Jan Cempírek

Crystal chemistry and origin of grandidierite, ominelite, boralsilite, and werdingite from the Bory Granulite Massif, Czech Republic

Abstract A mineral assemblage involving grandidierite, ominelite, boralsilite, werdingite, dumortierite (locally Sb,Ti-rich), tourmaline, and corundum, along with the matrix minerals K-feldspar, quartz, and plagioclase, was found in a veinlet cutting leucocratic granulite at Horní Bory, Bory Granulite Massif, Moldanubian Zone of the Bohemian Massif. Zoned crystals of primary grandidierite to ominelite enclosed in quartz are locally overgrown by prismatic crystals of boralsilite and Fe-rich werdingite. Boralsilite also occurs as separate cross-shaped plumose aggregates with Fe-rich werdingite in quartz. Grandidierite is commonly rimmed by a narrow zone of secondary tourmaline or is partially replaced by the assemblage tourmaline + corundum ± hercynite. Grandidierite (XFe = 0.34-0.71) exhibits dominant FeMg-1 substitution and elevated contents of Li (120-1890 ppm). Boralsilite formula ranges from Al15.97B6.20Si1.80O37 to Al15.65B5.29Si2.71O37 and the formula of werdingite ranges from (Fe,Mg)1.44Al14.61B4.00Si3.80O37 to (Fe,Mg)1.22Al14.86B4.25Si3.55O37. Dumortierite and Sb,Ti-rich dumortierite occur as zoned crystals with zones poor in minor elements (≤0.12 apfu Fe+Mg) and zones enriched in Sb (≤0.46 apfu) and Ti (≤0.25 apfu). Secondary tourmaline (XFe = 0.44-0.75) of the schorlmagnesiofotite- foitite-olenite solid solution occurs as a replacement product of grandidierite, rarely boralsilite. Other accessory minerals in the veinlet include monazite-(Ce), ilmenite, rutile, ferberite, srilankite, löllingite, arsenopyrite, and apatite. Formation of the borosilicate-bearing veinlet post-dates the development of foliation in the host granulite and is related to the decompressional process. The assemblage most probably originated from a H2O-poor system at T ~ 750 °C and P ~ 6-8 kbar. Textural relations as well as geological position of the borosilicate veinlet suggest that it represents the earliest intrusion related to pegmatites in the Bory Granulite Massif. Younger granitic pegmatites in the area are characterized by high contents of B, Al, P, Fe, and minor concentrations of W, Ti, Zr, Sc, and Sb. All pegmatite types probably formed within a short time period of ~5 Ma.

Oxy-schorl, Na(Fe2+2Al)Al6Si6O18(BO3)3(OH)3O, a new mineral from Zlatá Idka, Slovak Republic and Přibyslavice, Czech Republic

