64:180–186. Setkova, T.V., et al. (2009): Experimental growth and structural-morphological characteristics of Co-tourmaline. Dokl. Earth Sci., 424:82–85. Smith, F.G. (1949): Transport and deposition of the non-sulphide vein minerals. IV. Tourmaline. Econ. Geol., 44:186–192. Taylor, A.N. and Terrell, B.C. (1967): Synthetic tourmalines containing elements of the first transition series. J. Cryst. Growth., 1: 238–244. Tomisaka, T. (1968): Synthesis of some endmembers of the tourmaline group. Mineral. J., 5:355–364. van Hinsberg, V.J. (2011): Preliminary experimental data on trace-element partitioning between tourmaline and silicate melt. Canad. Mineral., 49:153–163. Vorbach, A. (1989): Experimental examination on the stability of synthetic tourmaline in temperatures from 250°C to 750°C and pressures up to 4kb. Neues Jahrb. Mineral. Abh., 161:69–83. von Goerne, G., et al. (1999a): Upper thermal stability of tourmaline + quartz in the system

MgO - Al2O3 - SiO2 - B2O3 - H2O and Na2OMgO - Al2O3 - SiO2 - B2O3 - H2O - HCl in hydrothermal solutions and siliceous melts. Canad. Mineral., 37:1025–1039. von Goerne, G., et al. (1999b): Hydrothermal synthesis of large dravite crystals by the chamber method. Eur. J. Mineral., 11:1061– 1077. von Goerne, G., et al. (2011): Experimental determination of Na-Ca distribution between tourmaline and fluid in the system CaONa2O-MgO-Al2O3-SiO2-B2O3-H2O. Canad. Mineral., 49:137–152. Werding, G. and Schreyer, W. (1984): Alkalifree tourmaline in the system MgO-Al2O3B2O3-SiO2-H2O. Geochim. Cosmochim. Acta., 48:1331–1344. Werding, G. and Schreyer, W. (2002): Experimental studies on borosilicates and selected borates. In: Grew, E.S. and Anovitz, L.M. (eds) Boron: Mineralogy, Petrology and Geochemistry, 2nd edn. Mineralogical Society of America, Rev. in Mineral., 33:117–163. Wolf, M.B. and London, D. (1997): Boron in granitic magmas: stability of tourmaline in equilibrium with biotite and cordierite. Contrib. Mineral. Petrol., 130:12-30. Wunder, B., et al. (2015): Synthetic and natural ammonium-bearing tourmaline. Am. Mineral., 100:250–256.


Authigenic tourmaline in Carboniferous and Permian siliciclastic rocks of western Poland 1

Julita Biernacka#


Institute of Geology, Poznan University, Poland, # [email protected]

Key words: authigenic tourmaline, shale, sandstone, conglomerate, diagenesis, Poland


ferous. This tectonic pulse was also responsible for the development of abundant subvertical fractures healed by various minerals at the temperature range of 150300oC (Speczik and Kozłowski 1987). The Carboniferous rocks were at places covered by mafic and acid volcanic rocks that originated at the Carboniferous/Permian transition. The Permian siliciclastic sediments were deposited after a remarkable hiatus of ~30 Ma, in aeolian, alluvial and playa environments and capped with 400-500m thick Upper Permian evaporites. They were buried to the maximum depth of 3-5 km in the Jurassic. An intersecting fault system cutting Carboniferous and Permian-Mesozoic rocks was documented by seismic studies. The most prominent fault system, cutting across the upper and middle crust, is the Dolsk Fault.

Although it has been known for many decades that tourmaline may crystallize in sedimentary rocks, the temperature and pressure of its crystallization have not been well constrained, and the source of chemical components as well as the crystal chemistry relatively rarely have been recorded (reviewed by Henry and Dutrow 2012). Diagenetic mineral assemblages often form in the course of multiple processes, operating during many million years, and rarely, if ever, attain thermodynamic equilibrium. They rather reflect reaction kinetics. Therefore, each successive case study gives insight into this still enigmatic field. In this presentation, I examine the distribution, modes of occurrence and chemical composition of authigenic tourmaline in various Carboniferous and Permian siliciclastic rocks of western Poland.


Authigenic tourmaline occurs in Upper Carboniferous turbiditic greywackes, shales and quarzarenites, in Permian alluvial conglomerates, and in Permian dune porous sublithic arenites. The latter case was described by Pieczka et al. 2011. The largest amounts of the mineral have been noticed in the rocks from W IG-1 borehole situated in the proximity of the Dolsk Fault Zone. Tourmaline is present there in the upper 300m of the Carboniferous verical section, it spans the Carboniferous/Permian unconformity and appears in the 100m thick Permian conglomerates. Authigenic tourmaline shows blue color under the microscope in the all rock types,


The Carboniferous and Permian siliciclastic rocks of western Poland (the area of the ForeSudetic Monocline) are concealed under the Mesozoic-Cenozoic cover of various thickness, from ~1 km to the south to > 4 km in the north, and are known only from borehols. It is believed that Carboniferous and Permian rock successions underwent different diagenetic paths in different times. The Carboniferous sediments, predominantly composed of marine shales, mudstones and sandstones of the Variscan foreland basin, were folded, thrust and underwent peak diagenesis (locally lowgrade metamorphism) in the Late Carboni-


therefore the dense aggregates of the mineral are easily recognizable. On the other hand, individual crystals are exceedingly small, the majority thinner than 1 m, with aspect ratios between 10 and 30. The thinnest crystals observed with the available SEM equipment were tens nm in width. The acicular tourmaline occurs in small bundles or radial aggregates at privileged sites. In the Carboniferous rocks, five modes of occurrence were distinguished. (1) In clay-rich rocks (greywacke, mudstone, shale) it is dispersed in clay matrix, however not evenly in the rock, but in patches that could have been kaolinite accumulations before. Some detrital mica flakes are expanded in a way characteristic for mica kaolinitization, yet instead of kaolinite authigenic tourmaline is present between the expanded flakes. Moreover, XRD studies revealed a reversed relationship between the presence of kaolinite and tourmaline in the < 2m fractions separated from the samples. (2) Tourmaline aggregates, associated with successive quartz, calcite, or anhydrite, replaced some detrital grains that had been totally dissolved. (3) In black, organic matter rich rocks, fine tourmaline crystals concentrate around plant fragments; they are particularly abundant in microfractures cutting these fragments. (4) In quartzarenites, tourmaline needles occur in thin fractures filled with quartz, calcite, or kaolinite. (5) In sandstones containing detrital tourmaline grains, authigenic crystals form subtle monopolar overgrowths on rounded tourmaline nuclei. In the case of tightly cemented quartzarenites, the overgrowths were observed only in the proximity of microfractures. In Permian, coarse-grained volcaniclastic conglomerate tourmaline has been found in clay matrix, in small voids after dissolved grains, and in microfractures, but not in rhyolitic clasts. Tourmaline postdated ubiquitous authigenic K-feldspar. The most conspicuous mode of occurrence exhibits authigenic tourmaline in Permian dune sandstone. The tiny crystals are pore-lining, they coat the entire grain surfaces except for grain contacts, where they are absent. Most frequently, tourmaline crystals form small aggregates oriented perpendicularly or obliquely to grain surfaces. Interestingly, rounded detrital tourmaline grains were also overgrown by authigenic crystals on entire available surfaces, although overgrowths along the C axis are slightly thicker.

MINERAL CHEMISTRY Only largest crystals analyzed with the use of an electron microprobe yielded good results. Tourmaline formulae calculated from electronmicroprobe data show high amounts of vacancies at the X-site (47 to 66%). Besides, the tourmaline crystals are relatively rich in aluminium (Al from 6.56 to 7.17 apfu) and poor in Ca and Ti (Ca in the range of 0.01-0.06 apfu; Ti below the detection limit). The contents of Mg and Fe vary among samples from different depths and cores, but also in individual crystals. Some of them exhibit distinct zoning with Mg/(Mg+Fe) ratios between 0.78 and 0.39. The majority of authigenic tourmalines have magnesio-foititic to foititic composition. The chemical composition of authigenic tourmaline is controlled by the substitutions FeMg-1 and Al(Fe,Mg)-1Na-1. CONCLUSIONS

The acicular habit and exceedingly small sizes of the tourmaline indicate its rapid crystallization. Similar sizes and habits were achieved in days-lasting experimental synthesis of tourmaline. Authigenic tourmaline must have formed as a result of the flow of boron-bearing fluids, percolating through microfracture networks in lithified Carboniferous rocks and through pore system in overlying sandstones. The process took place at elevated temperature (min. 150-200oC) in post-Permian times, during the reactivation of the regional faults. Clay minerals, particularly kaolinite, were the source of aluminum and silica. Boron and magnesium could have been delivered from the overlying Permian evaporites. REFERENCES Henry, D.J. and Dutrow, B.L. (2012): Tourmaline at diagenetic to low-grade metamorphic conditions: Its petrologic applicability. Lithos, 154:16–32. Pieczka, A., et al. (2011): Si-deficient foitite with [4]Al and [4]B from the ‚Ługi-1‘ borehole, southwestern Poland. J. Geosci., 56:389–398. Speczik, S. and Kozłowski, A. (1987): Fluid inclusion study of epigenetic veinlets from Carboniferous rocks, SW Poland. Chem. Geol., 61:287–298.


Tourmaline structure and crystal chemistry 1


Ferdinando Bosi#

Dipartimento di Scienze della Terra, Sapienza Università di Roma, Piazzale Aldo Moro, 5, I-00185 Rome, Italy, # [email protected]

Key words: tourmaline, short-range structure, long-range structure, order-disorder, nomenclature INTRODUCTION

order-disorder mechanisms over the structural sites. Short-range constraints depend on the charge of ions and tend to favor specific chemical arrangements. Of particular importance are the local atomic arrangements around the O1 and O3 sites (linked to 3Y and Y+2Z, respectively) which show a greater chemical variability. Valuable information can be derived from bond valence theory (e.g., Bosi 2011), which allows to predict specific stable local charge arrangements around O1(W– ) such as Y(3R2+)Σ6+ and Y(2R2+R3+)Σ7+, or around O1(O2–) such as Y(R2+2R3+)Σ8+ and Y (3R3+)Σ9+. The corresponding substitution mechanism, m YR2++ n ZR3+ + O1(OH,F)– = m Y 3+ R + n ZR2+ + O1O2– (where m = 3 and n = 2 for R3+ = Cr3+, Fe3+, V3+, while m = 2 and n = 1 for R3+ = Al3+), may explain the chemical relations between hydroxy/fluor- and oxyspecies rich in Mg. Local atomic arrangements may be detected by vibrational spectroscopy (IR and Raman), which is however limited to arrangements involving (OH). In this regard, each distinct local arrangement (YYY)-O1(OH)(X) will give rise to a distinct spectral signal.

The tourmaline supergroup minerals are cyclosilicates rich in boron that typically crystallize in the space group R3m. They are the most common and the earliest boron minerals formed on Earth. Tourmalines are widespread in Earth’s crust, typically occurring in granites and granitic pegmatites but also in sedimentary and metamorphic rocks. The general formula may be written as: XY3Z6T6O18(BO3)3V3W, with X = Na, K, Ca, , Fe3+, Cr, V3+, Mg, Fe2+, Mn2+, Li1+; Z = Al, Fe3+, Cr, V3+, Mg, Fe2+; T = Si, Al, B3+; B = B3+, V = (OH)–, O2–; W = (OH)–, F–, O2– being the most common constituents. The letters in the formula (X, Y, Z, T and B, not italicized) represent groups of cations at the [9]X, [6]Y, [6]Z, [4]T and [3]B crystallographic sites (letters italicized). As for the letters V and W, they represent groups of anions at the [3]O3 and [3]O1 sites, respectively. The H atoms occupy the H3 and H1 sites. The structure may be regarded as a ZO6 octahedral three-dimensional framework that surrounds the “islands” made of XO9, YO6, BO3 and TO4 polyhedral groups. The compositional range of tourmaline is remarkable, with at least 27 major/minor constituents identified in synthetic and natural samples (Tab. 1). The tourmaline supergroup currently consists of 33 species approved by the IMA-CNMNC.

Table 1: Major and minor constituents occurring in tourmaline.


SHORT- AND LONG-RANGE CONSTRAINTS The complex chemical composition of tourmaline is associated with an intermediate structural complexity, which can accommodate ions of a wide range of size and charge. Nevertheless, tourmaline site populations are mainly controlled by short-range (bondvalence) and long-range (geometric fit) constraints, which determine substitution and











Na K+ Li+ Ag+ (NH4) + H+ F– Cl –  

Ca Mg2+ Fe2+ Mn2+ Ni 2+ Co2+ Cu2+ Zn2+ O2–

Al Cr3+ V3+ Fe3+ B3+ Mn3+ Be 3+    

Si Ti 4+              

Notes: Vacancy is considered as a constituent in accord with the IMA-CNMNC rules. In bold are indicated the constituents characterizing the 33 mineral species of the tourmaline supergroup.


3%). Larger amounts have been observed for Z Mg2+ in (V,Cr,Fe)3+-rich tourmalines: up to  33% atoms/site.

However, each distinct arrangements (YZZ)O3 (OH) will not give rise to distinct spectral signal because the principal O–H stretching band of O3(OH) is affected by the three local arrangement 3(YZZ) associated with all three of the O3(OH) groups in the R3m symmetry (Watenphul et al. 2016). The larger clusters (X)–(O1)–(YYY)–[(O3)(O3)(O3)]–(ZZZZZZ) may be chemically identified by combining theoretical and experimental information. Long-range constraints depend mainly on the size of ions and may determine the structural stability of tourmaline. In this regard, disordering of R2+ and R3+ over Y and Z may be necessary to reduce the structural misfit between the YO6 and ZO6 octahedra by maintaining the difference between and smaller than 0.15 Å (e.g., Bosi and Lucchesi 2007). The disordering reaction associated with long-range constraints can be described as YR2++ ZR3+ = YR3+ + ZR2+, and does not require the incorporation of O2– at the O1 site. As an example, consider the empirical formula of fluor-dravite, which shows Mg partially disordered over Y and Z, instead of being ordered at Y, and is O1(O2–)-free (Clark et al. 2011). The disorder reaction YMg + ZAl = Y Al + ZMg occurs to shorten (by introducing Al3+) and enlarge (by introducing Mg2+) in fluor-dravite, and thus accommodate the potential misfit between Y MgO6 and ZAlO6. A similar mechanism for relief of misfit between YO6 and T6O18 could explain the occurrence of TB in Al-rich tourmalines.

SOME NOMENCLATURE ISSUES The current tourmaline nomenclature involves some problems. For instance, application of the rules in force (Henry et al. 2011, 2013) results in ambiguous naming of oxy-tourmalines (Bosi et al. 2017). Moreover, the recommended procedure for the Y and Z site allocation applies to hydroxy- and fluor-tourmalines, but appears unsuccessful for naming oxytourmalines mainly because of the incorrect Z site preference of V3+ and Cr3+, which should be reversed, and the qualitative Al site distribution. Improvements based on an updated Y and Z site allocation procedure are possible considering the above cation site preferences and the strong correlation between Z Al and [6]Al (r2 = 0.99; data points = 71): Z Al = – 0.13 + 1.19 ∙ [6]Al – 0.05 ∙ [6]Al2 Other issues concern, for example, tourmalines refined with symmetry lower than R3m: in accordance with the IMA-CNMNC guidelines on polymorph nomenclature, are they actually new species? REFERENCES Bosi, F. (2011): Stereochemical constraints in tourmaline: from a short-range to a longrange structure. Canad. Mineral., 49:17–27. Bosi, F. and Lucchesi, S. (2007): Crystal chemical relationships in the tourmaline group: structural constraints on chemical variability. Am. Mineral., 92:1054–1063. Bosi, F., et al. (2017): Crystal-chemical relations and classification problems of tourmalines belonging to the oxy-schorl–oxydravite–bosiite–povondraite series. Eur. J. Mineral., DOI 10.1127/ejm/2017/0029-2616 Clark, C.M., et al. (2011): Fluor-dravite, NaMg3Al6Si6O18(BO3)3(OH)3F, a new mineral species of the tourmaline group from the Crabtree emerald mine, Mitchell County, North Carolina: description and crystal structure. Canad. Mineral., 49:57–62. Henry, D.J., et al. (2011): Nomenclature of the tourmaline supergroup minerals. Am. Mineral., 96:895–913. Henry, D.J., et al. (2013): Erratum. Am. Mineral., 98, 524. Watenphul, A., et al. (2016): Exploring the potential of Raman spectroscopy for crystallochemical analyses of complex hydrous silicates: II. Am. Mineral., 101:970– 985.

OCTAHEDRAL SITE PREFERENCES Tourmaline site populations depend on the charge and size of atoms forming specific arrangements that obey both the short-range and long-range constraints. Because is always greater than in tourmaline, the Y and Z sites will tend to incorporate relatively large and small cations, respectively. Many empirical structural formulas show that the site preference for R2+- and R3+-cations is controlled by their sizes. The R3+ preference for Y is YFe3+ > YV3+ > YCr3+ > YAl3+, and vice versa for Z, ZAl3+ > ZCr3+ > ZV3+ > ZFe3+. Similarly, for R2+: YMn2+ > YFe2+ > YCo2+ > Y Mg2+ > YNi2+, and ZNi2+ > ZMg2+ > ZCo2+ > Z Fe2+ > ZMn2+. Further, the amounts of ZR3+ replaced by ZR2+ seem to be an effect of cationsize mismatch. For instance, the amounts of Z 2+ R -cations in Al-rich tourmalines decrease with increasing R2+-size: ZNi2+ (up to 20% atoms/site), ZMg2+ (up to 17%), ZCo2+ (up to 12%), ZFe2+ (up to 8%) and ZMn2+ (up to


Schorlitic and foititic tourmalines from the Arbus pluton, Sardinia, Italy: A first detailed crystal-chemical study 1

Ferdinando Bosi#, 2Stefano Cuccuru, 3Stefano Naitza, 4Henrik Skogby, 2Francesco Secchi, 5 Aida M. Conte, 4Ulf Hålenius, 6Nathaly De La Rosa, 6Per Kristiansson, 6E.J. Charlotta Nilsson, 6Linus Ros, 1Giovanni B. Andreozzi

Dipartimento di Scienze della Terra, Sapienza Università di Roma, Piazzale Aldo Moro, 5, I-00185 Rome, Italy, # [email protected] 2 Dipartimento di Scienze della Natura e del Territorio - Università di Sassari, Sassari, Italy 3 Dipartimento di Scienze Chimiche e Geologiche - Università di Cagliari, Cagliari, Italy 4 Department of Geosciences, Swedish Museum of Natural History, Box 50007, SE-10405 Stockholm, Sweden 5 CNR-Istituto di Geoscienze e Georisorse, UOS Roma, Piazzale Aldo Moro 5, I-00185 Roma, Italy 6 Division of Nuclear Physics, Department of Physics, Lund University, Box 118, SE-22100 Lund Sweden. 1

Key words: tourmaline, composition, structure, Sardinia, Arbus pluton



The tourmaline minerals are boron-rich cyclosilicates. They are the most common and the earliest boron minerals formed on Earth and are widespread in the Earth’s crust, occurring typically in granites and granitic pegmatites but also in sedimentary and metamorphic rocks. The general formula of tourmaline is: XY3Z6T6O18(BO3)3V3W, with X = Na, K, Ca, (= vacancy); Y = Al, Cr3+, V3+, Fe, Mg, Mn2+, Li; Z = Al, Cr3+, V3+, Fe, Mg; T = Si, Al, B3+; B = B3+, V = (OH), O2–; W = (OH), F, O2– being the most common constituents. The chemical and structural complexity of tourmaline reflects the host environment where they formed. It is well-know that the more complicated a mineral is, the more information it may contain about its crystallization environment and subsequent evolution. Therefore, a detailed crystal-chemical characterization of tourmaline is essential to revel the story of an evolving Earth (Dutrow and Henry 2011). In the present study, seven tourmaline samples from the Arbus pluton (SW Sardinia, Italy) were fully characterized from a chemical and structural viewpoint.