Abstract Oxy-schorl (IMA 2011-011), ideally Na(Fe22+Al)Al6Si6O18(BO3)3(OH)3O, a new mineral species of the tourmaline supergroup, is described. In Zlatá Idka, Slovak Republic (type locality), fan-shaped aggregates of greenish black acicular crystals ranging up to 2 cm in size, forming aggregates up to 3.5 cm thick were found in extensively metasomatically altered metarhyolite pyroclastics with Qtz+Ab+Ms. In Přibyslavice, Czech Republic (co-type locality), abundant brownish black subhedral, columnar crystals of oxy-schorl, up to 1 cm in size, arranged in thin layers, or irregular clusters up to 5 cm in diameter, occur in a foliated muscovite-tourmaline orthogneiss associated with Kfs+Ab+Qtz+Ms+Bt+Grt. Oxy-schorl from both localities has a Mohs hardness of 7 with no observable cleavage and parting. The measured and calculated densities are 3.17(2) and 3.208 g/cm3 (Zlatá Idka) and 3.19(1) and 3.198 g/cm3 (Přibyslavice), respectively. In plane-polarized light, oxy-schorl is pleochroic; O = green to bluish-green, E = pale yellowish to nearly colorless (Zlatá Idka) and O = dark grayish-green, E = pale brown (Přibyslavice), uniaxial negative, ω = 1.663(2), ε = 1.641(2) (Zlatá Idka) and ω = 1.662(2), ε = 1.637(2) (Přibyslavice). Oxy-schorl is trigonal, space group R3m, Z = 3, a = 15.916(3) Å, c = 7.107(1) Å, V = 1559.1(4) Å3 (Zlatá Idka) and a = 15.985(1) Å, c = 7.154(1) Å, V = 1583.1(2) Å3 (Přibyslavice). The composition (average of 5 electron microprobe analyses from Zlatá Idka and 5 from Přibyslavice) is (in wt%): SiO2 33.85 (34.57), TiO2 <0.05 (0.72), Al2O3 39.08 (33.55), Fe2O3 not determined (0.61), FeO 11.59 (13.07), MnO <0.06 (0.10), MgO 0.04 (0.74), CaO 0.30 (0.09), Na2O 1.67 (1.76), K2O <0.02 (0.03), F 0.26 (0.56), Cl 0.01 (<0.01), B2O3 (calc.) 10.39 (10.11), H2O (from the crystal-structure refinement) 2.92 (2.72), sum 99.29 (98.41) for Zlatá Idka and Přibyslavice (in parentheses). A combination of EMPA, Mössbauer spectroscopy, and crystal-structure refinement yields empirical formulas (Na0.591Ca0.103□0.306)Σ1.000(Al1.885Fe2+ 1.108Mn0.005Ti0.002)Σ3.000(Al5.428Mg0.572)Σ6.000(Si5.506Al0.494)Σ6.000O18 (BO3)3(OH)3(O0.625OH0.236F0.136Cl0.003)Σ1.000 for Zlatá Idka, and (Na0.586Ca0.017K0.006□0.391)Σ1.000(Fe2+1.879Mn0.015 Al1.013Ti0.093)Σ3.00(Al5.732Mg0.190Fe3+0.078)Σ6.000(Si5.944Al0.056)Σ6.000O18(BO3)3(OH)3(O0.579F0.307OH0.115)Σ1000 for Přibyslavice. Oxy-schorl is derived from schorl end-member by the AlOFe-1(OH)-1 substitution. The studied crystals of oxy-schorl represent two distinct ordering mechanisms: disorder of R2+ and R3+ cations in octahedral sites and all O ordered in the W site (Zlatá Idka), and R2+ and R3+ cations ordered in the Y and Z sites and O disordered in the V and W sites (Přibyslavice).

Complexly zoned niobian titanite from hedenbergite skarn at Písek, Czech Republic, constrained by substitutions Al(Nb,Ta)Ti-2, Al(F,OH)(TiO)-1 and SnTi-1

Abstract Euhedral crystals ofcomplexly zoned niobian titanite (up to 0.3 mm) are enclosed in hedenbergite (Hd53-81Di15-43Jh3-5) and quartz from a hedenbergite vein skarn at Kamenné doly near Písek, Czech Republic. They are associated with minor clinozoisite-epidote (Ps3-22), calcite, plagioclase (An95), scapolite (Me80-82), scheelite, pyrrhotite, fluorapatite, arsenopyrite, native bismuth and Bi,Te-minerals. The following textural and compositional subtypes were recognized: (I) Nb-rich titanite, (II) Nb-moderate titanite in the central zone, (III) Nb-poor, Sn-enriched titanite and (IV) Nb-poor, Al,F-rich titanite in the outer zone. The substitution Al(Nb,Ta)Ti-2 is dominant in subtypes I and II, the titanite subtype I being characterized by elevated contents of Al ≤0.257 atoms per formula unit (a.p.f.u.), Nb (≤0.161 a.p.f.u.) and Ta (≤0.037 a.p.f.u.). Amounts of Al, Nb and Ta in subtype II are smaller and more variable. The minor substitution SnTi-1 occurs chiefly in titanite subtype III with a content of Sn ≤0.039 a.p.f.u.. The substitution Al(F,OH)(TiO)-1 is typical for titanite subtype IV exhibiting elevated contents of Al (≤0.221 a.p.f.u.), F (≤0.196 a.p.f.u.) and Fe (≤0.039 a.p.f.u.). The negative relationship of substitutions Al(F,OH)(TiO)-1 vs. SnTi-1 and Al(Nb,Ta)Ti-2 is constrained chiefly by crystal structure rather than by the composition of parent medium alone. Textural relations suggest that the Nb-moderate titanite in the core zone and entire outer zone are products off luids-induced dissolution-reprecipitation processes. The studied niobian titanite represents a new F-enriched type from a medium-grade, calc-silicate rock.

Fe-rich and As-bearing vesuvianite and wiluite from Kozlov, Czech Republic

Abstract Green vesuvianite crystals occur with garnet and calcite in a hand specimen from the Nedvědice marble near Kozlov (near Štěpánov nad Svratkou, Svratka Crystalline Complex) in the Czech Republic. The average electron microprobe composition of the vesuvianite shows 12.10 wt% Fe2O3 (4.66 Fe pfu), 2.77 wt% B2O3 (2.45 B pfu), 1.71 wt% As2O5 (0.46 As pfu), and 1.40 wt% F (2.26 F pfu). The Fe concentration is the highest ever recorded for a vesuvianite-group mineral. The boron contents are extremely variable and two of the five compositions show more than the 2.50 B pfu needed for wiluite, and the average is only slightly less than this. The crystal structure [a = 15.7250(4), c = 11.7736(3) Å was refined in space group P4/nnc to an R1 value of 0.0221. The site refinement and Mössbauer spectroscopy results show Fe2+ substituting for Ca at the X3 site and filling the Y1 position, and Fe3+ substituting for Al at the Y3 position. Most of the Fe (70% from the site refinements and 78% from the Mössbauer interpretation) is ferric. The main effect of the high-Fe concentration is to increase the mean Y3-O distance to an unusually large 2.018 Å. Boron occurs at the T1 site, where it is coordinated by oxygen atoms at two O7B and two O11 positions, and at the T2 sites where it is coordinated by O atoms at one O10 and two O12A sites. When the nearby X3 site contains Fe, the T2 position is either vacant or [3]-coordinated by some combination involving an O10 site and two O12B positions, in which case the B atom is likely offset from the T2 site to reduce the B-O12B distance. Fluorine and OH occupy the O11 positions when there is a vacancy at the adjacent T1 position. Pentavalent As substitutes for Si at the Z2 site and Al at the Y2 site. The P4/nnc symmetry indicates that this vesuvianite formed at high temperatures (400-800 °C) and the predominance of Fe3+ and As5+ suggests under oxidizing conditions. The results showing Fe at three different sites with three different coordinations attests to the flexibility of the vesuvianite crystal structure. The incorporation of As at two different sites in the structure shows that rock-forming silicate minerals such as vesuvianite can be a reservoir for this heavy element

Lucchesiite, CaFe2+3Al6(Si6O18)(BO3)3(OH)3O, a new mineral species of the tourmaline supergroup