The Arbus pluton (SW Sardinia) emplaced during late Variscan post collisional phases at 304±1 Ma (U-Pb dating on zircons; Cuccuru et al. 2016), along a thrust surface separating a pile of greenschist allochtonous tectonic units from the anchimetamorphic para-autochtonous Foreland. Shallow crustal levels of emplacement (< 2 kb) are constrained by quite minute medium-grain rock-textures of intrusives, by a narrow contact aureole dominated by andalusite-cordierite spotted schists and as well as by the Al-in hornblende barometer in granodiorites. The pluton is made up of an outer zone that includes several granodioritic pulses representing different degrees of magmatic evolution (GD1, GD2 and GD3 granodiorites) and of a core zone of cordierite-bearing leucogranites (LG) (Secchi et al. 1991). Field relations suggest that cordierite-bearing leucogranites post-date the granodiorites. The entire pluton forms a roughly E-W trending elliptical body, emplaced within a narrow E-W trending shear zone, that reactivate the allochtonous/ parautochtonous contact (Fig. 1). All the intrusives belong to high-K, ilmenite-series. Overall, a complex petrogenetic scenario of pyroxene and amphibole rock-series may be


Figure 1. Geological sketch map of the Arbus pluton: a) metamorphic basement (dark green - allochtonous units, light green - Foreland); b) monzogabbro intrusion; c) GD1 (dark pink), GD2 (intermediate pink), and GD3 (light pink) granodiorites; d) LG leucogranite intrusions; e) post-Variscan covers; f) main thrust.

envisaged; among them, GD3 granodiorites and LG leucogranites represent the last and most All the intrusives belong to high-K, ilmenite-series. Overall, a complex differentiated magmatic pulses that also produced numerous swarms of granophyric dikes. The strong and widespread subsolidus alteration (e.g. uralitization of pyroxene) of the granodiorites, frequent pegmatitic pods and layers, miarolitic facies with interstitial quartz/Fe-cordierite graphic intergrowths and greisens developed in GD3 and LG, are indications that magma emplacement was associated with a large and long-lasting circulation of fluids. A distinctive feature of the Arbus granitoid intrusions, recognized at regional scale, is the presence of tourmaline, which could lead to identify a B-rich series of granitoids. The occurrence of tourmaline in the Arbus pluton, firstly recognized in 1933 in the granophyres and in hornfels facies in the contact aureola, has never been studied in detail. Recent field surveys (Cuccuru et al. 2016) recognized tourmalines in various occurrences: (a) in pegmatitic layers and in greisenized facies in the GD3 granodiorites; (b) in miarolitic facies and pegmatite pods in LG leucogranites; (c) in nests and fine aggregates in granophyres; (d) in pegmatitic veins and disseminations in contactmetamorphic facies.

TOURMALINE LOCATIONS The sampling locations of the seven samples here investigated are displayed in Fig. 1. AP23 and AP31 (occurrence a) come from tourmaline nests in greisenized facies and grains in pegmatitic layers, respectively, within medium-grained granodiorite labelled as GD3; AP25 (occurrence b) comes from pegmatite pods in medium-grained LG leucogranites; AP43 (occurrence c) comes from fine-grained tourmaline nests in granophyric dikes; AP47 dark, AP47 bright and AP57 (occurrence d) come from contact aureole. In particular, AP47 dark and bright come from tourmaline grains associated with quartz and K-feldspar in hydrothermal-pegmatitic veinlets embedded in the schists about 100 m far from the contact with the GD3 granodiorite; AP57 comes from tourmaline nests in thermometamorphic schists at the contact with granodiorites GD1. MINERAL CHEMISTRY Tourmaline crystals were structurally and chemically characterized by single-crystal Xray diffraction, electron microprobe analysis, Mössbauer, infrared and optical absorption spectroscopies. Measurements of boron and lithium by nuclear reaction analysis are in progress. Unit-cell parameters are typical of Fe2+rich, Al-dominant tourmalines: a-parameter


is typical of tourmalines crystallized in moderately fractionated granitic rocks. Sample AP57 shows the highest concentration of Mg (0.9 apfu), Ca (0.3 apfu) and Ti (0.2 apfu). Samples AP23, AP25, AP31 and AP43 show a similar bulk composition dominated by the schorlitic phase: Fe2+ about 2 apfu, Mg < 0.3 apfu, Ca and Ti < 0.1 apfu. In sample AP47 two tourmaline phases (dark and bright) were identified, separated by a quartz layer. The dark phase hasa shorlitic composition whereas the bright one has a foititic composition. REFERENCES Figure 2. Ternary system for tourmaline groups based on the X-site occupancy.

Cuccuru, S., et al. (2016): Structural and metallogenic map of late Variscan Arbus Pluton (SW Sardinia, Italy). J. Maps, 12:860–865. Dutrow, B.L. and Henry, D.J. (2011): Tourmaline: A Geologic DVD. Elements, 7: 301–306. Secchi, F., et al. (1991): The Arburese igneous complex: an example of igneous fractionation leading to peraluminous granites as residual melts. Chem. Geol., 92:213–249.

from 15.95 to 16.00 Ǻ, c-parameter from 7.15 to 7.18 Ǻ. Chemical results indicate that the analyzed tourmalines are alkali and X-vacant species (Fig. 2) belonging to (fluor-, oxy- and hydroxy-) schorl–foitite series, with Si ~ 6.0 atoms per formula unit (apfu), Al ~ 6.3-7.0 apfu, Fe2+ ~1.2-2.1 apfu, Mg ~ 0-0.9 apfu, Na ~ 0.4-0.7 apfu, F ~ 0-0.4 apfu and a number of X-site vacancy of ~ 0.3-0.6. This composition


Inspecting tourmaline distribution in granites using 3D tomography 1


Igor Broska#, Ján Madarás, Michal Kubiš, Juraj Šurka, Tomáš Mikuš

Earth Science Institute of the Slovak Academy of Sciences, Dúbravská cesta 9, 840 05 Bratislava, # [email protected]

Key words: tourmaline, nodules, dendrites, granite, aplite, Gemer unit, Himalaya


apical granites contain also tourmaline clusters, nodules and dendrites. Nodular tourmaline form oval shaped tourmaline clusters with intercalation of alkaline feldspars and quartz but also in form of massive tourmaline. Nodular tourmaline and especially dendritic shaped indicates rapid crystal growth.

Tourmaline in granites forms scattered individual grains, clusters of grains, nodules, but also dendrites. Its antagonistic relationship against to other mafic minerals such as biotite is typical. High resolution X-ray microtomography at conditions 220 kV/350µA offers new insights to tourmaline distribution in granites where tourmaline is a main mafic heavy mineral. The 3D inspection of tourmaline distribution in Gemeric granites of the Western Carpathians and leucogranites from the Himalayan chain offers new implications for understanding of genesis of tourmaline nodules.

TOURMALINE FROM GRANITES OF THE EVEREST AREA The Everest region contains large volume of crustal melt leucogranites with tourmaline where monazite records a crystallisation age of 20.5 – 21.3 Ma - ages; Simpson et al. 2000). The granite forms veins and sills with varying thickness up to 100 m, or large ballooning bodies – spectacular domed shaped granite is the Nuptse massif. High Himalayan leucogranites penetrated all rock complexes within Himalayan ridge. Granites have been originated in the several phases during the continental collision of the India and Tibet Plates. The Everest – Lhotse – Nuptse tourmaline bearing leucogranite massif form massive ballooning sill up to thickness 3000 m, which reaches 7800 m on the Kangshung face of Everest or on the south face of Nuptse, and is responsible for the extreme altitude of both mountains (Searle et al. 2003). Tourmaline in the leucogranite from Everest area forms individual grains but also clusters and nodules. Some of tourmaline are in dendritic form due to rapid growths (Fig. 1).

TOURMALINE FROM GRANITES OF THE WESTERN CARPATHIANS The granitic rocks of the Gemeric unit are Permian in age (ca 264 Ma) and they are named as the specialised S-type granites because of the occurrence of special mineralisation in their granite apical parts. These granites are mainly biotite-muscovite leucogranites with their evolved rare metal phases in cupolas. Granites characterise high SiO2 content, prevalence of K over Na, peraluminousity, low Ca, Mg, Ba, Sr, Zr and REEs, but high Li, Rb, Sn, Nb, Ta, F, B, and P (e.g. Broska et al. 2002). The granites are crop out in eight bodies which represent the apexes of larger hidden granite pluton. Tourmaline in the deeper-seated granites form scattered isolated schorlitic tourmaline crystals but the


that are equilibrated in different proportions. The origin of tourmaline clusters and veins in granites of the Gemeric unit thus resulted from immiscible vapor-rich borosilicate liquid produced during late stages of crystallisation and proces is very complex and combination of different regimes. Generaly tourmaline precipitation during this late stages of granite consolidation should be connected with liberation of fluids as proposed already London (1986). Acknowledgement: This study was funded by the Slovak Scientific Grant Agency (VEGA 2/0084/17). Figure. 1. Tourmaline distribution within nodule on selected layer form 3D tomography (Betliar granite). Note the cavities (black) in cluster of tourmalines (bright). Matrix of albites and quartz is in grey.

REFERENCES Balen, D. and Broska, I. (2010): Tourmaline nodules: products of devolatilization within the final evolutionary stage of granite melt? Geol. Soc. London Spec. Paper, 350:53–68. Broska, I., et al. (2002): The compositions of rock-forming and accessory minerals from the Gemeric granites (Hnilec area, Gemeric Superunit, Western Carpathians). Bull. Czech Geol. Survey, 77:147–155. Buriánek, D. and Novák, M. (2007): Compositional evolution and substitutions in disseminated and nodular tourmaline from leucocratic granites: examples from the Bohemian Massif, Czech Republic. Lithos, 95:148–164. Drivenes, K. et al. (2015): Late-magmatic immiscibility during batholith formation: assessment of B isotopes and trace elements in tourmaline from the Land`s End granite, SW England. Contrib. Mineral. Petrol., 169, 56–83. London, P. (1986): Formation of tourmalinerich gem pocket in miarolitic pegmatites. Amer. Min., 71:396–405. Searle, M.P., et al. (2003): The structural geometry, metamorphic and magmatic evolution of the Everest massif, High Himalaya of Nepal–South Tibet. Journal of the Geological Society, London, 160:345– 366. Simpson, R.L., et al. (2000): Two episodes of monazite crystallization during metamorphism and crustal melting in the Everest region of the Nepalese Himalaya. Geology, 28:403–406.

TOURMALINE CLUSTERS The formation of nodular tourmaline is product of late-stage magmatic crystallization and hydrothermal fluids penetrated granites (e.g. Buriánek and Novák 2007). Balen and Broska (2012) interpreted the tourmaline nodules as a result of decompression and ascent of fluids during shallow emplacement of the granite body and during the formation of vapour and boron rich bubbles migrating through the granite body. On the other hand, Drivenes et al. (2015) suggest formation of quartztourmaline nodules in granites from immiscible hydrous borosilicate melts producing during late-stage crystallisation. The role of fluids in late-stage crystallisation and dendritic flat precipitation indicates versatility in hydrothermal processes during the formation of tourmaline. Both interpretations are probably in relation to genesis of tourmaline nodules in granites form the Gemeric unit and the Everest area. The 3D investigation of granites from Betliar area in the Gemeric unit of the Western Carpathians indicates excpet massive form also the precipitation in irregular cavities. Here, the individual idiomorphic tourmaline crystals grow within vugs after degasation of residual melts. The tourmaline forms both (1) compact, almost pure tourmaline masses and also (2) mixture of tourmaline with albite and quartz


Hydrothermal replacement of amblygonite-montebrasite by blue tourmaline – examples from the Moldanubicum, Czech Republic 1


Zbyněk Buřival#, 1Milan Novák

Department of Geological Sciences, Kotlářská 2, 611 37, Brno, Czech Republic # [email protected]

Key words: subsolidus, hydrothermal alteration, replacement, tourmaline, amblygonite


primary phosphates are replaced by a finegrained whitish material with blue rims. Microscopically, the rims are not obvious and secondary minerals usually form irregular zones (Fig. 1). The zones are typically formed by fine-grained mixture of quartz + tourmaline + secondary fluorapatite, clay minerals, and muscovite; the latter frequently mixed with crandallite-group minerals in Dolni Bory.

Amblygonite-montebrasite LiAlPO4(F,OH) is widespread in LCT pegmatites with a high content of P and lack of Ca, Fe or Mn to form other phosphates (e.g. London 2008). Mineral assemblages replacing primary amblygonitemotebrasite from LCT pegmatites Dolní Bory, Dobrá Voda and Puklice were examined to assess composition of hydrothermal fluids in granitic pegmatites. Blue tourmaline (Fe-rich fluor-elbaite > fluor-schorl) in subsolidus reaction rims around amblygonite, garnet and/or primary Fe,Mn-phosphates is quite common in complex (Li) granitic pegmatites (Buřival and Novák 2015; Němec 1981). It is along with other subsolidus tourmalines in pegmatites good indicator of early subsolidus fluid composition because its refractory behavior prevents reequilibriation in low-T conditions typical for some associated minerals. MINERAL ASSEMBLAGES AND CHEMISTRY Primary mineral assemblages in the pegmatites include K-feldspar, albite, quartz, schorl, muscovite, lepidolite, elbaite and two generations of amblygonite-montebrasite – skeletal amblygonite I from albite unit and blocky amblygonite II from lepidolite unit. Amblygonite I is completely replaced in all studied samples whereas amblygonite II is typically less altered and never replaced by tourmaline. Replacement reactions with tourmaline were typically observed in amblygonite I adjacent to Li-enriched parts of albite unit. Macroscopically, grains and aggregates of

Figure 1. BSE photos of amblygonite replacement textures from Puklice and Dolní Bory. Photo: P. Gadas.


Secondary blue tourmaline (Fe-rich fluorelbaite > fluor-schorl ) is generally Na- (0.700.95 apfu) and F-rich (0.65-0.91 apfu). Contents of Licalc. (0.21-1.05 apfu), Al (6.427.92 apfu) and Fe (up to 2.17 apfu) (Fig. 2.) are variable, amounts of Mn > Ca, Ti, and Mg are low to negligible.

DISCUSSION A complex multi-stage replacement process is apparent. In early subsolidus stage amblygonite was replaced by the assemblage tourmaline + quartz + fluorapatite + muscovite. During the low-T stage fluorapatite and muscovite were partially replaced by clay minerals plus gorceixite and goyazite whereas tourmaline-quartz assemblage is stable. Negligible to none concentrations of Mg and Ti in secondary fluor-elbaite and muscovite suggest that the system was closed to the host rocks during the first stage of hydrothermal alteration. The presence of Ba and Sr in late crandallite minerals indicates an open system during the second stage. Formation of blue tourmaline and associated muscovite and fluorapatite requires the following elements: Na, K, Li, Ca, Al, Fe, Si, B, P, F and H2O. Lithium, Al, P and F are present in the replaced amblygonite. Apparent input of B, alkalis, Ca and H2O in fluids is required. Due to absence of Fe in amblygonite, it also must be present in the fluids. Replacement of garnet by blue tourmaline (Buřival and Novák 2015) is very similar to amblygonite replacement but garnet may provide Fe, and Mn but no Li and F. Consequently, input of B, alkalis, Ca, H2O, Li, Fe (Mn) is assumed during early replacement processes in complex (Li) granitic pegmatites. REFERENCES Buřival, Z. and Novák, M. (2015): Hydrothermal replacement of garnet by tourmaline in elbaite-subtype pegmatites. 7th symposium PEG 2015 Książ, Poland, 7-8. London, D. (2008): Pegmatites. Canad. Mineral., Spec. publ., 10, 347 p. Němec, D. (1981): Ein Pegmatit mit LiMineralisierung von Dolní Bory in Westmahren (ČSSR). Chem. Erde, 40:146– 177.

Figure 2. Ternary diagrams showing chemical composition of blue tourmaline after amblygonite at X-site (A) and Y+Z sites (B). Puklice: blue rhomb, Dolní Bory: green square, Dobrá Voda: red triangle; after garnet from Ctidružice: brown full triangle (Buřival and Novák 2015).


Synthetic study of the sources and origin of the gold deposit in the western Hoggar prospect (South of Algeria) 1

Ismahane Chaouche #, 1 Mohamed Talbi, 2 Yves Fuchs


Department of Geology, FSTGAT/ USTHB. Algeria, #[email protected], [email protected] 2 Laboratoire Géomatériaux et Environnement, Institut Francilien des Sciences Appliquées (IFSA),Université Paris-Est Marne la Vallée, France, EA 4508, 5 bd Descartes, 77454, Noisy-Champs, cedex 2, France

Key words: orogenic gold, terranes, gold, hydrothermal alteration, tourmaline INTRODUCTION


In the district of Tesnou-Ifetessene gold occurrences are present in tourmaline-rich quartz veins. In order to understand the conditions of deposition of gold and the nature the composition of the involved fluids and the zonality of the hydrothermal system a peculiar attention was devoted to the variation in the composition of tourmaline.

Under microscope it is possible, based on their morphology, to distinguish two types of tourmaline. The older one consists of elongated groups of more or less deformed tourmaline crystals or strongly brecciated tourmaline rosettes. The latter appears as needles. MINERALOGY


EMP analyses of the first type and calculated structural formula allowed to classify this one as a foitite whereas the second type is a solid solution between the schorl and the dravite end members. On the basis of the results of the electronic microprobe and Mössbauer analyzes, the structural formula can be written as follows.

The gold-bearing veins of Assouf Mellène, Sedrar and Idereksi crosscut diorite and granodiorite bodies from the Affedafeda complex. They show dominant N-S, NE-SW and NWSE orientation and develop preferentially where these orientations crosscut the major 4°30’ and NE running faults. The gold bearing veins are sometimes elongated and follow the cleavage planes in the strongly catalases host rocks. In veins the main mineralogical components are quartz (92%), tourmaline (∼2%), muscovite (∼1%) and sericite (∼2%) whereas calcite remains a minor component. Sulfides are scarce and the paragenesis is limited to pyrite and chalcopyrite and associated alteration products like Fe hydroxides and covellite. Black tourmaline develops in fissures of the quartz.

(□0,27Na0,611Ca0,106K0,004)(Mg1,427Mn0,004 Fe2+1,048Feivct0,171Fe3+0,288Sb0,001Cr0,003Ti0,024 Al0,120)Al6Si6O18(BO3)3(OH)3(OH) 0,992F0, 008 Gold is present as native gold, disseminated in quartz and tourmaline. It fills micro-fissures between the different grains but it is mainly present as coating of the tourmaline grains where it can also be present as microinclusions. It can also be found in the alteration aureoles of chalcopyrite and pyrite.