Abstract Lucchesiite, CaFe2+3 Al6(Si6O18)(BO3)3(OH)3O, is a new mineral of the tourmaline supergroup. It occurs in the Ratnapura District, Sri Lanka (6°35′N, 80°35′E), most probably from pegmatites and in Mirošov near Strážek, western Moravia, Czech Republic, (49°27′49.38″N, 16°9′54.34″E) in anatectic pegmatite contaminated by host calc-silicate rock. Crystals are black with a vitreous lustre, conchoidal fracture and grey streak. Lucchesiite has a Mohs hardness of ∼7 and a calculated density of 3.209 g/cm3 (Sri Lanka) to 3.243 g/cm3 (Czech Republic). In plane-polarized light, lucchesiite is pleochroic (O = very dark brown and E = light brown) and uniaxial (-). Lucchesiite is rhombohedral, space group R3m, a ≈ 16.00 Å, c ≈ 7.21 Å, V ≈ 1599.9 Å3, Z = 3. The crystal structure of lucchesiite was refined to R1 ≈ 1.5% using ∼2000 unique reflections collected with MoKα X-ray intensity data. Crystal-chemical analysis for the Sri Lanka (holotype) and Czech Republic (cotype) samples resulted in the empirical formulae, respectively: X(Ca0.69Na0.30K0.02)Σ1.01Y(Fe2+1.44Mg0.72Al0.48Ti4+0.33V3+0.02Mn0.01Zn0.01)Σ3.00Z(Al4.74Mg1.01Fe3+0.25)Σ6.00 [T(Si5.85Al0.15)Σ6.00O18](BO3)3V(OH)3W[O0.69F0.24(OH)0.07]Σ1.00 and X(Ca0.49Na0.45□0.05K0.01)Σ1.00Y(Fe2+1.14Fe3+0.95Mg0.42Al0.37Mn0.03Ti4+0.08Zn0.01)Σ3.00Z(Al5.11Fe3+0.38Mg0.52)Σ6.00[T(Si5.88Al0.12)Σ6.00O18] (BO3)3V[(OH)2.66O0.34]Σ3.00W(O0.94F0.06)Σ1.00. Lucchesiite is an oxy-species belonging to the calcic group of the tourmaline supergroup. The closest end-member composition of a valid tourmaline species is that of feruvite, to which lucchesiite is ideally related by the heterovalent coupled substitution ZAl3+ + O1O2- ↔ ZMg2+ + O1(OH)1-. The new mineral was approved by the International Mineralogical Association Commission on New Minerals, Nomenclature and Classification (IMA 2015-043).

Sc- and REE-rich tourmaline replaced by Sc-rich REE-bearing epidote-group mineral from the mixed (NYF+LCT) Kracovice pegmatite (Moldanubian Zone, Czech Republic)

Abstract Primary black thick-prismatic Al-rich schorl to rare fluor-schorl (TurP1), locally overgrown by brownish-green Li-rich fluor-schorl to fluor-elbaite (TurP2) from the Kracovice pegmatite (mixed NYF+LCT signature), was partly replaced by secondary Li-rich fluor-schorl to fluor-elbaite (TurS) plus the assemblage REE-bearing epidote-group mineral + chamosite. Primary Al-rich schorl (TurP1) shows high and variable contents of Sc (33-364 ppm) and Y+REE (40-458 ppm) with steep, LREEenriched REE pattern. Overgrowing (TurP2) and replacing (TurS) Li-rich fluor-schorl to fluor-elbaite is typically depleted in Sc (21-60 ppm) and Y+REE (3-47 ppm) with well-developed tetrad effect in the first (La-Nd) and the second (Sm-Gd) tetrads. Scandium- and REE-rich black tourmaline (TurP1) crystallized earlier from the melt, whereas crystallization of primary Li-rich fluor-schorl to fluor-elbaite (TurP2) most likely took place during late magmatic to early hydrothermal conditions. Both the secondary Li-rich fluor-schorl to fluor-elbaite (TurS) and the unusual assemblage of REE-bearing epidotegroup mineral + chamosite are likely coeval products of subsolidus reactions of the magmatic Al-rich schorl (TurP1) with evolved REE-poor, Li,F-rich, alkaline pegmatite-derived fluids. Well-crystalline REE-bearing epidote-group mineral (Y+REE = 0.42-0.60 apfu) confirmed by Raman spectroscopy has a steep, LREE-rich chondrite-normalized REE pattern with significant negative Eu anomaly and shows variable and high contents of Sc (≤3.3 wt% Sc2O3) and Sn (≤1.0 wt% SnO2). Substitution ScAl-1 and minor vacancy in the octahedral sites are suggested in the REE-bearing epidote-group mineral.