Baksheev, I.A., et al. (2011): Ferric-iron-rich tourmaline from the Darasun gold deposit, Transbaikalia, Russia. Canad. Mineral., 49: 263–276. Beziat, D., et al. (1999): The Guibaré and Fété Kolé Gold-bearing tourmaline-quartz veins in the Birrimian greenstone Belts of Burkina Fasso. Canad. Mineral., 37, 575–591. Bosi, F. and Lucchesi, S. (2004): Crystal chemistry of the schorl-dravite series. Eur. J. Min., 16:335–344. Dyar, M.D., et al. (1998): Inclusive chemical characterization of tourmaline: Mössbauer study of Fe valence and site occupancy. Am. Mineral., 83:848–864. Ferrow, E.A., et al. (1988): Mössbauer effect study on the mixed valence state of iron in tourmaline. Min. Mag., 52:221–228. Fuchs, Y. (1989): Hydrothermal alteration at the Novazza volcanic field and its relation to the U-Mo-Zn Novazza deposit, northern Italy. In: Panel Proceedings Series, Metallogenesis of uranium deposits: Vienna, International Atomic Energy Agency: 137– 151. Paulet, P.H., et al. (1991): Zonalité des différents types de borosilicates dans les système hydrothermal à or-argent de la Humboldt Range, Pershing County, Nevada, USA. C.R. Acad. Sci., 313, 10:1155–1162.

Chemical analyses performed by SEM and EMP technique, structural studies by Mössbauer technique (MES) and FTIR allow to precise the distribution of the different types of tourmaline on the basis of their composition and structure. A part of the tourmaline of the gold occurrence of Seldrar show a composition of foitite, that, following Y. Fuchs and R. Maury (1995) would characterize the central spot of the hydrothermal system (T=250°330°C; Paulet et al. 1991) whereas the tourmaline of the Assouf Mellene, Idereksi and an area of Seldrar have a higher Fe3+ content reflecting the more oxidizing condition prevailing in the outer zone of the hydrothermal system with lower temperatures (T= 160-180°C). The presence of gold seems to be related to a late phase of the hydrothermal process and is associated either to the 2nd stage of tourmalinization or to the sulfides. REFERENCES Andreozzi, G.B., et al. (2008): Linking Mössbauer and structural parameters in elbaite-schorl-dravite tourmalines. Am. Mineral., 93:658–666.


Hydrothermal tourmaline from the Panasqueira W-Sn-Cu deposit, Portugal: chemical and B-isotope constraints on fluid source and evolution 1

Marta S. Codeço#, 1 Robert B. Trumbull, 1Philipp Weis, 2 Filipe Pinto, 3 Pilar Lecumberri-Sanchez


GFZ German Research Centre for Geosciences, Telegrafenberg, 14473 Potsdam, Germany, [email protected] 2 Beralt Tin & Wolfram (Portugal) S.A., Barroca Grande, 6225-051, Aldeia de S. Francisco de Assis, Castelo Branco, Portugal 3 Department of Geosciences, University of Arizona, 1040 E 4th Street, Tucson, Arizona 85721, USA #

Key words: Tourmaline, Boron isotopes, SIMS, ore deposits, Panasqueira, magmatichydrothermal systems



The Panasqueira deposit consists of a swarm of sub-horizontal, hydrothermal W-Sn-Cubearing quartz veins hosted by metasediments (Fig. 1) and is spatially associated with a greisen cupola of a peraluminous, S-type, twomica granite (Kelly and Rye 1979). Tourmaline is a ubiquitous mineral in wallrock alteration zones that predate the main ore stage, and it also occurs in late-stage vugs within the ore veins. We report results from in-situ microprobe and SIMS analyses showing the variations in chemical and boron isotopic compositions of tourmaline from different settings within the Panasqueira deposit.

Hydrothermal tourmaline from Panasqueira is Fe-Mg-Na-Al-rich and belongs to the schorldravite series with up to 0.6 apfu Al in octahedral sites. Tourmaline crystals display optical and chemical zoning, with Fe-F-Na increase and Mg-Al-Ca decrease from core to rim. F contents are inversely correlated with octahedral Al and X-site vacancies. These variations are consistent throughout the deposit and from the main-stage (Fig. 2 a-c) to latestage tourmaline (Fig. 2 d-f).

Figure 1. Schematic cross section of the Panasqueira vein system, greisen and Panasqueira granite (modified after Oosterom et al. 1984). L – Level.


Figure 2. Chemical compositions of tourmaline from the wallrocks (a,b,c; proximal (crosses) and distal (squares) to the vein) and late-vug fillings (c,d,f) showing Fe-Mg (a,d), F-Al (b,e) and Fe-Mg-Na (c,f) variations.

Peru. Fluid inclusion studies of vein quartz show a temperature range of 230° to 360° C throughout the deposit (Kelly and Rye 1979). We suggest that wallrock alteration and tourmaline formation likely began at a higher temperature than recorded in the quartz veins. If one accepts the model of a magmatic fluid source, the early-stage fluids likely had temperatures of at least 500 °C. Cooling can also explain the offset of some tourmaline to lower 11B values (down to -13 ‰). Hydrothermal tourmaline is isotopically lighter than the source fluid by a temperaturedependent factor, so cooling results in progressively lower 11B values. Assuming that tourmaline started to crystallize at 500550° C, a cooling of ca. 300° C (the low end of fluid inclusion temperatures) produce a shift of about 3 ‰ in 11B (Meyer et al. 2008), which is consistent with the observed range. The presence of 11B values higher than the main range of -9 ± 2 ‰ (up to -4 ‰) requires a different mechanism than cooling. In a setting with a limited fluid B-reservoir (high rock/fluid ratio), crystallization of tourmaline progressively depletes the fluid in 10B so that later tourmaline is isotopically heavier than early growth. This process is logically most effective farthest from the veins where the fluid/rock ratio is lowest, and that may explain the weak but systematic correlation of 11B values in tourmaline according to distance from the vein contact (Fig. 3).


The Panasqueira tourmaline has a B-isotope composition ranging from -4 to -13 ‰ but over 90% of the analyses fall between -7 and -11 ‰ (mean = 9.1 ‰, n= 247, see Fig. 3). There is no significant B-isotope zoning nor any variations according to host-rock type or relative age of tourmaline in the deposit. However, there is a slight and systematic trend to lower 11B values in tourmaline from nearer to the vein contact (Fig. 3). Although the B-isotopic compositions are compatible with either a metasedimentary or granitic source of boron, mass-balance considerations make an origin from the S-type granite more likely (Fig. 3). FLUID-ROCK INTERACTION AND COOLING: IMPLICATIONS FOR FLUID EVOLUTION

The variations in tourmaline chemistry and 11B values can be explained by a combination of two processes: fluid-rock interaction and cooling of the hydrothermal fluid. The deposit-wide chemical zonation pattern, with Na and Fe/(Fe+Mg) increasing from core to rim, may reflect cooling. The partitioning of Na in tourmaline increases with decreasing temperature (von Goerne et al. 2001). Mlynarczyk and Williams-Jones (2006) invoked cooling to explain the progressive increase in Fe/(Fe+Mg) ratio in hydrothermal tourmaline from the San Rafael Sn-Cu deposit,


We suggest that tourmaline in the wallrocks first formed under rock-dominated conditions where replacement of chlorite, biotite and feldspar in the host schists contributed Al, Fe, Mg and Ca for tourmaline core growth. The progression to a more fluid-dominated system coupled with cooling contributed to the Fe- Frich rim formation.

REFERENCES Kelly, W.C. and Rye, R.O. (1979): Geologic, fluid inclusion, and stable isotope studies of the tin-tungsten deposits of Panasqueira, Portugal. Econ. Geol., 74:1721–1822. Meyer, C., et al. (2008): Boron-isotope fractionation between tourmaline and fluid: an experimental re-investigation. Contr. Mineral. Petrol., 156:259–267. Mlynarczyk, M.S.J. and Williams-Jones, A.E. (2006): Zoned tourmaline associated with cassiterite: implications for fluid evolution and tin mineralization in the San Rafael SnCu deposit, southeastern Peru. Canad. Mineral., 44:347–365. Oosterom, M.G. ; et al. (1984): Lithogeochemical studies of aureoles around the Panasqueira tin-tungsten deposit, Portugal. Miner. Deposita, 19:283–288. Van Hinsberg, V.J. et al. (2011): Tourmaline: An ideal indicator of its host environment. Canad. Mineral., 49:1–16. Von Goerne, G., et al. (2001): Synthesis of tourmaline solid solutions in the system Na2O-MgO-Al2O3-SiO2-B2O3-H2O-HCl and the distribution of Na between tourmaline and the fluid at 300 to 700° C and 200 MPa. Contr. Mineral. Petrol., 141:160–173.

Figure 3. Box and whiskers plot for the entire set of tourmaline data from Panasqueira and the comparison of δ11B values measured in tourmaline samples from Panasqueira with other W-Sn deposits and related settings from the Central Iberian Zone (modified after van Hinsberg et al. 2011). L – Level; SS – spotted schist; FGS – fine-grained schist; D and P – indicate analyses distal and proximal to the vein, respectively.


Simultaneous formation of fluor-schorl and fluor-elbaite in the Kracovice pegmatite, an effect of Fe and Mg on tourmaline stability 1

Renata Čopjaková#, 1Radek Škoda, 2.3Michaela Vašinová Galiová, 1 Milan Novák


Department of Geological Sciences, Masaryk University, Brno, Czech Republic, # copj[email protected] Department of Chemistry, Masaryk University, Brno, Czech Republic 3 Central European Institute of Technology (CEITEC), Masaryk University, Brno, Czech Republic 2

Key words: fluor-schorl, fluor-elbaite, tourmaline stability, Kracovice pegmatite


graphic, blocky K-feldspar and albite units or it overgrows together with Li-Al mica partlycorroded garnet in the albite unit. Tur2 corresponds to Li-rich fluor-schorl to fluorelbaite (0.1-1.1 wt. % Li, high Al 6.96-7.43 apfu, variable Fe 0.29-1.82 apfu, Mn enrichment 0.18-0.76 apfu, commonly very low Mg < 0.05 apfu and Ti < 0.03 apfu, X-site dominated by Na 0.71-0.99 apfu, high F 0.67– 1.00 apfu). Contents of Al, Mn, Na, F and Li increase and Fe decrease toward rims (for more details see Čopjaková et al. 2015). Only tourmaline grains (Tur2* in Fig. 1) close to the tourmalinized Li-Fe mica (see next chapter) shows higher Mg (0.03-0.45 apfu) and Ti contents (0.03-0.09 apfu). Crystallization of Tur2 may hit transition from late magmatic to hydrothermal conditions and it crystallized from highly evolved, F,Li-rich melt and fluid.

Kracovice pegmatite represents the most evolved pegmatite body of NYF pegmatites related to the Třebíč Pluton. The pegmatite consists of: a granitic unit (Kfs + Pl + Qz + Bt + Ttn), a graphic unit (Kfs + Qz + Tu + Grt + Li-Fe mica + Ms), a blocky K-feldspar unit (Kfs + Qz + Ab + Tu + Grt + Ms), and an albite unit (Ab + Qz + Tu + Grt + Li-Al mica + Ms + topaz) close to a small quartz core. Several primary and secondary tourmaline types were distinguished based on the textural relations and chemical composition. They were studied by combination of EPMA and LA-ICPMS methods. PRIMARY MAGMATIC SCHORL Black tourmaline Tur1 forms prismatic crystals or graphic intergrowths with quartz and garnet. It occurs in all units except of the granitic unit and corresponds to Al-rich schorl to fluorschorl (6.57-7.12 apfu Al; 1.89-2.23 apfu Fe; low Mg ˂ 0.09 apfu; Fe/(Fe + Mg) ~ 0.98, low Mn 0.07-0.15 apfu, Ti < 0.03 apfu, Li 80-500 ppm, and moderate Na 0.49-0.73 apfu, X-site vacancy 0.23-0.50 pfu and F 0.18-0.68 apfu; Fig. 1). Tur1 crystallized early from B,F-rich melt (Novák et al. 2012, Čopjaková et al. 2013, 2015).

TOURMALINIZATION OF Li-Fe MICA Tourmalinization of dark mica of Li-Fe series (2.86-3.45 apfu Si, 0.64-1.18 apfu VIAl, 0.331.26 apfu Fe, 0.03-0.72 apfu Mg, Fe/(Fe+Mg) = 0.46-0.98, 0.12-1.36 apfu Li, 0.43-1.87 apfu F) was observed in graphic unit, close to the albite unit. A fine-grade mixture of secondary tourmaline TurSMc and quartz replaces flakes of dark mica from the rim inwards. Locally, the intensive tourmalinization resulted in a full replacement of original mica by a rather homogeneous tourmaline with tiny quartz inclusions. In this case, TurSMc preserves original textural features of dark mica, a flakelike shape and oriented ilmenite-pyrophanite inclusions (Fig. 2). TurSMc corresponds to F-

Li-RICH TOURMALINE Later, light brown to greenish tourmaline Tur2 forms tiny prismatic crystals, common in the albite and blocky K-feldspar units. Moreover, it overgrows and partly replaces Tur1 in the


Figure. 2. BSE image of tourmaline TurSMc after dark mica with tiny quarz and ilmenite-pyrophanite inclusions.

available throughout their early magmatic crystallization (from granitic to the albite unit). Dark mica represents the sole primary carrier of Mg in the pegmatite. Both later F,Na-rich tourmaline generations, fluor-elbaite Tur2 and fluor-schorl TurSMc crystallized simultaneously at the presence of the highly evolved Li,F-rich, alkaline, B-saturated melt/fluid. Availability of Fe + Mg during the replacement of Li-rich dark mica seems to be the main factor affecting formation of fluor-schorl instead of fluorelbaite. Li-rich Tur2 formed from Li-rich source only at low activity of Fe + Mg and the availability of Fe + Mg from dark mica stabilized Li-poor fluor-schorl (TurSMc) composition.

Figure. 1. Chemical composition of tourmaline; occupations of the X-site and Y+Z sites.

schorl (5.82-6.78 apfu Al, high Fe 1.27-2.58 apfu, variable Mg 0.12-1.06 apfu, Fetot/(Fetot + Mg) = 0.59-0.94, minor Mn 0.02-0.23 apfu and Ti 0.03-0.19 apfu, low Li 270-2200 ppm, high Na 0.66-0.93 apfu and F 0.55-0.89 apfu; Fig. 1). Its chemistry is driven by composition of source dark mica and variability in Mg, Fe, Ti reflects their contents in relict mica. Textural and chemical features indicate tourmalinization of primary dark mica occurred during its interaction with the highly evolved Li,F,Narich, B-saturated melt simultaneously with crystallization of columnar fluor-elbaite Tur2 in the albite unit.

REFERENCES Čopjaková, R., et al. (2013): Distributions of Y + REE and Sc in tourmaline and their implications for the melt evolution; examples from NYF pegmatites of the Třebíč Pluton, Moldanubian Zone, Czech Republic. J. Geosci., 58:113–131. Čopjaková, R., et al. (2015): Scandium- and REE-rich tourmaline replaced by Sc-rich REE-bearing epidote-group minerals from the mixed (NYF + LCT) Kracovice pegmatite (Moldanubian Zone, Czech Republic). Am. Mineral., 100:1434–1451. Novák, M., et al. (2012): Contrasting origins of the mixed signature in granitic pegmatites; examples from the Moldanubian Zone, Czech Republic. Canad. Mineral., 50:1077–1094.


Crystallization of magmatic schorl Tur1 was terminated probably by melt depletion in Fe or alternatively by decreasing content of B in the melt. Magmatic Tur1 together with garnet and minor dark mica carry almost all Fe and it was


Tourmaline compositions: Reflecting the fluid phase 1


Barbara L. Dutrow#, 1Darrell J. Henry

Dept. Geology & Geophysics, Louisiana State University, Baton Rouge, LA 70803 USA, # [email protected]

Key words: tourmaline, fiber, fluids, pegmatite, salinity, hydrothermal


Similarly, tourmaline species may reflect components of the fluids phase, both in the cation and anion sites. For example, tourmaline compositions from oxy-dravite to povondraite (O-P) are described by the FeAl-1 exchange. These Al-deficient depronotated tourmaline compositions are commonly associated with saline environments with the Fe3+ implying oxidizing fluids. Tourmalines of the O-P trend are found in metamorphosed caps of salt domes (Henry et al. 1999), in active salt domes, in active geothermal systems (Lynch and Ortega 1997), and in Cu-Mo-Au ore deposits (e.g. Cavaretta and Puxeddu 1990). The occurrence of O-P tourmalines are consistent with high salinity fluids and may be an indicator of such fluids in the geologic past. Flourine incorporation into tourmalines is associated with interaction of flourine-bearing aqueous fluids. However, the incorporation of F into the tourmaline structure has crystallochemical constraints and is dependent on the X-site charge (Henry and Dutrow 2011). When X-site vacancies are > 0.6 pfu, a maximum amount of 0.1 F pfu is permitted into the structure. Consequently, fluor-species are found in alkali and calcic group tourmalines but are absent in vacancydominated tourmalines. Tourmaline-fluid partitioning experiments can be used to approximate cation compositions of the fluid phase (e.g. von Goerne 2001, 2011; van Hinsberg et al. 2017). Refitting partitioning data (von Goerne 2001, 2011) permits expressions to be obtained relating X-site cationic occupancy to Na and Ca contents in aqueous fluids in local equilibrium with tourmaline + albite + quartz for conditions near 300° C and 0.2 GPa.

Deciphering the geologic conditions of terranes relies on extracting signatures of geologic processes embedded in minerals. The tourmaline supergroup minerals, while accessory phases in most rocks, have the ability to retain and reflect not only the composition of the host rock(s) in which they form, but also the fluid phase with which they interact. In some geo/hydrothermal settings, tourmalines may form in a fluid-rich systém that is not rock buffered. Recent work suggests that compositions of tourmaline can be indicators of the fluid-phase composition and in some cases, provide quantitative estimates of ion activities (e.g. Dutrow and Henry 2016; van Hinsberg et al. 2017). TOURMALINE AND FLUIDS Tourmaline forms when boron becomes available to the system. Such boron can be internally derived from the breakdown of other B-bearing minerals or through increased concentration in the melt, or it can be an external fluid that infiltrates the system. In both cases, tourmaline preserves signatures of the fluid event. These signatures are in the form of overgrowths on preexisiting tourmaline, dissolution of prexisting tourmaline and reprecipitation of a new tourmaline composition (e.g. Dutrow and Henry 2000, 2016), or more subtly such as tourmaline pseudomorphing another phase (e.g. Dutrow et al. 1999). Fluids which stabilize tourmaline are neutral to acidic whereas alkaline fluids result in tourmaline dissolution (Henry and Dutrow 1996)



an additional method to decipher the evolution of the hydrothermal environments, particularly those associated with a dynamic, evolving fluid phase that has escaped the systém.

Fibrous tourmalines are particularly responsive to changing conditions in fluid-rich hydrothermal environments. They commonly form late in the petrogenetic sequence of a geologic environment where there is open fluid-filled space e.g. at the end of the tourmaline crystallization sequence in a pegmatite pocket or in a hydrous fluid-rich fracture system. Using these derived expressions suggest that fibrous tourmalines equilibrated with aqueous fluids having Na concentrations from 0.07 to 0.48 mol/l Na and with generally low Ca concentrations ( 0.35 mol/l Na. Most fibrous tourmalines are chemically zoned and the zoning tracks the chemical evolution trends in the host environment (Fig.1). For the case of Cruzeiro, Brazil fibrous tourmaline with three distinct chemical zones, tourmaline compositions evolve from foitite to schorl to elbaite/fluor-elbaite (Dutrow and Henry 2016). Here, zoning in a single fiber mirrors the overall fractionation trend in the pegmatite with increases in Na, Li and F, a feature of self-similarity.