Allanite-(Nd), CaNdAl2Fe2+(SiO4)(Si2O7)O(OH), a new mineral from Åskagen, Sweden

Abstract Allanite-(Nd), ideally CaNdAl2Fe2+(SiO4)(Si2O7)O(OH), the Nd-analog of allanite-(Ce), occurs in the Åskagen pegmatite, central Sweden. It forms fine-grained aggregates within altered thalénite-(Y). The other associated minerals include: iimoriite-(Y), keiviite-(Y), allanite-(Y), and tengerite-(Y). Allanite-(Nd) is biaxial (-); with refractive indices α = 1.723(5), β = 1.754(7), γ = 1.772(5), and measured 2V = 82°(±3°). Larger fragments of allanite-(Nd) are moderately pleochroic (α = pale grayishbrown, γ = grayish-brown); small fragments are colorless. Allanite-(Nd) is monoclinic, space group P21/m, with the following unit-cell parameters: a = 8.8897(5), b = 5.7308(2), c = 10.1010(6) Å, β = 115.166(7)°, V = 465.75(4) Å3, Z = 2. The strongest five peaks from the X-ray powder diffraction patterns [d (Å)(I)(hkl)] are: 3.51(46)(2̅11), 2.89(100)(1̄13), 2.87(45)(020), 2.70(60)(120), 2.61(60)(3̅11). Allanite-(Nd) has a density of 3.98 g/cm3, vitreous luster, grayish-brown streak, and the Mohs hardness is ca. 6. It is brittle, with imperfect cleavage, and conchoidal fracture. The Mössbauer spectroscopy revealed Fe3+/Fetot ratio of 0.27 in the type sample. The refined empirical formula of the crystal used for the structure determination is: A1(Ca0.957Mn0.043)Σ1.000A2(Nd0.337Ca0.154Sm0.131Ce0.125Y0.112Gd0.062Pr0.051Dy0.016La0.007Er0.005)Σ1.000M1(Al0.860Fe3+0.140)Σ1.000M2AlM3(Fe2+0.760Al0.112Fe3+0.070Mg0.058)Σ1.000T1,2,3Si3O11.978(OH)1.032. The allanite-(Nd) was approved by CNMNC (IMA 2010-060).

Crystallographic control on lithium isotope fractionation in Archean to Cenozoic lithium-cesium-tantalum pegmatites

The age distribution of LCT pegmatites largely overlaps with major phases of collisional orogenic events and assembly of super-continents. Some of the largest known LCT pegmatite deposits formed in very short intervals, 2.7-2.5 and 1.9-1.8 billion years ago (Ga), corresponding to two major pulses of continental crust growth. However, the exact process of generation and segregation of large volumes of Li-bearing pegmatite liquids, perhaps involving disequilibrium fractional crystallization and leaving residual melts enriched in fluxing elements such as B, F, H2O, Li, and P, remains largely obscure. The new data on Li contents and isotope compositions in major mineral phases from temporally and geographically separated pegmatite bodies document extreme variations in d7Li values among individual large LCT pegmatites, in particular Archean occurrences. The observed >10‰ variations in d7Li values for the same mineral phases from different localities (i.e., beryl, petalite, spodumene, lepidolite, amblygonite, muscovite) contrast with globally homogeneous Li isotope systematics of major mineral phases from unmodified mantle rocks. Consistent Li isotope offsets between coexisting mineral phases are best explained by Li isotope fractionation as a function of the bond length between Li and neighboring ions (O, OH, F). We suggest that spatially distinct Li isotope patterns act as fingerprints for different pegmatites and can be explained by the pre-existing Li isotope differences of their crustal sources at the time of pegmatite formation owing to differences in crustal age and evolution. This would imply secular evolution of the continental crust over Earth history toward present-day globally broadly uniform crustal 7Li/6Li ratios (d7Li ~0‰). The differences among Archean occurrences could reflect possible Archean paleogeography and perhaps be linked with different thermal regimes of individual cratons as a consequence of variations in crustal thickness.