Figure. 1. Compositional variations of three generations of tourmaline developed within individual fibers. Variations from foitite to schorl to elbaite/ fluorelbaite mirrors the fractionation trend of a pegmatite (Cruzeiro mine, Brazil; data from Dutrow and Henry 2000, 2016).

CONCLUSIONS Tourmaline is a robust indicator of its host rock and fluid environment. The compositions of tourmaline and fibrous tourmaline provide


Cavarretta, G. and Puxeddu, M. (1990): Schorl–dravite–ferridravite tourmalines deposited by hydrothermal magmatic fluids during early evolution of the Larderello Geothermal Field, Italy. Econ. Geol., 85: 1236–1251. Dutrow, B.L. and Henry, D.J. (2000): Complexly zoned fibrous tourmaline, Cruzeiro mine, Minas Gerais, Brazil: A record of evolving magmatic and hydrothermal fluids. Canad. Mineral,. 38:131–143. Dutrow, B.L. and Henry, D.J. (2016): Fibrous Tourmaline: A sensitive probe of fliud composiitons and petrologic environments. Canad. Mineral., 54:311–335. Dutrow, B.L., et al. (1999): Tourmaline-rich pseudomorphs in sillimanite zone metapelites: Demarcation of an infiltration front. Am. Mineral., 84:794–805. Henry, D.J. and Dutrow, B.L. (1996): Metamorphic tourmaline and its petrologic applications. in Boron (Grew, E. S. and Anovitz, L. M., eds.) RIM 33:503–557. Henry, D.J. and Dutrow, B.L. (2011): The incorporation of fluorine in tourmaline: Internal crystallographic controls or external environmental influences? Canad. Mineral., 49:41–56. Lynch, G. and Ortega, S. (1997): Hydrothermal alteration and tourmaline–albite equilibria at the Coxheath porphyry Cu–Mo– Au deposit, Nova Scotia. Canad. Mineral., 35:79–94. van Hinsberg, V.J., et al. (2017): Determining subduction-zone fluid composition using a tourmaline mineral probe. Geochem. Persp. Let., 3:160–169. von Goerne, G., et al. (2001) Synthesis of tourmaline solid solutions in the system Na2O-MgO-Al2O3-SiO2-B2O3-H2O-HCl and the distribution of Na between tourmaline and fluid at 300 to 700 °C and 200 MPa. Contrib. Mineral. Petrol., 141:160–173. von Goerne, G., et al. V.J. (2011): Experimental determination of Na-Ca distribution between tourmaline and fluid in the system CaO-Na2O-MgO-Al2O3-SiO2B2O3-H2O. Canad. Mineral., 49:137–152

Tetrahedral substitutions in tourmaline and their applications in petrology 1,2

1 2

Andreas Ertl#, 2Ekkehart Tillmanns

Mineralogisch‑Petrographische Abt., Naturhistorisches Museum, 1010 Vienna, Austria, # [email protected] Institut für Mineralogie und Kristallographie, Geozentrum, Universität Wien, 1090 Vienna, Austria

Key words: tourmaline, mineralogy, crystallography, tetrahedral site, petrology, boron, aluminum range 550-900 °C/1-5 GPa. Similar tourmalines were synthesized later by different groups. A pronounced negative correlation (r2 = 1.0) between temperature during crystal growth (at constant pressure) and [4]B (from refinement) in synthetic Al-rich tourmalines, which were synthesized by London (2011), was found by Ertl et al. (2012). When plotting the relationship between [4]B (from structure refinements) in Al- and [4]B-rich tourmaline and the unit cell volume, a pronounced negative correlation can be observed (Fig. 1). An extrapolation suggests for an Al-rich tourmaline with 3 apfu [4]B a cell volume of

INTRODUCTION Tourmaline is an extremely important mineral because of its relevance in the geosciences. The tourmaline group consists of more than a dozen end members, which occur in many different geological environments (Henry et al. 2011). Tourmalines are complex aluminumborosilicates with strongly varying compositions because of isomorphous replacements. The tourmaline mineral group is chemically one of the most complicated groups of silicate minerals, with the general formula XY3Z6[T6O18](BO3)3V3W. The structure is characterized by six-membered tetrahedral rings (T sites), whose apical oxygens point toward the (-) c-pole, producing the acentric nature of the structure. The T sites are usually occupied by Si and sometimes, especially in high-grade metamorphic rocks, by significant amounts of Al (e.g., Cempírek et al. 2006). BORON AT THE T SITE Within the last 20 years a significant substitution of Si by B was found in natural and synthetic tourmalines (Ertl et al. 1997, 2005, 2006, 2007, 2008, 2012, 2015; Hughes et al. 2004; Schreyer et al. 2002). It is still not easy to verify the presence of B at the T site. Especially just by using only microprobe data, as it is usually done, tetrahedrally coordinated B can be overlooked. The incorporation of B at the T site in tourmaline can be important for petrology because it indicates lower temperatures during crystallization (Schreyer et al. 2000, Ertl et al. 2008). The first description of synthetic Al-rich tourmaline with verified [4]B was by Schreyer et al. (2000). They successfully synthesized tourmaline in the

Figure 1. Relationship between tetrahedrally coordinated boron (from structure refinements) in Aland [4]B-rich tourmaline and the unit cell volume. Note: Natural and synthetic Al- and [4]B-rich tourmaline samples (refined [4]B 0.4 apfu; FeO + MnO 4 GPa metamorphic conditions), minimal volume diffusion results in retention of the imprinted chemical signature (ibid). Tourmaline is also hard, chemically resistant, and persists in clastic sedimentary environments. The utility of tourmaline as a provenance indicator was earlier demonstrated using compositions of tourmaline as a function of their rock type when plotted on an Al-Mg-Fe triangular diagram (Henry and Guidotti 1985). The diagram correlates and predicts the lithology in which a given tourmaline composition crystallized because it was found that tourmaline is in chemical equilibrium with its host rocks. This diagram has been used in a wide range of studies, including source region for metasediments (Henry and Dutrow 1992), the source of sands in modern and ancient drainage systems (e.g. Viator 2003; Morton et al. 2002), and the source of hydrothermal fluids (e.g. Trumbull et al. 2011; Hazarika et al. 2005; Nascimento et al. 2007). Consequently, tourmaline is an excellent provenance indicator, but could be more

METHODS A total of 189 crystals from 58 samples representing five lithologic categories form the preliminary dataset. The samples are wellcharacterized in terms of the lithology of formataion and include tourmaline from pelitic metamorphic rocks, Li-rich pegmatites, calcareous metamorphic rocks, hydrothermal deposits, and Li-poor pegmatites (grouped with two samples from silicic igneous rocks). The tourmalines were analyzed using an Ocean Optics 2500+ LIBS system with a Nd-YAG laser operated at 1064 nm wavelength and 63 mJ energy per pulse. Between 50 and 200 spectra were collected for each crystal, depending on the size of analyzable surfaces. The spectra were averaged to produce one spectrum per crystal. The chemometric technique, Partial Least Squares Regression (PLSR), was used to create a matching algorithm that sequentially identifies tourmaline spectra from the five lithologic categories (Fig. 1.). Each PLSR model classifies spectra into two groups: the lithologic category being distinguished and the group of all other lithologic categories. After a category has been distinguished, it is removed model distinguishes between two categories.


spectra contain information on every element, fully utilizing the complex compositional variations in tourmaline. REFERENCES Hazarika, P., et al. (2015): Diverse tourmaline compositions from orogenic gold deposits in the Hutti-Maski Greenstone Belt, India: Implications for sources of ore-forming fluids. Econ. Geol., 110:337–353. Henry, D.J., and Dutrow, B.L. (1992): Tourmaline in a low-grade clastic metasedimentary rock - an example of the petrogenetic potential of tourmaline. Contri, Mineral. Petrol., 112:203–218. Henry, D.J. and Dutrow, B.L. (1996): Metamorphic tourmaline and its petrologic applications. in Boron (Grew, E. S. and Anovitz, L. M., eds.) RIM 33:503–557. Henry, D.J. and Guidotti, C. V. (1985): Tourmaline in the staurolite grade metapelites of NW Maine: a petrogenetic indicator mineral. Am. Mineral., 70:1–15. Hubert, J.F. (1962): A zircon–tourmaline–rutile maturity index and the interdependence of the composition of heavy mineral assemblages with the gross composition and texture of sandstones. J. Sed. Petrol., 32:440–450. Krynine, P.D. (1946): The tourmaline group in sediments. J. Geology, 54: 65–87. Morton, A., et al. (2002): Correlation of reservoir sandstones using quantitative heavy mineral analysis. Petroleum Geosc., 8:251– 262. Nascimento, M.D., et al. (2007): Provenance of Albian sandstones in the Sao Luis-Grajau Basin (northern Brazil) from evidence of PbPb zircon ages, mineral chemistry of tourmaline and palaeocurrent data. Sed. Geology, 201:21–42. Trumbull, R.B., et al.. (2011): Fluid sources and metallogenesis in the Blackbird Co-Cu-Au-BiY-REE district, Idaho, USA: Insights from major-element and boron isotopic compositions of tourmaline. Canad. Mineral., 49:225–244. van Hinsberg, V.J. et al. (2011): Tourmaline as a petrologic forensic mineral: A unique recorder of its geologic past. Elements, 7:327–332. Viator, D.B. (2003): Detrital tourmaline as an indicator of provenance: A chemical and sedimentological study of modern sands from the Black Hills, South Dakota. MS Thesis, Louisiana State University, 130 pp.

Figure 1. Matching algorithm for determination of lithology in which tourmaline crystallized. Each model classifies spectra into one of two groups: the category of interest and the group consisting of all other categories. Success rates are given as percentage of correctly classified validation spectra.

from subsequent models. As a result, the finalLIBS spectra from half of the tourmalines were used to calibrate the models. The remaining half of the spectra were used to validate the models; success rates are based on the percent of correctly classified validation spectra. RESULTS The matching algorithm has an overall sucess rate of 96.3%, correctly classifying 288 of 299 spectra. Invidicual models have success rates ranging between 92.2% and 99.0% (Fig. 1). Success rates increase as the number of samples increases. DISCUSSION AND CONCLUSIONS Based on these prelimary data, this tourmaline analysis method can be applied to sediment provenance, tectonic reconstruction, and mineral exploration problems. This method has two advantages: (1) the ease of analysis and lack of sample preparation that allows projects to use hundreds of samples, and (2) LIBS


Tourmaline crystal growth in gem-bearing pegmatites 1


Federico Pezzotta#

Natural History Museum, Corso Venezia 55, 20121 Milano, Italy, #[email protected]

Key words: tourmaline supergroup, crystal growth, gem-bearing pegmatites aggregates, or (b) “comb” structures composed by tourmaline single crystals or graphic aggregates with quartz. Tourmaline “Comb” textures, which are typical of the intermediate zones of the most fractionated pegmatites, belonging to the experience of the author systematically grow from the footwall toward the core of the vein, strongly contributing to the textural and compositional asymmetry of the zoned gem-LCT pegs. This textural feature evidences that the gravity-vector significantly influences the geochemical differentiation of the gem-LCT pegs bodies. Core zones and miarolitic cavities: large tourmaline crystals (or graphic aggregates with quartz) grown in the intermediate zones can project into the core zone up to the massive quartz core and the cavities, forming large and strongly zoned crystals. Tourmaline compositions can develop from “primitive” to “very evolved” and the represented species are schorl, fluor-schorl, foitite, elbaite, fluorelbaite, fluor-liddicoatite, tsilaisite, fluortsilaisite, and rossmanite. A further, compositionally high-evolved, late generation tourmaline can occur forming crystals of small size (“pencils”) dispersed in massive quartz, or in cavities.

INTRODUCTION Tourmaline supergroup minerals are present as both rock forming minerals and accessory minerals in gem-bearing pegmatites belonging to the LCT and the mixed LCT-NYF families (indicated as gem-LCT pegs in the following text). Such minerals typically crystallize in multiple generations and with different textures and compositions across the following pegmatitic units: endo-contact reaction zones; border and intermediate zones; core zones. The varieties of gem-LCT pegs compositions, zonation, structures and crystallization histories, combined with the exceptional crystallochemical features of tourmaline, are responsible for the occurrence of several species such as: dravite, fluor-dravite, uvite, feruvite, schorl, fluor-schorl, foitite, elbaite, fluor-elbaite, fluor-liddicoatite, tsilaisite, fluortsilaisite, and rossmanite. In most cases two or more species can be present in a single zoned crystal. TOURMALINE TEXTURES Pegmatite border reaction zones (or “inner” reaction zones) – B-rich Pegmatitic medium can react with the host rocks with massive crystallization of tourmaline characterized by significant contents of FeMg-Ti-Ca (dravite, fluor-dravite, uvite, feruvite, schorl, fluor-schorl, foitite). Textures include granular aggregates and radiating groups of elongated blackish prismatic crystals inside quartz, feldspars, and micas, with more or less apatite, projecting into the pegmatite and rimming host rock enclaves. Intermediate zones - Coarsening of the pegmatitic grain size accompanies the formation of: (a) dispersed tapering tourmaline single crystals or quartz-tourmaline graphic


Most of gem-LCT pegs are characterized by tourmaline crystals (or graphic aggregates with quartz), in the contact zones and in the intermediate zones, quiet exclusively grown in the direction of the analogous pole (-c), or in alternative in the direction of the antilogous pole (+c). This phenomena is particularly evident when crystals and crystals aggregates grew up to projecting inside cavities, forming


drusy surfaces. This situation is in general persistent with entire pegmatite fields. “Analogous pole pegmatites” (-cP) numerically dominate on the “antilogous pole pegmatites” (+cP). Significant examples of -cP are in California and Elba Island (Italy). Many examples of +cP are in Madagascar, and in Malkhan, Russia. A more rare group of Gem-LCT pegs is characterized, in the border zones and the intermediate zones, of mixed occurrences of tourmaline crystals grown in the direction of both the analogous or the antilogous pole. A representative example of these “mixed” pegmatites (cP) occurs in the famous locality of Paprok, Nooristan, Afghanistan. Depending on the available documentation, the renewed Jonas pegmatite in Governador Valadares, Minas Gerais, Brazil, probably belongs to the same group. It is important to notice that the latest generation of high geochemically evolved tourmaline crystals in the core zone (the “pencils”) is not correlated with the growth direction of the tourmaline crystals of previous generations in the pegmatite. Such “pencils” typically grow toward both –c and +c directions, with a strong preference at the very latest stages for the +c direction.

(3) Boundary-layer mechanism during crystal faces growth in the pegmatitic medium – Complex rhythmic colors variations are very likely the result of boundary-layer mechanism during crystallization, related to imperfect elements diffusion in the pegmatitic medium at the fronts of crystals growth (London 2008). (4) External disturbs – These phenomena, such as partial re-opening of the geochemical system related to surrounding-rock structural instability, are responsible for introducing geochemical disturbs in the crystallizing medium. Clear examples occur in Elba Island pegmatites where the typical “black cap” overgrowth of tourmaline crystals, immediately follow in the paragenesis the pocket rupture at the latest stages of crystallization of the cavities. DISCUSSION AND CONCLUSIONS Tourmaline crystals in “Comb” textures, which typically grow from the footwall toward the core of the pegmatitic bodies, evidence a major role of the gravity vector in the geochemical segregation of the pegmatitic units. Analogous versus Antilogous crystals growth is very likely associated with significant geochemical or physical factors still to be investigated. The variety of the patterns of color-zoning of tourmaline crystals from different localities is explained by the geochemical fractionation during crystallization, including paragenetic effects, more or less in combination with other phenomena such as the different growth vectors and energy of the crystal forms, “boundary layer” mechanism during crystallization, and external disturbs of the geochemical environment. This study once again evidences how crystals of tourmaline-group minerals exceptionally recorded the many events and phenomena occurred during gem-LCT pegs formation.

CRYSTALS COLOR ZONATION Color zonation in tourmaline crystals occur in response of crystallochemical variations during disequilibrium crystal growth. This phenomena is related to the following factors: (1) Geochemical fractionation during crystallization, including paragenetic effects – This process is responsible for a “linear” geochemical evolution; from primitive species rich in Mg, Fe, Ti, to high evolved species rich in Li, with Mn, Al and vacancies playing various roles. From dark colored varieties, crystals grow to greenish-bluish varieties up to colorless and pinkish and red. “Watermelon” pattern could be related to paragenetic effects. (2) Different growth vectors and surface energy of the crystal forms – Prisms, and pedion and pyramidal forms at the analogous and antilogous poles are characterized by strongly different growth vectors and surface energies, producing complex color-zoning patterns, in part described in Van Hinsberg et al. (2011) as “hourglass” sectors zoning.

REFERENCES London, D. (2008): Pegmatites. Canad. Mineral., Spec. publication 10, 347 p. van Hinsberg V.J., et al. (2011) Tourmaline: an ideal indicator of its host environment. Canad. Mineral., 49:1–16.


Bosiite or not bosiite? That is the question… 1


Adam Pieczka#, 1Mateusz Sęk

Department of Mineralogy, Petrography and Geochemistry; AGH University of Science and Technology, 30059 Cracow, Mickiewicza 30, Poland, #[email protected]

Key words: bosiite, tourmaline supergroup, Julianna pegmatitic system, Poland


bearing tourmaline. It is associated with Libearing annite (Li calculated on the basis of mica group stoichiometry), Cs-dominant annite and a Fe-bearing chlorite-group mineral. The tourmaline probably formed due to late hydrothermal alteration of a primary dark mica by a (B, Li, Rb, Cs)-enriched alkaline fluid.

Bosiite, NaFe3+3(Al4Mg2)(Si6O18)(BO3)3(OH)3 O, is one of the youngest minerals in the tourmaline supergroup. The mineral was firstly found in the Darasun gold deposit, VershinoDarasunskiy, Transbaikal Krai, EasternSiberian Region, Russia (Ertl et al. 2015; 2016), in the form of compositionally heterogeneous crystals intimately associated with oxy-dravitic and dravitic tourmalines. A new occurrence of bosiite was briefly reported from the Řečice pegmatite, Czech Republic (Flégr et al. 2016); however this bosiite, secondary after annite and highly heterogeneous, has low Mg (0.60–1.12 apfu) and high Fetot (3.07–4.15 apfu). The compositions of the bosiite were explained as a solid solution between end-members bosiite, schorl and povondraite. A very similar in composition tourmaline was found in the Julianna pegmatitic system at Piława Górna, the Góry Sowie Block, Lower Silesia, SW Poland. It is relatively common in the graphic unit of highly-evolved LCT-type pegmatite, at the border with the zone of metasomatic albite. Here, the tourmaline is associated with various mica-group minerals and very often overgrown by multicolored Li-tourmalines. It crystallized as secondary due to tourmalinization of the micas, and contains numerous inclusions of apatites, ilmenite, titanite, scheelite, and others. The associated mica-group minerals typically grade from a dark mica through a light green one, a beige one, finally to a colorless mica, all enriched slightly in F, Rb, Cs and probably Li. The studied crystal represents a dark-colored, compositionally heterogeneous tourmaline with numerous inclusions of all the mentioned phases and quartz, grading outwards to a Li-

METHODS Electron microprobe analyses were performed at the Inter-Institute Analytical Complex for Minerals and Synthetic Substances at the University of Warsaw, using a Cameca SX Five FE electron microprobe operating in wavelength-dispersive mode. Formulae of tourmaline were normalized in relation to 6 Si apfu, with B2O3 amounts calculated with the assumption of 3 B apfu, Li2O calculated from stoichiometry, assuming that Li fills octahedral sites up to 9 apfu, and H2O amounts, occurring in the form of OH groups, calculated also from stoichiometry assuming the presence of 4 (OH,F) apfu. These conditions correspond to the maximum OH– and minimum Fe3+ contents in the tourmaline. RESULTS AND DISCUSSION The X site in the tourmaline is almost completely occupied (0.89–0.90 apfu) with the dominant Na, and subordinate Ca and K. Due to Al < 6 apfu, the Z site is only partly filled by the component (4.06–5.22 apfu), and surely completed by Fe3+ and divalent octahedral cations. The Fetot ranges from 2.24 to 3.60 apfu, Mg 0.51–0.56 apfu, Ti 0.13–0.27 apfu, Mn 0.02–0.04 apfu and calculated Li 0.43– 0.64 apfu. In respect of the Z-site occupation, bosiite is a specific member of the oxy-dravite


– povondraite series, with the ratio between the end-members 4:3 (Fig. 1). Compositional plot of the Piława Górna tourmaline in the Fetot– Mg–Al triangle reveals clear, straight arrangement of data points, close to that observed by Flégr et al. (2016) in the Czech ‘bosiite’, but slightly different of that solid solution marked out by oxy-dravite and povondraite. The observed replacement trend is rather in agreement with the substitution leading from dravite to povondraite and vice versa, which can be presented in the form of an exchange vector as Fe3+7O2–[Mg2+Al3+6(OH)–]–1. For other members of the dravite-schorl series it leads to the end-member composition Na(Fe2+2Fe3+)Fe3+6B3Si6O27(OH)3O = NaFe3+3(Fe3+4Fe2+2)B3Si6O27(OH)3O, which can be recognized as Z(Fe3+,Fe2+)-analogue of bosiite. Positions of the Piława Górna tourmaline in the triangle corresponding to the solid solution: dravite – schorl – NaFe3+3(Fe3+4Fe2+2)B3Si6O27(OH)3O tourmaline suggest the content of the tourmaline endmember ~32–33 mol. %, less than end-member schorl (47–48 mol. %), thus the tourmaline still

should represent Al-deficient schorl. However, the most Al-depleted Czech tourmaline shows already ~39–40 mol. % content of the endmember, higher than of dravite (~30 mol. %) and schorl (~31 mol. %). Therefore it could represent a new mineral in the tourmaline supergroup. Assuming the substitution mentioned as valid in both tourmalines, they should exhibit low amounts of WO2–, rather distinctly < 1 apfu as the content typical for bosiite. Bosiite can also be considered as a member of solid solution between dravite and NaFe3+3Fe3+6B3Si6O27O3OH hypothetical tourmaline, being the Fe3+-analogue of olenite, with the ratio 1:2. Starting from Mg-bearing schorl, Na(Fe2+2Mg)Al6B3Si6O27(OH)3OH, the substitution produce a hypothetical NaFe3+3(Al4Fe2+2)B3Si6O27(OH)3O analogue of bosiite. The substitution, however, would lead to VO2– contents between 1.5 and 2.0 apfu in the Czech tourmaline and around 1 apfu in the tourmaline from Piława Górna. As a result, the tourmaline from Piława Górna would be closer to the end-member mentioned. As both hypothetical end-members differ in Fe3+/Fetotal ratio and amounts of O2– in the V and W sites, final confirmation of the crystal-chemical status of the Czech and Polish ‘bosiite’-type tourmalines needs LA-ICP-MS, Mossbauer spectroscopic and structural studies, explaining the presence of hypothetical Li, degree of Fe oxidation as well as occupation of the V and W sites. Acknowledgement: The study was supported by the National Science Centre (Poland) grant 2015/19/B/ ST10/01809 to AP. REFERENCES

Figure 1. Compositional relationships in Fe-bearing tourmaline from Piława (dark crosses) vs. tourmaline of Flégr et al. (2016) (grey field) and the end-member bosiite (green star). Additional symbols: green line – oxy-dravitepovondraite s.s., red lines – Fe3+7O(MgAl6)-1(OH)-1 replacement trends between dravite and povondraite and other members of the dravite–schorl series, blue lines – trends of Fe3+9O3(Mg3Al6)-1[(OH)3]-1 replacement, dark star – Fe2+-analoge of bosiite.


Ertl, A., et al. (2015): Bosiite, IMA 2014-094. CNMNC Newsletter No. 24, April 2015, page 249; Miner. Mag., 79, 247–251. Ertl, A., et al. (2016): Bosiite, NaFe3+3 (Al4Mg2)(Si6O18)(BO3)3(OH)3O, a new ferric member of the tourmaline supergroup from the Darasun gold deposit, Transbaikalia, Russia. Eur. J. Miner., 28, 581–591. Flégr, T., et al. (2016): New occurrence of bosiite in the Řečice pegmatite, Czech Republic. Book of Abstracts. International Scientific Symposium – New Minerals and Mineralogy in the 21th Century, Jachymov 2016.

Hydrothermal alteration of tourmaline from pegmatites enclosed in serpentinites; examples from the Moldanubian Zone (Czech Republic) Jiří Prokop, 1Zdeněk Losos#,1Milan Novák, 1Petr Gadas



Department of Geological Sciences, Faculty of Science, Masaryk University, Kotlářská 2, 611 37 Brno, Czech Republic, # [email protected]

Key words: tourmaline, hydrothermal alteration, pegmatite, serpentinite INTRODUCTION

resembling pegmatites also show strong alteration of feldspars (prehnite, albite, zeolites), quartz (kerolite, smectite and serpentine minerals) and biotite (vermiculite, chlorite).

Tourmaline is a complex borosilicate mineral commonly found as an accessory phase in a wide variety of geological settings. Because of its negligible diffusion up to high temperatures and refractory behaviour tourmaline is considered an excellent petrogenetic indicator (van Hinsberg et al. 2011). However, the alterations of tourmaline have been reported in some granitic pegmatites (e.g. Novák et al. 2013; Čopjaková et al. 2015). Recent study on tourmaline from pegmatites enclosed in serpentinites (Tab. 1) revealed that tourmaline was altered to several minerals during hydrothermal stage with contribution of fluids infiltrating from the host serpentinite. Mostly black tourmalines (dravite, oxy-dravite, schorl, osy-schorl, uvite) show elevated contents of Mg (and Ca) as a result of external contamination of pegmatite melt from host serpentinite (Novák et al. 2017). These pegmatites and plagioclase-tourmaline rocks


At two localities tourmaline occurs in coarsegrained plagioclase-tourmaline rocks (Tab. 1) At Nová Ves near Oslavany plagioclase (labradorite–bytownite) is often replaced by prehnite and minor chlorite and thompsonite. Associated tourmaline is replaced by pumpellyite-(Al) (Fig. 1) and minor prehnite. In Drahonín VI plagioclase is almost completely altered to natrolite and thompsonite-(Ca). Biotite was replaced by vermiculite. Rims of tourmaline are corroded and replaced by natrolite, muscovitealuminoceladonite and chlorite (XMg = 0.85 – 0.89). In simply zoned barren pegmatite at Bernartice near Dolní Kralovice tourmaline is

Table 1. Localities, mineral assemblages and products of replacement of the studied tourmalines.


Parental rock type

Major minerals

Products of mineral replacement Tourmaline composition after tourmaline after feldspar, mica and quartz

Nová Ves near Oslavany

Pl-Tur rock

Pl, Tur

dravite, uvite

Pmp-(Al), Prh

Drahonín VI

Pl-Tur rock

Pl, Tur

dravite, uvite

Ntr, Ms, Chl

Ntr, Thm-(Ca), Vrm, Chl, Ms


barren pegmatite

Kfs, Pl, Ms

schorl, dravite

Kfs, Ms, Chl

Kfs, Chl, Ms, Srp, Sme

beryl-columbite pegmatite

Kfs, Pl, Qz, Bt

schorl, dravite

Kfs, Arf

Vrm, Chl, Ms, kerolite

Věžná I


Prh, Chl, Thm-(Ca)

replaced (Fig. 2) by K-feldspar (Or99Ab1), muscovite (with elevated contents of Si, Mg and Fe) and chlorite (XMg = 0.86 – 0.87). The Věžná I pegmatite is more complex zoned body with apparently higher degree of

bearing minerals - milarite, bavenite, bertrandite, epidydimite, eudydimite, secondary beryl suggests highly variable conditions of these alteration processes. Prismatic crystals of black dravite from outer part of intermediate unit are replaced in veintlets of K-feldspar and masses of arfvedsonite whereas schorl and elbaite from inner units are not altered. CONCLUSION These alterations manifest that tourmaline is apparently unstable during specific hydrothermal conditions. Hydrothermal alterations are likely ranging from early to late subsolidus stage and the alteration processes were locally enhanced by intensive rock deformation. Elevated activities of Na, K and/or Ca are documented by the secondary assemblages (Tab. 1). Highly variable mineral assemblages indicate variable and likely alkaline conditions of these alteration processes although the host rock is the same.

Figure 1. Nová Ves u Oslavan – Alteration of tourmaline (Tur) by pumpellyite (Pmp) association with chlorite (Chl).

REFERENCES Čopjaková, R., et al. (2015): Scandium- and REE-rich tourmaline replaced by Sc-rich REE-bearing epidote-group mineral from the mixed (NYF + LCT) Kracovice pegmatite (Moldanubian Zone, Czech Republic). Am. Mineral., 100:1434–1451. Novák, M., et al. (2013): Geological position, mineral assemblages and contamination of granitic pegmatites in the Moldanubian Zone, Czech Republic; examples from the Vlastějovice region. J. Geosci., 58:21–47. Novák, M., et al. (2017): Tourmaline, an indicator of external Mg-contamination of granitic pegmatites from host serpentinite; examples from the Moldanubian Zone, Czech Republic. Mineral. Petrol., DOI 10.1007/s00710-017-0512-4, 17pp. van Hinsberg V.J., et al. (2011): Tourmaline: an ideal indicator of its host environment. Canad. Mineral., 49:1–16.

Figure 2. Bernartice - Tourmaline (Tur) extensively altered by chlorite (Chl) and secondary Kfeldspar (Kfs II) in association with primary Kfeldspar (Kfs I) and serpentine minerals.

fractionation manifested by occurrence of beryl, Be-rich cordierite, elbaite, lepidolite, and pollucite. Pegmatite underwent several alterations processes including e.g., replacement of cordierite by a mixture of micaceous minerals; alteration of beryl to several Be-minerals; extensive replacement of quartz by kerolite; and vermiculitization of phlogopite. A high variety of secondary Be-


Major and trace element composition of tourmaline in aplites and granites as indicator for magmatic and hydrothermal development (Elba, Italy) 1

Charlotte Redler#, 2Marika Vespa, 3Meinert Rahn


Institute of Earth and Environmeltal Sciences, University of Freiburg, Albertstr. 23b, D-79104 Freiburg, [email protected] 2 Karlsruhe Institute of Technology, Institute for Nuclear Waste Disposal, Hermann.van-Helmholtz Platz 1, D76344 Eggenstein-Leopoldshafen 3 Eidgenössisches Nuklearsicherheitsinspektorat –ENIS–, Industriestr. 19, CH-5200 Brugg #

Key words: Tourmaline, Elba, aplites, major elements, trace elements, magmatic, hydrothermal



Tourmaline (XY3Z6(T6O18)(BO3)3V3W; with the most common site occupancies being X = Na+, Ca2+, K+, X (vacancy); Y = Fe2+, Mg2+, Mn2+, Ti4+, Li+, Fe3+, Al3+ and Cr3+; Z = Al3+, Mg2+, Fe3+, V3+, Cr3+; T = Si4+, Al3+, B3+; B = B3+; V= OH-, O2-; W = F-, OH-, O2-) is a complex borosilicate, typically reflecting the environment of its formation. It is widespread in numerous rock types and is stable over a wide pressure and temperature range (e.g. Dutrow and Henry 2011; Henry and Dutrow 1990; van Hinsberg et al. 2011a,b). Tourmaline occurs in western and central Elba within aplites and pegmatites, which are related to different phases of the magmatic activity during the late Miocene. It is well known from its occurrence as gemstone within pegmatites and miarolitic cavities, which occur at the border to the Monte Capanne pluton on western Elba (e.g. Pezzotta 2001; Bosi et al. 2005a), and was first described by Vernadsky (1914) in leucogranitic dykes of e.g. San Piero in Campo and nearby Sant´Ilario in Campo. In contrast, the occurrence of tourmaline in aplites is relatively unstudied. The aplites consist of mostly of quartz and feldspar with accessory mica, apatite and zircon and frequent tourmaline of variable shape, size and composition. Field and thin section descriptions as well as major and trace element chemical data of tourmalines from various localities on Elba are the base for this study with the goal to distinguish between tourmalines of magmatic and hydrothermal origin.

Tourmaline in the studied aplites from western and central Elba is predominantly alkali-group tourmaline with a schorl-dravite solid solution (Fig. 1a,b). X-vacant-group tourmaline occurs additionally at three of five localities being always foitite with variable Mg contents. As Al is the dominant cation on the Y-site beside Fe and Mg, all tourmalines have an olenite component, but Al and calculated Li are generally too low to classify as elbaite (Fig. 1c). The increase of measured F (always lower than 0.5 atoms per formula unit) is related to an increase in the X-site charge and vice versa. The Li (calculated) and F contents in tourmaline of the aplites are too low to have been formed at late-pegmatitic to postmagmatic hydrothermal conditions. The overall decrease in Fe and Al consistent with an increase in Mg and Ca (and Na) points to a development of tourmaline in the aplites from early-magmatic to late-magmatic conditions as a result of fractional crystallisation within the magmatic reservoir of the Monte Capanne pluton. This is consistent with the general magmatic history of western and central Elba that has been interpreted as a result of mixing between crustal anatectic melts and mantlederived magmas. EASTERN ELBA Three major types of tourmaline are found: (1) prismatic crystals within aplite that cross cut the crystalline basement (Calamita schists); (2) tourmalines occurring within K-feldspar of the


La Serra-Porto Azzurro monzogranite; and (3) black, very fine-grained zones of variable size occurring within the Calamita schists and the monzogranite, which are themselves completely impregnated with tourmaline. The three types of tourmaline differ in their major element contents. Magmatic tourmalines in the aplites (type 1 = Group A) have in general low alkali (Na + Ca) and calculated Li contents together with high Fe and Al contents, and are mostly schorl and foitite. Tourmalines of hydrothermal origin (types 2 and 3 = Group B; Fig. 2a–c) have higher Ca, Na, Mg, and calculated Li contents, and are more variable in composition (schorl, dravite and magnesiofoitite). In thin section, accumulations of fibrous, blue tourmalines (BTA) occurr within cracks and at rims of earlier tourmaline crystals and show compositions similar to the corresponding tourmalines of group A and B (Fig. 2d–f). They indicate a second hydrothermal episode and are developed from local circulating fluids as a recrystallization of earlier tourmalines.


Tourmalines from western Elba are very variable in their trace element composition. Depending on the sample positive Eu anomalies (PC and SP), negative Eu and Ho anomalies (CB) or no anomaly occur, and LREE are generally more enriched than HREE. No clear variation can be given for further trace elements. The trace element compositions of tourmalines in eastern Elba show in general lower Li, Ga, Pb, Zn and Sc, but higher Sn, Sr and Co contents in the BTA than in group A and B tourmalines. REE in the BTA show enriched HREE and occasionally a positive Eu anomaly, whereas group A and B tourmalines have negative or none Eu anomalies. The trace element composition of zoned tourmalines is variable within cores and rims, and no clear evidence for enrichment or depletion of distinct elements can be given.

Figure 1. Classification of tourmaline in samples of western and central Elba: a) Primary tourmaline groups; b) Alkali-group tourmalines; c) X-group tourmalines.


Figure 2. Classification of tourmalines crystals in nine samples (a–c) and blue tourmaline accumulations (BTA; d–f) in five of nine samples from eastern Elba: a) Primary tourmaline groups; b) Alkali-group tourmalines; c) X -group tourmalines; d) Primary BTA groups of; e) Alkali-group BTA; f) X-group BTA. The black line in a–c indicated the division into group A and group B samples, which does not fit with the classification of BTA.


Pezzotta, F. (2001): Elba: die klassische Urlaubsinsel der Mineralogie. Weise, München. van Hinsberg, V.J., et al. (2011a): Tourmaline as a petrologic forensic mineral: a unique recorder of its geologic past. Elements, 7: 327–332. van Hinsberg, V.J., et al. (2011b): Tourmaline: An ideal indicator of its host environment. Canad. Mineral., 49:1–16. Vernadsky, W. (1914): Über die chemische Formel der Turmaline. Z. Kristallogr., 53: 273–288.

Dutrow, B.L. and Henry, D.J. (2011): Tourmaline: A geologic DVD. Elements, 7: 301–306. Henry, D.J. and Dutrow, B.L. (1990): Ca substitution in Li-poor aluminous tourmalines. Canad. Mineral., 28:111–124. Henry, D.J. and Dutrow, B.L. (2011): The incorporation of fluorine in tourmaline: internal crystallographic controls or external environmental influences? Canad. Mineral., 49:41–56.


Composition of tourmaline in orogenic gold deposits 1


Marjorie Sciuba#, 1Donald Grzela, 1Nelly Manéglia, 1Sheida Makvandi, 1Georges Beaudoin

Université Laval, #[email protected]

Key words: orogenic gold, major and trace elements, partial least square-discriminant analysis


veins cutting the La Ronde volcanogenic massive sulfide deposit (Canada), mineralized pegmatite in the Roberto gold deposit (Canada) and the Nevada gold deposit (USA) that has an epithermal affinity. In orogenic gold deposits, tourmaline is commonly associated with the mineralization. It may be found in quartzcarbonate veins or disseminated in vein selvages.

The indicator mineral technique is broadly used in exploration for many deposit types. Tourmaline has physical and chemical properties such as hardness, specific gravity and chemical stability to resist long transport distance and thus ability to be used as an indicator mineral. Tourmaline occurs in a wide range of geological settings including sedimentary, metamorphic, igneous rocks and minerals deposits such as porphyry, greisen, polymetallic veins and breccias, SEDEX, VMS and IOCG (Griffin and Ryan 1995; Slack 1996; Slack and Trumbull 2011). Thus, when tourmaline is found in heavy mineral concentrate from surficial sediments during indicator mineral survey, it is useful to constrain the nature of the source. This study aims to define discriminant geochemical features for tourmaline from orogenic gold deposits.

ANALYTICAL METHODS Tourmaline was analysed at Université Laval, with a CAMECA SX-100 Electron Probe Micro-Analyser for major and minor elements, and at UQAC with RESOlution M-50 Excimer 193 nm laser coupled to an Agilent 7700x ICPMS for minor and trace elements. Both methods are described in Grzela (2017). Partial Least Square-Discriminant Analysis (PLS-DA) was performed on the LA-ICP-MS data to test the potential to discriminate the hostrock composition using the method described in Makvandi et al. (2016). PLS-DA is a multivariate statistical analysis that emphasizes the maximum variation among groups and thus highlight the best separation between the groups (de Iorio et al. 2007).