Sodium scandium diphosphate, NaScP2O7, isotypic with α-NaTi(III)P2O7

Crystals of the title compound, NaScP2O7, were grown by a flux method. The crystal structure is isotypic with those of α-NaTiP2O7, NaYbP2O7 and NaLuP2O7, and is closely related to that of NaYP2O7. The structural set-up consists of a three-dimensional framework of P2O7 units that are corner-shared by ScO6 octahedra, forming tunnels running parallel to [010]. The Na atoms are situated in the tunnels and are surrounded by nine O atoms in a distorted environment.

Vránaite, ideally Al16B4Si4O38, a new mineral related to boralsilite, Al16B6Si2O37, from the Manjaka pegmatite, Sahatany Valley, Madagascar

Abstract The system B2O3-Al2O3-SiO2 (BAS) includes two ternary phases occurring naturally, boromullite, Al9BSi2O19, and boralsilite, Al16B6Si2O37, as well as synthetic compounds structurally related to mullite. The new mineral vránaite, a third naturally occurring anhydrous ternary BAS phase, is found with albite and K-feldspar as a breakdown product of spodumene in the elbaite-subtype Manjaka granitic pegmatite, Sahatany Valley, Madagascar. Boralsilite also occurs in this association, although separately from vránaite; both minerals form rare aggregates of subparallel prisms up to 100 μm long. Optically, vránaite is biaxial (–), nα = 1.607(1), nβ = 1.634(1), nγ = 1.637(1) (white light), 2Vx(calc) = 36.4°, X ≈ c; Y ≈ a; Z = b. An averaged analysis by EMP and LA-ICP-MS (Li, Be) gives (wt%) SiO2 20.24, B2O3 11.73, Al2O3 64.77, BeO 1.03, MnO 0.01, FeO 0.13, Li2O 1.40, Sum 99.31. Raman spectroscopy in the 3000–4000 cm−1 region rules out the presence of significant OH or H2O. Vránaite is monoclinic, space group I2/m, a = 10.3832(12), b = 5.6682(7), c = 10.8228(12) Å, β = 90.106(11)°; V = 636.97(13) Å3, Z = 1. In the structure [R1 = 0.0416 for 550 Fo > 4σFo], chains of AlO6 octahedra run parallel to [010] and are cross-linked by Si2O7 disilicate groups, BO3 triangles, and clusters of AlO4 and two AlO5 polyhedra. Two Al positions with fivefold coordination, Al4 and Al5, are too close to one another to be occupied simultaneously; their refined site-occupancy factors are 54% and 20% occupancy, respectively. Al5 is fivefold-coordinated Al when the Al9 site and both O9 sites are occupied, a situation giving a reasonable structure model as it explains why occupancies of the Al5 and O9 sites are almost equal. Bond valence calculations for the Al4 site suggest Li is likely to be sited here, whereas Be is most probably at the Al5 site. One of the nine O sites is only 20% occupied; this O9 site completes the coordination of the Al5 site and is located at the fourth corner of what could be a partially occupied BO4 tetrahedron, in which case the B site is shifted out of the plane of the BO3 triangle. However, this shift remains an inference as we have no evidence for a split position of the B atom. If all sites were filled (Al4 and Al5 to 50%), the formula becomes Al16B4Si4O38, close to Li1.08Be0.47Fe0.02Al14.65B3.89Si3.88O36.62 calculated from the analyses assuming cations sum to 24. The compatibility index based on the Gladstone-Dale relationship is 0.001 (“superior”). Assemblages with vránaite and boralsilite are inferred to represent initial reaction products of a residual liquid rich in Li, Be, Na, K, and B during a pressure and chemical quench, but at low H2O activities due to early melt contamination by carbonate in the host rocks. The two BAS phases are interpreted to have crystallized metastably in lieu of dumortierite in accordance with Ostwald Step Rule, possibly first as “boron mullite,” then as monoclinic phases. The presence of such metastable phases is suggestive that pegmatites crystallize, at least partially, by disequilibrium processes, with significant undercooling, and at high viscosities, which limit diffusion rates.