A set of tourmaline from 18 world-class and other orogenic gold deposits and districts are included in the study. Deposits were selected for the diversity in geological settings including hostrock composition (sedimentary, mafic to felsic rocks), age of hostrock (Archean to Paleozoic), metamorphic facies of the hostrock (from lower greenschist to upper amphibolite), and age of mineralization (Archean to Carboniferous). Additional samples from various geological environment include metamorphic tourmaline from the Tehery-Wager area (Canada), hydrothermal

RESULTS Tourmaline from orogenic gold deposits is alkalic and has dominantly dravitic to schorlitic composition (Fig. 1a,b) with an average composition X-, Y-, Z- and T-sites: (Na0.65Ca0.14X0.22)(Fe2+1.06Mg1.59Al0.14Ti0.08Cr0.02) (Al5.86Mg0.13)(Si5.97Al0.07)


Figure 1. Classification of tourmaline from orogenic gold deposits and other deposit types and localities (a) X-Vacant-Ca-(Na+K) ternary diagram (b) Mg/(Fe+Mg) vs Vac/(Na+K+Vac) diagram and (c) Mg/(Fe+Mg) vs Ca/(Ca+Na) diagram (Henry et al. 2011).

In some exceptions, tourmaline from orogenic gold deposits belong to the Mgfoitite, the foitite, the uvite and the feruvite fields (Fig. 1b,c). Core and rim defined by different color, have most commonly similar major element composition (Fig. 1). Tourmaline from orogenic gold deposits hosted in sedimentary, felsic, intermediate and mafic rocks show little compositional discrimination based on major elements including Si, Al, Fe, Mg and Na (Fig. 2). Tourmaline from deposits hosted in mafic rocks has an average Cr concentration higher than those hosted in sedimentary, felsic and intermediate rocks (Fig. 2). Tourmaline from deposits hosted in sedimentary rocks has average Sr, K and Mn concentrations higher than those hosted in felsic, intermediate and mafic rocks (Fig. 2). Tourmaline from orogenic gold deposits has similar major element composition to those from the Nevada gold deposit, hydrothermal veins from La Ronde, the Roberto pegmatite and metamorphic tourmaline from the TeheryWager area. However, tourmaline from

orogenic gold deposits has consistantly lower Mn concentrations and higher Sr concentrations than those from the Nevada gold deposit, hydrothermal veins from La Ronde, the Roberto pegmatite and metamorphic tourmaline from the TeheryWager area (Fig. 2). PLS-DA on trace elements in tourmaline from orogenic gold deposits was used to test the signature of the hostrock composition (Fig. 3). The component Qw1* is defined by high positive contributions of Sc, Co, Ni and Yb and high negative contributions of Eu, Mn, Li and Ti. The component Qw2* is defined by high positive contributions of Sr, Sn, Be and Ga and high negative contributions of Li and Eu. Tourmaline from deposits hosted in felsic and intermediate rocks have commonly positive t1 and positive t2, those from deposits hosted in mafic rocks have mostly negative t2, and those from deposits hosted in sedimentary rocks have mostly negative t1. The rock composition fields overlap partially with each other (Fig. 3).

Figure 2. Major and minor element average composition of tourmaline from orogenic gold deposits and other deposit types and localities.


Figure 3. PLS-DA of LA-ICP-MS data for tourmaline from orogenic gold deposits (a) Qw1*-Qw2* loadings and (b) t1-t2 scores.


REFERENCES Condie, K. (1993): Chemical composition and evolution of the upper continental crust: Contrasting results from surface samples and shales. Chem. Geol., 104:1–37. de Iorio, M., et al. (2007): Statistical Techniques in Metabolic Profilling In: Balding, D., et al. (eds) Handbook of statistical genetics. 3rd ed., 347–373. Goldfarb, R. and Groves, D. (2015): Orogenic gold; common or evolving fluid and metal sources through time, Lithos, 233:2–26. Grzela, D. (2017): Chemical composition of indicator minerals from orogenic gold deposits and glacial sediments of the Val-d’Or district (Québec, Canada), MSc thesis, Université Laval, Québec, Canada. Griffin, W. and Ryan, C. (1995): Trace-elements in indicator minerals - Area selection and target evaluation in diamond exploration. J. Geochem. Explor., 53:311–337. Henry, D., et al. (2011): Nomenclature of the tourmaline-supergroup minerals, Am. Mineral., 96:895–913. Makvandi, S., et al. (2016): Partial least squaresdiscriminant analysis of trace element compositions of magnetite from various VMS deposit subtypes: Appli. to min. explo., 78:388–408. Slack, J. (1996): Tourmaline associations with hydrothermal ore deposits. Rev. Mineral., 33: 559–643. Slack, J. and Trumbull, R. (2011): Tourmaline as a Recorder of Ore-Forming Processes. Elements, 7:321–326.

Tourmaline from orogenic gold deposits has dominantly dravitic to schorlitic composition. However, the major element composition does not vary systematically with the hostrock composition. The high Cr concentrations in tourmaline from deposits hosted in mafic rocks and the high Sr and K concentrations in tourmaline from deposits hosted in sedimentary rocks reflect the composition in the respective hostrocks (Condie 1993). The lower Mn and higher Sr concentrations in tourmaline from orogenic gold deposits, compared to those from the Nevada gold deposit, hydrothermal veins from La Ronde, the Roberto pegmatite and the metamorphic tourmaline from the Tehery-Wager area show that Mn and Sr may be useful minor elements to discriminate tourmaline from orogenic gold deposits, from those from other geological settings. The first two components, Qw1*-Qw2*, capture the highest variability in the trace element composition of tourmaline from orogenic gold deposits. The low variability of the trace element composition of tourmaline from orogenic gold deposits reflects a similar source of fluid for all orogenic gold deposits as mentioned in Goldfarb and Groves (2015). However, the trace element composition of tourmaline is modified by reaction with the hostrocks as shown by Sr, Mn, Cr and PLS-DA of trace elements.


Uvite and feruvite from lamprophyre dikes and metasediments near the O’Grady batholith in the Northwest Territories, Canada 1

Emily Scribner#, 2Jan Cempírek, 1Lee Groat


The University of British Columbia, 2207 Main Mall, Vancouver, BC Canada V6T 1Z4, [email protected] 2 Masaryk University, Kotlářská 267/2, 611 37 Brno, Czech Republic #

Key words: Uvite, feruvite, lamprophyre dikes, O’Grady batholith, Northwest Territories INTRODUCTION

to magnesiohornblende, anorthite-albite and Kfeldspar, with clinochlore, pyroxene group minerals, titanite, apatite, and minor tourmaline group minerals, pyrite, allanite(Ce), and zircon. Tourmaline occurs at the exocontact zone of the lamprophyre dikes (in a quartz vein) and in a endocontact zone near the margins. This endocontact zone is composed of actinolite to magnesiohornblende, clinochlore, titanite and quartz, with minor pyroxene group minerals and apatite.

A tourmaline-rich metasedimentary sequence hosts pegmatites, quartz-tourmaline veins, and lamprophyre dikes in an area near the O’Grady batholith in the Northwest Territories, Canada (62°46’8.33”N, 128°56’9.07”W; Fig. 1). The O’Grady batholith was emplaced at 94 Ma (Anderson 1983) as part of the hornblendebearing Tombstone suite of the larger calcalkaline Selwyn plutonic suite (Gordey and Anderson 1993). A band of pegmatitic granite located on the west side of the O’Grady batholith contains a small area of highly evolved, Li-bearing, mixed NYF-LCT-type pegmatite dikes (Ercit et al. 2003). Pegmatite dikes also occur in the same area as the lamprophyre dikes but they are relatively depleted in rare-elements and contain foititeschorl (sometimes F-rich) as the dominant tourmaline species. The lamprophyre dikes trend north-south and range from 2-6 m in length. They have not been dated, but other lampro-phyre dikes associated with the Selwyn plutonic suite have been dated from 84 to 96 Ma (Gordey and Anderson 1993). They are composed of quartz, actinolite

RESULTS AND DISCUSSION Three generations of tourmaline are recognized within the lamprophyre dikes: Tur I and II occur at the margins of the dikes, in a quartz vein at their exocontact, and Tur III occurs in the endocontact zone near the margins of the dikes (Fig. 2). Tur I is zoned with compositions mainly alternating between dravite and uvite. Some small areas in the crystals adjacent to the lamprophyre dike are more enriched in Fe and reach compositions of feruvite. Tur II has patchy chemical zoning: its composition ranges from uvite to feruvite to fluor-uvite. Tur III occurs in altered masses of mostly dravite, but also uvite, feruvite, and minor fluor-uvite. In general, Tur III is richer in Ca and Fe (and Ti) than Tur I (Fig. 3). Feruvite is enriched in Ti (up to 2.17 wt. % TiO2; 0.276 apfu Ti) and fluor-uvite contains up to 0.45 apfu F. Although contents of V and Cr in all tourmaline is low, Tur III is the most enriched in these elements with up to 0.13 wt. % V2O3 in feruvite and 0.35% wt. % Cr2O3 in dravite. The composition of Tur I is similar to tourmaline from the metasedimentary host rock. This, together with its position at the margin of the lamprophyre dikes, suggests that

Figure 1. Location of the study area.


lamprophyre dike has a lighter δ11B isotopic signature than tourmaline from nearby pegmatite and aplite dikes (average δ11B = -3.5 and -4.3 respectively). All δ11B values are in the range of the average continental crust (Chaussidon and Albarède 1992) indicating no influence of other fluid sources.

Figure 2. Zoned, primary Tur I and Tur II with patchy chemical zoning at the margin of a lamprophyre dike, adjacent to greenish-gray Tur III within the lamprophyre dike.

Tur I represents the original composition of tourmaline prior to the intrusion of the lamprophyre dikes. The intrusion of the dikes caused the remobilization of B-rich fluids, which reacted with Ca,Fe,Ti-bearing minerals in the dikes to form the more Ca,Fe,Ti-rich Tur III. The composition of Tur II is intermediate to Tur I and Tur III but more similar to Tur III, suggesting that the effect of the remobilized fluids increases towards the interior of the lamprophyre dikes (Fig. 3). Single-crystal X-ray diffraction measurements of the Tur II (Ca,OH-dominant, Al-deficient feruvite) were made using a Bruker X8 APEX II diffractometer at the University of British Columbia. The unit cell dimensions are a = 15.9787(4) Å, c = 7.2163(2) Å, V = 1595.61 Å3, similar to feruvite and particularly to lucchesiite (Bosi et al 2017). Compared to lucchesiite, the studied tourmaline is strongly Al-deficient (Al < 5.5 apfu) and also OH-dominant; an estimation of Fe3+ content will be necessary to determine if the tourmaline is actually a Fe3+-rich lucchesiite. Boron isotopes from a tourmaline crystal from the margin of a lamprophyre dike were measured using a secondary ion mass spectrometer at the University of Manitoba. In general, the core of the tourmaline crystal (average δ11B = -7.9) is isotopically lighter than the rim (average δ11B = -7.3). A zone that is brighter in BSE imaging at the outermost rim of the tourmaline has the lightest isotopic signature of δ11B = -8.7. Tourmaline from the

Figure 3. Variation in Ca versus Mg / (Mg+Fe) in Tur I, II, and III.

REFERENCES Anderson, R.G. (1983): Selwyn plutonic suite and its relationship to tungsten skarn mineralization, southeastern Yukon and District of Mackenzie. In: Current Research, Part B. Geological Survey of Canada Paper 83-1B:151–163. Bosi, F., et al. (2017): Lucchesiite, CaFe2+3Al6(Si6O18)(BO3)3(OH)3O, a new mineral species of the tourmaline group. Mineral. Mag., 81:1–14. Chaussidon, M. and Albarède, F. (1992): Secular boron isotope variations in the continental crust: an ion microprobe study. Earth Planet. Sci. Lett., 108:229–241. Ercit, T.S., et al. (2003): Granitic pegmatites of the O’Grady batholith, N.W.T., Canada: a case study of the evolution of the elbaite subtype of rare-element granitic pegmatite. Canad. Mineral., 41:117–137. Gordey, S.P. and Anderson, R.G. (1993): Evolution of the northern Cordilleran miogeocline, Nahanni map area (105I), Yukon and Northwest Territories. Geological Survey of Canada: Memoir 428. 214 pp.


The problems of tourmaline single crystals growth 1

1 2

Tatiana Setkova#, 1Vladimir Balitsky, 2Oleg Vereshchagin

Institute of Experimental Mineralogy RAS, Russia, # [email protected] Institute of Earth Science, SPbSU, Russia

Key words: tourmaline, crystal growth, hydrothermal solutions INTRODUCTION


Tourmaline is one of the most widespread minerals in nature and one of the most popular gems and promising piezoelectric material. The growth of large crystals is today a topical task.

In the Laboratory of Synthesis and Modification of Minerals of the Institute of Experimental Mineralogy RAS was made an attempt to develop a hydrothermal technique of tourmaline crystals growth on the seed in the temperature range 400-750°C and pressures of 100-150 MPa. The experimental data on stability and features of the crystallization of tourmaline and thermodynamic calculations showed a sufficiently low solubility of tourmaline in boric and boric-chloride hydrothermal solutions. In boric-alkali, boricfluoride and boric-chlorine-fluoride solutions, the intensity of dissolution of tourmaline appreciably increases and is accompanied by the formation of aluminosilicate and fluoride phases (Setkova et al. 2009). In all solutions tourmaline grow as overgowth on the seed and as spontaneous nucleation crystals, at using quartz, corundum and stoichiometric tourmaline-forming component as a nutrient. The nucleation increases with boric acid content in the solution. As a result, the growth parameters of tourmaline were significantly reduced to 450°C and 100 MPa, and technique of colored Fe, Ni, Co, (Ni, Fe), (Ni, Cr), (Co, Ni, Cr), (Ga,Fe)-bearing tourmaline crystals growth on seed was developed (Setkova et al. 2011, Rozhdestvenskaya et al. 2012; Vereshchagin et al. 2016). The maximum growth rate of the crystals was 0.05 mm/day by the faces of the trigonal pyramid in optic axis direction and was observed for Co-bearing tourmaline, which was formed in boric solutions of cobalt chloride crystallohydrate at a temperature range of 400 up to 700°С and pressure 100 MPa. The color intensity of the

PREVIOUS WORKS Investigations on the tourmaline crystals growth began in the middle of the twentieth century. Crystals were synthesized at temperatures between 400 and 800°C and pressures of 70–4000 MPa using glasses of tourmaline composition (Smith 1948), mixtures of oxides and other compounds, and minerals and rocks containing silicon and aluminum oxides (sillimanite, kyanite, grandidierite, pyrophillite, almandine, labradorite, etc.) (Fuchs et al. 1998, Marler et al. 2002). The tourmaline single crystals can be grown on a seed crystal (Voskresenskaya and Barsukova 1968; Goerne et al. 1999). In particular, Voskresenskaya has developed a method of growth of colored (Fe, Mg, Co, Ni, Mn, Cr, etc.) species of tourmaline. The biggest overgrowth (up to 3 mm) on a seed crystal was observed in the case of Co-bearing tourmaline. Crystal growth was performed under very high temperatures (750°C) and pressures (200 MPa) using highly concentrated (up to 80 wt%) boron–chloride solutions that required usage of very expensive materials. The latest works on the tourmaline synthesis (London 2011; Berryman et al. 2014; Ertl et al. 2015; Kutzschbach et al. 2016) were carried out mainly to study the stability, mineral equilibria and the structure of this mineral. The crystals size in the above works do not exceed 1 mm.


Figure 1. Synthetic tourmalines grown in hydrothermal solutions.

crystals depends on the content of cobalt in tourmaline crystals (Fig. 1a-c). In boric acid, boric-alkaline boric-chloride and boric-fluoride solutions the Fe-bearing tourmaline was formed. The highest iron content was observed in crystals formed in boric acid solutions. Green (Fe, Ni)-bearing tourmaline (see Fig. 1g) was formed in boricfluoride solutions, emerald green (Ni, Cr)bearing tourmaline - in boric-chlorine-fluoride solutions (see Fig. 1d). In complex boronchloride solutions, the crystallization of polychromic (Co, Ni, Cr)-bearing tourmaline (see Fig. 1h) was also established. The color of this tourmaline varies from pink to green, with a change of Co content on to Ni, Cr.

REFERENCES Berryman, E., et al. (2014): Am. Mineral., 99:539–542. Ertl, A., et al. (2015): Canad. Mineral., 53:209–220. Fuchs, Y., et al. (1998): Am.Mineral., 83:525– 534. Goerne, G., et al. (1999): Eur. J. Miner., 11. 1061-1077. Kutzschbach, M., et al. (2016): Am. Mineral., 101, 93–104. London, D. (2011): Canad. Mineral., 49:117136. Marler, B., et al. (2002): Eur. J. Mineral., 14: 763–771. Rozhdestvenskaya, I.V., et al. (2012): Cryst. Rep., 57, 1:57-63. Setkova, T., et al. (2011): J. Crystal Growth., 318:904–907. Setkova, T., et al. (2009): Dokl. Earth Sci., 425 (3):490–493. Smith, F.G. (1948): Econ. Geol., 44:186–192. Vereshchagin, O., et al. (2016): Eur. J. Mineral., 28, 3:593–599. Voskresenskaya, I.E. and Barsukova, M.A. (1968): Hydrother. Syn. of Mineral.:175–192 (in Russian).


The problems of tourmaline single crystal growth associated with: complex chemical composition of tourmaline, the high stability in hydrothermal solution, the low growth rate (relatively high growth rate only in optical axis direction), high P-T parameters of growth and highly concentrated solutions and, as a consequence, the use of expensive equipment and materials.


Development of boron isotopic analysis of tourmaline by LA-MC-ICP-MS and its application to natural samples 1

Ryuichi Shinjo#, 1,2Kazuki Kinjo, 3Kazuo Nakashima, 4Teruyoshi Imaoka


Dept. Physics & Earth Sci., University of the Ryukyus, JAPAN, # [email protected] Cabinet office, Okinawa General Bureau, JAPAN 3 Dept. Earth and Environmental Sciences, Yamagata University, JAPAN 4 Dept. Geosphere Sciences, Yamaguchi University, JAPAN 2

Key words: boron isotope, LA-MC-ICP-MS, Yamaguchi, Japan

INTRODUCTION Table 1: Boron isotopic data for SLT.

We have developed the analytical method for in situ boron isotopic composition of tourmaline using LA-MC-ICP-MS (Laser ablation multi-collector inductively coupled plasma mass spectrometry). Boron is a major constituent in tourmaline; its B isotopic composition reflects the nature of fluids involved in tourmaline crystallization. Variation in 11B/10B at grain-size or internal single grain in rock/vein may reveal temporal change of fluid composition and provide important constraints on dynamical petrogenetic and tectonic processes.


d11 B ± 2SE (2SD)

Acid digestion

-13.403 ± 0.057

Alkali fusion

-13.442 ± 0.061

Laser ablation

-13.38 ± 0.18 (n=9)

Table 2: Laser and MC-ICP-MS operating settings. Laser settings

METHOD DEVELOPMENT As a in-house standard, large grain (~2×2×5 cm) of tourmaline (named as SLT) from Brazil was selected to monitor data quality in routine analysis. Powders made from a portion of SLT were decomposed by both acid decomposition and alkali fusion methods. Boron separation were done by using a semi-automated separation machine with a 6-port valve conected to mini-column filled with boron specific resin (Amberlite IRA 743). Our wet separation procedure requires no evaporation step, which sometime introduce potential B loss, fractionation, and blank contribution during sample dry-down on hot plate. Faraday detectors in MC-ICP-MS were used to analyze a tourmaline. The δ11B values of SLT by acid decomposition and alkali fusion were δ11B = -

Laser probe

ESI NWR femto


1 J/cm2

Repetition rate

1 Hz

Spot diameter

65 microns

Ablation time

60 s

Wash-out time

40 s

MC-ICP-MS settings MC-ICP-MS:

Thermofisher Neptune Plus

Inlet system Cool gas (Ar)

15.0 L/min

Auxiliary gas (Ar)

0.70 L/min

Sample gas (Ar)

0.75 to 0.85 L/min

Additional gas (He)

0.8 L/min

RF power

1200 W

Interface cones

Nickel sampler and X-skimmer




-2000 V


-700 V

13.40 ± 0.06 ‰ and δ11B = -13.44 ± 0.06 ‰, respectively (Tab. 1). Laser ablation and MC-ICP-MS settings and parameters are listed in Tab 2. The ‘standard-sample-standard’ bracketing procedure was adopted to correct for isotopic fractionation. We used tourmaline standard B4, distributed from IAEA, as matrix-matched primary standard, instead of NIST SRM 610 which has been used as external standard (e.g., Tiepolo et al. 2006). Ablation was carried out under He flux; He and the ablated material are mixed with Ar downstream of the ablation cell in a PFA chamber. Data were calculated offline by using Iolite software (Paton et al. 2011). The average of δ11B of traverse-analysis of SLT grain by LA-MC-ICP-MS was -13.38 ± 0.18 ‰ (n=9). It is confirmed that SLT is homogeneous, suitable for micro analyses. The δ11B values by wet analyses and LA-MC-ICPMS traverse-analysis were identical within the analytical error. Therefore SLT is suitable for the use as an in-house standard material. Longterm reproducibility for SLT was δ11B = -13.65 ± 0.48 (2SD) ‰. 2SD of 0.4 to 0.6 ‰ has been reported in literatures (le Roux et al. 2004; Tiepolo et al. 2006; Mikova et al. 2014; Lin et al. 2014; Öztürk et al. 2015).

the Hikoshima and Murodu areas are classified into 4 types in each area. In the Hikoshima area, δ11B of tourmaline range from -10.07 to +0.7 ‰. In contrast, δ11B of host rocks range from -12.04 to +0.84 ‰. The high δ11B of tourmaline appears to be affected by intrusive rocks in the Hikoshima area. In the Murodu area, δ11B of tourmaline (-7.62 to -5.32 ‰) overlap roughly with δ11B of host rocks (-7.95 to -5.59 ‰). REFERENCES Hou, K., et al. (2010): In situ boron isotope measurement of natural geological materials by LA-MC-ICP-MS. Chinese Sci. Bull., 55: 3305–3311. le Roux, P.J., et al. (2004): In situ, multiplemultiplier, laser ablation ICP-MS measurement of boron isotopic composition (11B) at the nanogram level. Chem. Geol., 203:123–138. Lin, L., et al. (2014): Determination of boron isotope compositions of geological materials by laser ablation MC-ICP-MS using newly designed high sensitivity skimmer and sample cones. Chem. Geol., 386:22–30. Mikova, J., et al. (2014): Matrix effects during laser ablation MC ICP-MS analysis of boron isotopes in tourmaline. Royal Soc. Chem., 29: 903–914. Paton, C., et al. (2011): Iolite: Freeware for the visualisation and processing of mass spectrometric data. J. Anal. Atom. Spect., 26: 2508–2518. Tiepolo, M., et al. (2006) Laser ablation multicollector ICPMS determination of d11B in geological samples. Applied Geochem., 21: 788–801.

APPLICATIONS TO NATURAL SAMPLES Using the developed method, boron isotopic composition of tourmalines from Shimonoseki prefecture, Japan were analyzed. Boron concentration and boron isotopic composition were also analyzed for the basement rocks from the region, by ICP- AES (Inductively coupled plasma atomic emission spectrometry) and MC-ICP-MS, respectively. Tourmalines in


Fractionation in Li-tourmaline from the Lhenice lepidolite pegmatite - "darrellhenryite loop" in F-saturation 1

1 2

Lenka Skřápková, 1Jan Cempírek, 2Michaela Vašinová Galiová

Department of Geological Sciences, Faculty of Science, Masaryk University, Kotlářská 2, 611 37 Brno Department of Chemistry, Faculty of Science, Masaryk University, Kotlářská 2, 611 37 Brno

Key words: pegmatite, tourmaline, lepidolite, elbaite, darrellhenryite, fluorine


amblygonite, minor blocky K-feldspar), and a lepidolite core (medium-grained lepidolite + quartz + pink and white tourmaline). The latter two zones are locally rich in cassiterite, whereas columbite-tantalite is less frequent.

The lepidolite-subtype pegmatite at Lhenice, South Bohemia, Czech Republic, is a moderate-sized, highly fractionated pegmatite with abundant lepidolite, elbaite, amblygonite, and cassiterite. The pegmatite was discovered few years ago, and has never been studied so far. As the first mineralogical study at the locality, we acquired EMPA and LA-ICP-MS data of pink and colorless tourmaline from the most fractionated pegmatite units. The studied pegmatite is located in the southern part of the Moldanubian Zone, characterized by presence of high-grade metamorphic rocks retrogressed at MP-MT conditions (gneisses, granulites, serpentinites). The area contains several occurrences of lepidolite-subtype pegmatites (besides multiple less fractionated dikes) within the SouthBohemian pegmatite field (Novák and Cempírek 2010), including the type locality of darrellhenryite (Nová Ves – Novák et al. 2013). In general, the pegmatites are characterized by high contents of P2O5, F and B2O3 (amblygonite-montebrasite, lepidolite, tourmaline) and a high degree of fractionation.

METHODS Tourmaline samples were mapped in backscattered electron images and analyzed using electron microprobe Cameca SX100 (Department of Geol. Sci., MU). The EMPA-analyzed spots were subsequently measured using LA-ICP-MS system (Department of Chemistry, MU), using a laser ablation system Analyte G2 (Teledyne CETAC Technologies) connected to a sector field ICPMS spectrometer Element 2 (Thermo Fischer Scientific). TOURMALINE COMPOSITION Tourmaline from the intermediate zone (Tur 1; zoned, prismatic crystals up to 5 cm long and 1 cm thick, green, blue and pink in color) shows a gradual compositional change from Cabearing fluor-elbaite, darrellhenryite and to Mn-bearing fluor-elbaite (Fig. 1). Subsequent fracturing was followed by minor hydrothermal replacement of the earlier tourmaline along fractures and overgrowths of new fibrous tourmaline (Tur 2) with elevated Na and F contents, Fe-bearing fluor-elbaite (Fig. 1). The process was followed by crystallization of the latest, fracture-filling and interstitial Fe-rich fluor-elbaite (Tur 3) that overgrowths all earlier tourmaline generations.

PEGMATITE STRUCTURE The pegmatite dike forms flat-laying tabular body, ca. 4 m thick and 20 m long in its most fractionated part. The dike has concentric zoning (from border to center): border zone (granitic texture, minor black tourmaline), wall zone (graphic feldspar with abundant black tourmaline), intermediate zone (albite + quartz + blue, green and pink tourmaline +


Tourmaline from the lepidolite core (white, fibrous crystals ca. 2 cm long) is darrellhenryite (Tur 1), fractured by the latest Fe-rich fluor-elbaite (Tur 3). The compositional evolution therefore shows a marked trend of a strong decrease in F (and minor decrease in Na) during crystallization of the primary tourmaline, towards darrellhenryite and rossmanite, followed by the increase of F in crystal rims. Later tourmaline stages have much higher Na and F contents than the zoned primary tourmaline. TOURMALINE EVOLUTION When looking in detail at the tourmaline assemblages, we can conclude that the primary tourmaline (Tur 1) crystallized from Na-rich melt that produced abundant albite. The tourmaline crystallization overlaps with emergence of the first mica n the albite zone; we suggest that the sudden drop in the tourmaline F content was caused by start of mica crystallization. The F-depleted tourmaline (darrellhenryite) in the lepidolite core is also either coeval or slightly older than the associated lepidolite. The evolution path of the early tourmaline therefore follows a “loop” of Na,F decrease and increase, turning at darrellhenryite composition (Fig. 1). Late development of the primary tourmaline is characterized by an increase in Na, OH, F and Fe contents. The new tourmaline generation (Tur 2) is associated by late K-feldspar and muscovite, suggesting hydrothermal crystallization. Textures of latest Fe-enriched (but not Mgor Ca-bearing) fluor-elbaite (Tur 3) indicate hydrothermal origin; we suggest that its high Fe content might reflect composition of primary hydrothermal fluid in a closed system. REFERENCES Novák, M., Cempírek, J. (2010): Granitic pegmatites and mineralogical museums in Czech Republic. IMA 2010 Field trip guide CZ2. Acta Mineral. Petrogr., Field guide series, 6:1–56. Novák, M., et al. (2013): Darrellhenryite, Na(LiAl2)Al6(BO3)3Si6O18(OH)3O, a new mineral from the tourmaline supergroup. Am. Mineral., 98:1886–1892.

Figure 1. Compositional evolution of tourmaline from most fractionated units of the Lhenice pegmatite.


Petrographic study of tourmaline-rich metamorphic rocks in the Sopron area, Eastern Alps (W-Hungary) 1

Tamás Spránitz#, 1Zoltán Kovács, 1Benjámin Váczi, 2 Kálmán Török, 1Sándor Józsa

1 Department of Petrology and Geochemistry, Eötvös Loránd University, Budapest, Hungary, # [email protected] 2 Geological and Geophysical Institute of Hungary, Budapest

Keywords: tourmaline, petrography, mineral chemistry, orthogneiss, tourmalinite, quartzite, Sopron, Eastern Alps


investigated well in a few old quarries. However, small natural exposures such as outcrops in valleys and autochtonous debris of bedrocks show a diverse composition including tourmaline-rich rocks, which litologies have not yet been described from the area. Such rocks are tourmaline-rich pegmatitic orthogneisses, tourmalinites, kyaniteleuchtenbergite-muscovite schists and quartzites crosscut by subordinate quartztourmaline veins and layers. The aim of this study is to give a detailed petrographic description of these rocks, based on optical and scanning electron microscopy supplied by chemical (SEM-EDS) and structural (Raman-spectroscopy, X-ray diffraction) analyses.

The Sopron area is situated in the western margin of the Pannonian basin and tectonically belongs to the Grobgneiss Series of the Lower Austroalpine nappe system (Eastern Alps) (Török 1996). The metamorphic complex is built up by orthogneisses, micaschists, leucophyllites, quartzites, and amphibolites. The metamorphic evolution of the complex has been studied by several authors in the last three decades (e.g. Lelkes-Felvári et al. 1984; Kisházi-Ivancsics 1987; Török 1996, 2001; Balogh and Dunkl 2005). Based on the mineral assemblages of the mica-schists Variscan, Permo-Triassic and Alpine metamorphic imprints can be distinguished. The protolith of the orthogneiss is a Variscan peraluminious two-mica leucogranite (Török 1998). The gneisses (regardless of the presence of foliation) show an Alpine HP amphibolite facies metamorphic event. After the Alpine peak metamorphism, in the close vicinity of shear zones Mg-metasomatism caused the formation of leucophyllites and quartzites in micaschists and in gneisses as well. REE-rich phosphate mineralisation, caused by late highsalinity fluids, postdates the formation of leucophyllites and a final stage of limoniteclay mineralisation can also be recorded during the Alpine retrogression (Török 2001). Although, the Sopron area is relatively large (ca. 40 km2), the quality and amount of outcrops are very poor, the composing and well described litologies can only be

RESULTS The ortogneisses have pegmatitic texture which is displayed by coarse-grained rock forming minerals such as quartz, plagioclase, tourmaline, garnet and white mica. Most of them are slightly foliated, but some unfoliated metagranites are also present. Quartz shows different types of high-temperature grain boundary migration recrystallisation textures, but sometimes it has polygonal fabric. Kfeldspars are only present in metagranites, all feldspars of the gneisses show nearly clear albitic composition. Feldspars contain numerous small white mica needles and flakes as inclusions, which show phengitic composition (Si up to 6.65 apfu.). Tourmaline


is mostly euhedral, cracked and shows strong colour-zoning (pale yellowish-brown, pale brown and light bluish-green from core to rim). The chemical composition of all zones belongs to schorl-dravite solid solution. A tendency can be observed: Mg content increases from core to rim, but the tourmalines in metagranites show gradual Fe enrichment from core to rim. Fe-rich schorl also appears in crack fillings. In these cracks high-Si white mica can be also observed. The white micas always show an increasing Si content from core to rim. Garnet cores are Fe-Mn rich (Alm-Sps), and a narrow Ca-rich rim is identified. Metagranites contain beryl in small amounts, which proves to be a new mineral in this region. In some orthogneisses feldspars are completely replaced by fine-grained white mica, Mg-rich chlorite and quartz. Additionally kyanite and different phosphate minerals (lazulite, florencite, goyazite-plumbogummitecrandallite, monazite and apatite) are also present in significant quantities. Apatite is quite frequent as inclusions in feldspar and as subhedral crystals near white mica rich finegrained veins as well. Tourmalinites are, a newly discovered litology in this region, containing tourmaline up to 80 vol%. These rocks are built up by alternating quartz and tourmaline layers. Tourmaline is generally coarse-grained (few cms), deformed, contains chlorite inclusions, shows complex and oscillatory zoning or mosaic pattern with polygonal fabric. The composition of tourmaline varies in a wide range: the light blue cores and brownish regions with oscillatory zoning are schorldravite solid solutions, dark brown rims and crack fillings are almost pure schorl, but in some bluish-colourless cracks nearly clear dravite composition is measured. Inside tourmaline crystals often dark brown micashaped fabric or relicts can be traced with higher amount of Ti and Fe than the neighbouring parts. Between and inside tourmaline grains rutile and ilmenite can be observed with fine-grained scheelite inclusions. In quartz-rich zones apatite, white mica, albite, garnet, biotie, zircon and monazite are identified. White micas show high Si content (more than 6.8 apfu.), garnets are enriched in Ca. Deformed tourmaline-rich layers and veins can be observed in the kyanite-chloritemuscovite schists and quartzites as well.

Euhedral, zoned but deformed schorl-dravite tourmaline appears along kyanite, Mg-chlorite (leuchtenbergite) rutile and muscovite, the latter contains sillimanite needles as relict phases. A narrow colourless tourmaline rim enriched exclusively in Mg (FeO content is below 1%) can be identified. White mica is also enriched in Mg (#Mg>0,9). INTERPRETATION Coarse-grained pegmatitic orthogneisses and metagranites with a significant amount of primary tourmaline-beryl assemblage indicates that a fluid-rich (B, Be) environment prevailed during the final crystallization of the Variscan peraluminious leucogranite. High Si white mica (phengite) rims and Ca-enrichment in garnet rims show HP Alpine metamorphic overprint. REE-rich phosphate mineralisation, kyanite and leuchtenbergite in the orthogneisses imply that high salinity fluids metasomatized the gneisses along the preexisting shear zones after the Alpine peak metamorphic event. Tourmaline grains in the sheared veins and layers in kyanite-bearing quartzites and schists may originate from a micaschist underwent a strong Mgmetasomatism (clear dravite rims) during the formation of leucophyllites. Based on textural data (e.g. polygonal fabric, phosphate mineralisation in the cracks), garnet and white mica chemistry, tourmalinites seem to have undergone the Alpine metamorphism, therefore the formation of the rock is pre-dating this event. The following scenario seems to be a possible explanation for the formation of this rock. During the crystallization of the granitoids a B-rich fluid could exsolve from the residual melt causing tourmalinisation in the country rocks (London et al. 1996). Due to the escaped tourmalinizing fluids, rock forming minerals are partially or completely replaced (Henry and Dutrow 1996). As tourmalinites are closely associated with Sn-W greisen deposits, where scheelite is a main ore mineral (e.g.: Slack 1996), so relict scheelite in rutile in the described samples may indicate a similiar environment of formation. According to these, tourmalinites may have formed by contact boron metasomatism related to fluids from the Variscan granitoid intrusion, then underwent the Alpine metamorphic event, therefore the described metamorphic fabric and assemblage can be observed.



London, D., et al. (1996): Boron in granitic rocks and their contact aureoles. Rev. Mineral. Geochem., 33(1):299–330. Slack, J.F. (1996): Tourmaline associations with hydrothermal ore deposits. Rev. Mineral. Geochem., 33(1):559–643. Török, K. (1996): High-pressure/low temperature metamorphism of the Kő-hegy gneiss, Sopron (W-Hungary); Phengite barometry and fluid inclusions. – Eur. J. Miner., 8:917–925. Török, K (1998): Magmatic and high-pressure metamorphic development of orthogneisses in the Sopron area, Eastern Alps (WHungary). N. Jb, Miner. Abh., 173:63–91. Török, K. (2001): Multiple fluid migration events in the Sopron Gneisses during the Alpine high-pressure metamorphism, as recorded by bulk-rock and mineral chemistry and fluid inclusions. N. Jb. Miner. Abh., (177):1–36.

Balogh, K. and Dunkl, I. (2005): Argon and fission track dating of Alpine metamorphism and basement exhumation in the Sopron Mts. (Eastern Alps, Hungary): thermocronology or mineral growth? – Mineral. Petrol., 83:191– 218. Henry, D.J., and Dutrow, B.L. (1996): Metamorphic tourmaline and its petrologic applications. Rev. Mineral. Geochem., 33 (1): 503–557. Kisházi, P. and Ivancsics, J. (1987): Petrogenesis of the Sopron Gneiss Formation. Földtani Közlöny 119:153–166. (in Hungarian with English abstract). Lelkes-Felvári, Gy., et al. (1984): Pre-Alpine and Alpine developments of the Austridic basement in the Sopron area (Eastern Alps, Hungary). Rend. Soc. It. Min. Petr., 39:593– 612.


Tourmaline in graphitic material as a potential provenience indicator in archaeology 1

1 2

Vladimír Šrein#, 1Petr Bohdálek, 2Jan Blažek, 2Marek Půlpán

Česká geologická služba (Czech Geological Survey), Prague, Czech Republic, # [email protected] cz Ústav archeologické a památkové péče severních Čech v.v.i. (Department of archaeological conservation in Northern Bohemia), Most, Czech Republic

Key words: tourmaline, graphite, archaeology, Bronze Age


brecciated/crushed mix of graphitic rock, quartz, plagioclase, and mica. Mineralogical examination of main rockforming minerals compositions revealed presence of minute zoned V-bearing tourmaline crystals in mica and feldspar matrix. So far known graphite-bearing materials from the Czech Republic show rather different petrographic characteristics and distinctly different composition of tourmaline (e.g. Litochleb et al. 1983; Losos and Selway 1998; Cempírek et al. 2013). The observed mineral assemblage and especially tourmaline composition show a strong similarity to the tourmaline-bearing graphite rocks from the locality Amstall in Austria (Ertl et al. 2008).

The archaeological location Místo in the Ore Mountains / Erzgebirge (dist. Chomutov, Ústí Region, North Bohemia) was discovered during surface prospection in the 1960's. Based on results of minor excavations conducted in following decades it was interpreted as a mountain settlement from the Final Bronze Age. We present the main results of comprehensive excavations that took place on the site in 2013 in the framework of the international project ArchaeoMontan. This research has not only significantly expanded the source base (e.g., with dozens of metal and graphite artefacts), but has also shown longer and more intensive use of the site from a Late through the Final Bronze Age to the Hallstatt period. Finally, considered are possible specific functions of this unusual location, which is situated on the steep slopes, high up (altitude > 600 m) in the Ore Mountains, and is therefore completely outside the traditional prehistoric settlement oecumene.

REFERENCES Cempírek, J., et al. (2013): Crystal structure and compositional evolution of vanadiumrich oxy-dravite from graphite quartzite at Bítovánky, Czech Republic. Journ. Geosci., 58:149–162. Ertl, A., et al. (2008): V3+-bearing, Mg-rich, strongly disordered olenite from a graphite deposit near Amstall, Lower Austria: A structural, chemical and spectroscopic investigation. Neues Jb. Min., Abh., 184, 3: 243–253. Litochleb, J., et al. (1983): Goldmanit ze Struhadla u Klatov. Časopis pro mineralogii a geologii 28:103–104. Losos, Z. and Selway, J.B. (1998): Tourmaline of dravite-uvite series in graphitic rocks of the Velké Vrbno Group (Silesicum, Czech Republic). Journ. Czech Geol. Soc., 43:45– 52.

GRAPHITE MATERIAL During archeological research at locations with metallurgical indications, we found graphitecontaining fragments along with bronze artefacts. Detailed examination of the graphite fragments showed two different materials. Larger fragment is relatively compact, solid piece of graphite gneiss with weakly visible foliation. The second fragment formed by secondary consolidation/cementation of a


Tourmaline from Věžná I pegmatite; an indicator of external contamination and internal fractionation Jiří Toman#, 1Milan Novák, 1Jan Loun, 1Jan Cempírek


Department of Geological Sciences, Faculty of Science, Masaryk University, Kotlářská 2, 611 37 Brno, Czech Republic, #[email protected] 2 Department of Mineralogy and Petrography, Moravian Museum, Zelný trh 6, 659 37, Brno, Czech Republic 1

Key words: tourmaline, contamination, fractionation, pegmatite, serpentinite, Věžná I


most Mg-rich tourmalines also show elevated contents of Ca. Apparent enrichment in Mg is evident also in secondary dravite (f) after cordierite (Fig. 1); however, XMg is the same in both minerals. Manganese and F increase steadily from primitive black tourmalines in graphic (a) and intermediate unit (b) via black tourmaline close to albite-pollucite unit (d) to elbaite in rims around black schorl (c) and individual solitary grains of fluor-elbaite (e) in albite-pollucite unit (Fig. 1).

Highly evolved granitic pegmatites exhibit extreme geochemical fractionation within a suite of granitic rocks. However, due to small size and long propagation of low-volume melt from their sources through rock complexes pegmatites locally show strong external contamination from host rocks (Černý et al. 2012; Novák et al. 2017). Typical elements illustrating high fractionation and contamination are Li, Cs, Mn, Ta, F and Mg (Ca), respectively, most of them compatible with tourmaline. The zoned Věžná I pegmatite cutting serpentinite is an excellent example of the pegmatite with both high degree of fractionation (elbaite, polylithionite, pollucite, sokolovaite, topaz) and strong contamination (phlogopite, cordierite, dravite), and, moreover, tourmaline is a common accessory mineral in most textural-paragenetic units. Consequently, it is an ideal indicator of both processes – geochemical fractionation and external contamination.

CONCLUSION Chemical compositions of tourmaline vary from Mg-rich in outer units to Mn, F-enriched and mostly Mg-free elbaite to fluor-elbaite from the most evolved albite-pollucite unit and from rare rims of pink elbaite around large crystal of schorl (b) in intermediate blocky unit. These suggest that external contamination from host serpentinite acted before and during pegmatite melt emplacement. Almost immediately after emplacement the pegmatite behaved as a close system and Mg was exhausted from the system during crystallization phologopite → cordierite → tourmaline. Minor amounts of Mg found in some analyses of elbaite (e, c) suggest that pegmatite became open in subsolidus during crystallization of the latest primary tourmalines and other minerals. This process may be related to strong kerolitization of quartz associated with recrystallization of oligoclase to adularia, pectolite, albite and zeolites (Dosbaba and Novák 2012). However, subsolidus Mg-contamination likely proceeded in several stages and requires more thorough study.


Several textural, paragenetic and compositional types of tourmaline were distinguished (Tab. 1). Chemical compositions of the individual types of tourmaline (Fig. 1) show apparent differences. The highest concentrations of Mg are typically found in tourmaline from outer units (a,b) and then decrease steadily to both types of elbaite. Elbaite in overgrowths of black tourmaline from intermediate unit is mostly Mg-free but some analyses of tourmaline from albitepollucite unit are Mg-enriched (Fig. 1). The



Table 1. Paragenetic, textural and compositional types of tourmaline.


REFERENCES Černý, P., et al. (2012): Granitic pegmatites as reflections of their sources. Elements, 8:289– 294. Dosbaba, M. and Novák M. (2012): Quartz replacement by "kerolite" in graphic quartzfeldspar intergrowths from the Věžná I pegmatite, Czech Republic; A complex desilicification process related to episyenitization. Canad. Mineral., 50, 6: 1609–1622. Novák, M. (1998): Fibrous blue dravite; an indicator of fluid composition during subsolidus replacement processes in Li-poor granitic pegmatites in the Moldanubicum, Czech Republic. Journ. Czech Geol. Soc., 43:24-30. Novák, M., et al. (2017): Tourmaline, an indicator of external Mg-contamination of granitic pegmatites from host serpentinite; examples from the Moldanubian Zone, Czech Republic. Mineral. Petrol., 96, in print.



Figure 1. Chemical composition of tourmaline. a) Na+K-Ca-X-site vacancy; b) Al-Fe+Mn-Mg; c) F/Mg; d) F/Mn; symbols: see the chapter RESULTS and the table 1.


Tourmaline applications to ore systems 1


Robert B. Trumbull#

GFZ German Research Centre for Geosciences, Telegrafenberg, 14473 Potsdam, Germany, [email protected]


Key words: B-isotopes, trace elements, exploration, fluid evolution


what still holds true: tourmaline is a common and locally abundant gangue mineral in four main classes of ore deposits: (1) stratiform VMS and SEDEX deposits, (2) porphyry-type Cu-Mo-Au deposits in continental arcs, (3) metamorphic-hosted "orogenic" Au deposits and (4) granite-related Sn-W deposits and rareelement pegmatites. Why these settings are commonly rich in boron is not entirely clear. It is likely that a first stage of B-enrichment in clay-rich marine sediments and seawateraltered oceanic crust, with or without evaporite deposits, was relevant for the stratiform and porphyry deposits. The advanced level of differentiation reached by parental granites for Sn-W deposits and pegmatites may explain high B contents in residual melts and fluids although London et al. (1996) argued for preenrichment in the magma source. Orogenic Au deposits are commonly associated with major shear zones and tourmaline enrichment in these cases may relate more to an effective focus of crustal fluids than to any specific protolith.

Tourmaline is found as a gangue mineral in many hydrothermal and magmatic ore deposits, and in places tourmaline itself is mined as gemstock. Ideally, the geochemical, isotopic and textural diversity of tourmaline in a given ore deposit can constrain important ore-forming conditions and processes. In magmatic systems like granites and pegmatites, applications typically relate to magma provenance, open- vs. closed-system fractionation and the development of a separate fluid phase. In hydrothermal deposits the applications include determining the fluid provenance and evidence for fluid mixing, redox state and boiling. The tendency of tourmaline to form and show compositional zoning and multiple episodes of growth gives a record of how conditions changed with time. On a more pragmatic level, the composition of tourmaline may prove useful in mineral exploration as a pathfinder for ore. Thus, there are ample reasons for economic geologists to study tourmaline. Since the extensive review of tourmaline in oreforming systems by Slack (1996) and a summary of more recent developments by Slack and Trumbull (2011) the number of studies has continued to increase. This contribution highlights some of the progress in this field, which relates especially to the application of microbeam techniques for insitu analysis of trace elements and boron isotopes by LA-ICPMS and SIMS.


One of the hallmarks of tourmaline is its extreme chemical diversity, which can be useful as an indicator of source (e.g. Van Hinsberg et al. 2011). However, in hydrothermal systems it must be kept in mind that some, if not most of the chemical variations in tourmaline may be controlled by the local host rock rather than the mineralizing fluid, which complicates using tourmaline geochemistry as an exploration tool (Slack 1996). Trace-elements in tourmaline have been evaluated in this context in several studies (e.g., Griffin 1996; Jiang et al. 2004; Yavuz et al. 2011; Iveson et al. 2016; Duchoslav et al.


Slack (1996) provided an extensive review of tourmaline in ore associations and pointed out


2016). In specific cases, ore-metal concentrations (e.g., Cu, Zn, Sn) in tourmaline from mineralized zones are higher than in barren rocks and of potential exploration significance. But these studies also found highly variable trace element contents and only limited inter-element correlations that complicate the interpretation and call for more systematic work on element partitioning. Another application of tourmaline is for a redox indicator (Williamson et al. 2010; Baksheev et al. 2012). Tourmaline commonly shows compositional zoning and other expressions of multistage growth (mosaic textures, overgrowths, veining). These features are important because ore formation is typically multi-stage and commonly triggered by a change in physicalchemical conditions (e.g., redox, boiling, fluid mixing) that may be recorded in the tourmaline zonation. A good example among many is the study by Mlynarczyk and Williams-Jones (2006) of zoned tourmaline zoning in the San Rafael Cu-Au porphyry deposit, Peru. There, abundant early tourmaline formed in barren pre-ore veins whereas Fe-rich tourmaline fringes and rims coincide with cassiterite precipitation and with a possible switch from lithostatic to hydrostatic conditions (see also Williamson et al. 2010; Dutrow and Henry 2000).

classic example where B-isotopes cast new light on ore provenance is the discovery of highly negative δ11B values (ca. -20 ‰ in tourmaline from the Broken Hill stratiform deposits that rule out a marine source of boron (Slack et al. 1989). At the other extreme, Xavier et al. (2008) found tourmaline δ11B values from +14 to +26.5 ‰ that confirmed involvement of marine brines in the formation of iron oxide-copper-gold (IOGC) deposits in Brazil. B-isotope variations are sensitive to the temperature of mineralization and to changes in fluid composition caused by crystallization of B-phases, phase transitions such as fluid exsolution or boiling, and to fluid mixing. The isotopic signal of tourmaline is not always easy to interpret, nowhere more so than in zoned rare-element pegmatites (e.g. Siegel et al. 2016; Trumbull et al. 2013; Maner and London 2017). Each case needs individual scrutiny but there are many examples where the effort has paid off. It is important to stress that boron isotope analysis should not be a stand-alone method. One area that remains to be exploited in the future is the combined insitu isotope analyses of B, O and H, all of which can be analysed by SIMS but have rarely been combined (Xavier et al. 2013).


Baksheev, I.A., et al. (2012): Tourmaline as a prospecting guide for porphyry-style deposits. Eur. J. Mineral., 24:957–979. Drivenes, K., et al., (2015): Late‑magmatic immiscibility during batholith formation: assessment of B isotopes and trace elements in tourmaline from Land’s End Granite, SW England. Contrib. Mineral. Petrol., 169, 56. Duchoslav, M., et al., (2017) Changes in tourmaline composition during magmatic and hydrothermal processes leading to tin-ore deposition: The Cornubian Batholith, SW England. Ore Geol. Rev., 83:215–234. Dutrow, B.L. and Henry, D.J. (2000): Complexly zoned fibrous tourmaline, Cruzeiro mine, Minas Gerais, Brazil: a record of evolving magmatic and hydrothermal fluids. Canad. Mineral., 38:131–143. Griffin, W.L., et al. (1996): Trace elements in tourmalines from massive sulfide deposits and tourmalinites: geochemical controls and exploration applications. Econ. Geol., 91: 657–675. van Hinsberg, V.J., et al. (2011): Tourmaline as a petrologic forensic mineral: a unique



For most B-isotope studies, the presence of compositional zoning makes in-situ analysis useful if not necessary. B-isotopic zoning is not rare (e.g. Pal et al. 2010; Su et al. 2016) and should always be checked for, but in fact many studies find little or no variation of Bisotope ratios even in chemically stronglyzoned crystals (Krienitz et al. 2008; Drivenes et al. 2015, Zheng et al. 2016). This, and the common lack of correlation between isotopic and elemental composition in tourmaline, reflects the fact that boron is typically a fluid component whereas other elements may derive from the local host rocks. One of the most common applications of Bisotope studies in ore research is to constrain the fluid or magma source and its evolution. Source discrimination relies on the very distinctive positive δ11B values for marine reservoirs (e.g., +40 ‰ in seawater) in contrast to distinctly negative values for the crust and mantle (see Marshall and Foster 2017). A


recorder of its geologic past. Elements, 7: 327–332. Iveson, A.A., et al., (2016): Magmatic– hydrothermal fluids and volatile metals in the Spirit Lake pluton and Margaret Cu–Mo porphyry system, SW Washington, USA. Contrib. Mineral. Petrol., 171, 20. Jiang, S.-Y., et al. (2004): Trace and rare-earth element geochemistry in tourmaline and cassiterite from the Yunlong tin deposit, Yunnan, China: implication for migmatitic– hydrothermal fluid evolution and ore genesis. Chem. Geol., 209:193–213. Krienitz, M.-S., et al., (2008): Hydrothermal gold mineralization at the Hira Buddini Gold Mine, India: contraints on fluid sources and evolution from boron isotopic compositions of tourmaline. Mineral. Dep., 43:421–434. London, D., et al. (1996): Boron in granitic rocks and their contact aureoles. In: Boron – mineralogy, petrology and geochemistry (E.S. Grew and L.M. Anovitz, eds.). Rev. Mineral. Geochem., 33:299–330. Maner, J.L. and London, D. (2017): The boron isotope evolution of the Little Three pegmatites, Ramona, CA. Chem. Geol., 460: 70–83. Marschall, H.R. and Foster, G.L., eds. (2017) Boron Isotopes - The Fifth Element. Advances in Geochemistry, Springer-Verlag, Heidelberg, in press. Mlynarczyk, M.S.J. and Williams-Jones A.E. (2006): Zoned tourmaline associated with cassiterite: Implications for fluid evolution and tin mineralization in the San Rafael Sn– Cu deposit, southeastern Peru. Canad. Mineral., 44:347–365 Pal, D.C., et al. (2010): Chemical and boron isotope compositions of tourmaline from the Jaduguda U (-Cu-Fe) deposit, Singhbhum shear zone, India: implications for the source and evolution of the mineralizing fluid. Chem. Geol., 227:245–260. Siegel, K., et al. (2016): Stable isotope (B,H,O) and mineral-chemistry constraints on the magmatic to hydrothermal evolution of the Varuträsk rare-element pegmatite (northern Sweden). Chem. Geol., 421:1–16. Slack, J.F. (1996): Tourmaline associations with hydrothermal ore deposits. In: Boron – mineralogy, petrology and geochemistry (E.S. Grew and L.M. Anovitz, eds.). Rev. Mineral. Geochem., 33:559–644.

Slack, J.F. and Trumbull, R.B (2011): Tourmaline as a recorder of ore-forming processes. Elements, 7:321–326. Slack, J.F., et al. (1989): Boron isotope evidence for the involvement of non-marine evaporites in the origin of Broken Hill ore deposits. Nature, 342:913–916. Su, Z.-K. et al., (2016): Using elemental and boron isotopic compositions of tourmaline to trace fluid evolutions of IOCG systems: the world class Dahongshan Fe-Cu deposit in SW China. Chem. Geol., 441:265–279 Trumbull, R.B., et al., (2011): Fluid sources and metallogenesis in the Blackbird Co-CuAu-Bi-Y-REE district, USA: constraints from chemical and boron isotope variations in tourmaline. Canad. Mineral., 49:225–244. Trumbull, R.B., et al. (2013): The diversity of B-isotope variations in tourmaline from rareelement pegmatites in the Borborema Province of Brazil. Chem. Geol., 352:47–62. Yavuz, F., et al., (2011): Trace-element, rareearth element and boron isotopic compositions of tourmaline from a vein-type Pb–Zn–Cu–U deposit, NE Turkey. Intern. Geol. Rev., 53:1–24. Williamson, B.J., et al. (2010): Geochemical constraints from zoned hydrothermal tourmalines on fluid evolution and Sn mineralization: an example from fault breccias at Roche, SW England. J. Petrol., 41:1439–1453. Xavier, R.P., et al. (2008): Tourmaline Bisotopes fingerprint marine evaporites as the source of high salinity ore fluids in iron oxide-copper-gold deposits, Carajás Mineral Province (Brazil). Geology, 36:743–746. Xavier, R.P., et al. (2013) Sources of mineralizing fluids in Cu-Au systems from the Carajás Mineral Province (Brazil): constraints from in-situ microanalysis of hydrogen and boron isotopes in tourmaline. 12th SGA Biennial Meeting, Uppsala. Zheng, Z., et al., (2016): Fluid sources and metallogenesis in the Baiganhu W-Sn deposit, east Kunlun, NW China: insights from chemical and boron isotopic compositions of tourmaline. Ore Geol. Rev., 72:1129–1142.


Tourmalines of the siderite-quartz-sulphide hydrothermal veins, Gemeric Unit, Western Carpathians, Slovakia: crystal chemistry and genetic aspects 1

Pavel Uher#, 1Peter Bačík, 1Jakub Dikej, 2Tomáš Vaculovič

Department of Mineralogy and Petrology, Faculty of Natural Sciences, Comenius University, Ilkovičova 6, 842 15 Bratislava, Slovakia, [email protected] 2 Department of Chemistry, Faculty of Science, Masaryk University, Kotlářská 2, 611 37 Brno, Czech Republic 1

Key words: tourmaline, hydrothermal veins, crystal chemistry, trace elements, substitutions, compositional trends, Gemeric unit, Western Carpathians


tourmaline crystals are relatively homogeneous or they show concentric zoning. The compositions belong mostly to the alkali group, schorl to dravite series, rarely dominant X-site vacant foititic tourmaline (Vlachovo and Bindt) or oxy-dravite compositions (Hnilčík) were detected (Bačík et al. 2017a). Generally, the both Fe/(Fe+Mg) and X-vacancy/(X-vacancy+Na) ratios of the hydrothermal vein tourmalines show wide variations: 0.250.75 and 0.050.55, respectively. In some cases, the rim zones of schorlitic tourmalines show high concentrations of Ti (up to 2.35 wt.% TiO2, 0.30 apfu; Rožňavské Bystré). The tourmaline chemical composition is mostly controlled by the alkali- and protondeficient substitutions: XAlNa-1(Mg,Fe2+)-1 and AlO(Mg,Fe2+)-1(OH)-1, respectively. Titanium is incorporated into the tourmaline structure by several possible mechanisms, such Y Y Y Z as Ti (Mg,Fe)YAl-2, Ti MgYAl-1ZAl-1, Y TiO(YAlOH) or XCaYTiZMgO2X□-1Y,ZAl-2(OH)-2 (Bačík et al. 2017a). Mössbauer spectroscopy of the tourmaline has shown variable Fe3+ proportions (926 % of total Fe). The concentrations of trace elements in the studied tourmalines determined by LA-ICP-MS show following variations (in ppm): Li (410), Sc (6100), Ti (4002400), V (280720), Cr (5100), Co (5130), Ni (30140), Zn (40140), Ga (4070), and Sr (40570). Contents of other trace elements, including Be, Cu, Ge, Zr, In, Sn, and Pb were very low (