Thermal expansion of gehlenite, Ca2Al[AlSiO7], and the related aluminates LnCaAl[Al2O7] with Ln = Tb, Sm

The thermal expansion of gehlenite, Ca₂Al[AlSiO₇], (up to T=830 K), TbCaAl[Al₂O₇] (up to T=1100 K) and SmCaAl[Al₂O₇] (up to T=1024 K) has been determined. All compounds are of the melilite structure type with space group Thermal expansion data were obtained from in situ X-ray powder diffraction experiments in-house and at HASYLAB at the Deutsches Elektronen Synchrotron (DESY) in Hamburg (Germany). The thermal expansion coefficients for gehlenite were found to be: α₁=7.2(4)×10⁻⁶×K⁻¹+3.6(7)×10⁻⁹ΔT×K⁻² and α₃=15.0(1)×10⁻⁶×K⁻¹. For TbCaAl[Al₂O₇] the respective values are: α₁=7.0(2)×10⁻⁶×K⁻¹+2.0(2)×10⁻⁹ΔT×K⁻² and α₃=8.5(2)×10⁻⁶×K⁻¹+2.0(3)×10⁻⁹ΔT×K⁻², and the thermal expansion coefficients for SmCaAl[Al₂O₇] are: α₁=6.9(2)×10⁻⁶×K⁻¹+1.7(2)×10⁻⁹ΔT×K⁻² and α₃=9.344(5)×10⁻⁶×K⁻¹. The expansion mechanisms of the three compounds are explained in terms of structural trends obtained from Rietveld refinements of the crystal structures of the compounds against the powder diffraction patterns. No structural phase transitions have been observed. While gehlenite behaves like a ‘proper’ layer structure, the aluminates show increased framework structure behavior. This is most probably explained by stronger coulombic interactions between the tetrahedral conformation and the layer-bridging cations due to the coupled substitution (Ca²⁺+Si⁴⁺)–(Ln ³⁺+Al³⁺) in the melilite-type structure. This article has been mistakenly published twice. The first and original version of it is available at .     

Determination of structures of Ca2CoSi2O7, Ca2MgSi2O7, and Ca2(Mg0.55Fe0.45)Si2O7 in incommensurate and normal phases and observation of diffuse streaks at high temperature

 The structures of Ca₂CoSi₂O₇, Ca₂MgSi₂O₇, and Ca₂(Mg_0.55Fe_0.45)Si₂O₇ have been determined in the temperature range between 297 and 773 K with arbitrary intervals. The structures of the incommensurate phase of the three compounds are characterized by the presence of the six-, seven-, and eight-coordinated Ca–O polyhedra and of the bundles along the c-axes consisting of four arrays of the six-coordinated Ca–O polyhedra and an array of T¹O₄ (T¹: Co, Mg, or Mg–Fe) tetrahedra in the structures. The number of bundles in each material decreases at elevated temperatures. The incommensurate phase undergoes a phase transition into the normal phase at 493 K in Ca₂CoSi₂O₇, at 360 K in Ca₂MgSi₂O₇, and at 510 K in Ca₂(Mg_0.55Fe_0.45)Si₂O₇. The features of the structures of the normal phase are almost the same as those found in the basic structures (the averaged structures of the incommensurate structures), and this fact implies that the characteristics of the structures, such as the six-coordinated Ca–O polyhedra or fragments of the bundles, should be partially preserved at higher temperatures both in the incommensurate structures and also in the structures of the normal phase. Analyses of anisotropic displacement parameters clarified that disorder of the modulation waves is developed in the structures at higher temperatures. The evolution of a disorder in the structures was ascertained by observation of the circular diffuse streaks in the vicinity of the transition temperature between the incommensurate and normal phases. Received: 3 July 2000 / Accepted: 26 October 2000     

Distribution of cations at two tetrahedral sites in Ca2MgSi2O7-Ca2Fe3+AlSiO7 series synthetic melilite and its relation to incommensurate structure

Synthetic melilites on the join Ca₂MgSi₂O₇ (åkermanite: Ak)-Ca₂Fe³⁺AlSiO₇ (ferrialuminium gehlenite: FAGeh) were studied using X-ray powder diffraction and ⁵⁷Fe Mössbauer spectroscopic methods to determine the distribution of Fe³⁺ between two different tetrahedral sites (T1 and T2), and the relationship between ionic substitution and incommensurate (IC) structure. Melilites were synthesized from starting materials with compositions of Ak₁₀₀, Ak₈₀FAGeh₂₀, Ak₇₀FAGeh₃₀ and Ak₅₀FAGeh₅₀ by sintering at 1,170–1,350 °C and 1 atm. The average chemical compositions and end-member components, Ak, FAGeh and Geh (Ca₂Al₂SiO₇), of the synthetic melilites were Ca_2.015Mg_1.023Si_1.981O₇ (Ak₁₀₀), Ca_2.017Mg_0.788Fe _0.187 ³⁺ Al_0.221Si_1.791O₇ (Ak₇₈FAGeh₁₉Geh₃), Ca_1.995Mg_0.695Fe _0.258 ³⁺ Al_0.318Si_1.723O₇ (Ak₆₉FAGeh₂₅Geh₆) and Ca_1.982Mg_0.495Fe _0.449 ³⁺ Al_0.519Si_1.535O₇ (Ak₄₉FAGeh₄₄Geh₇), respectively. Rietveld refinements using X-ray powder diffraction data measured using CuK _α -radiation at room temperature converged successfully with goodness-of-fits of 1.15–1.26. The refined Fe occupancies at the T1 and T2 sites and the Mg and Si contents determined by electron microprobe analysis gave the site populations of [0.788Mg + 0.082Fe³⁺ + 0.130Al]_T1[0.104Fe³⁺ + 0.104Al + 1.792Si]_T2 for Ak₇₈FAGeh₁₉Geh₃, [0.695Mg + 0.127Fe³⁺ + 0.178Al]_T1[0.132Fe³⁺ + 0.144Al + 1.724Si]_T2 for Ak₆₉FAGeh₂₅Geh₆ and [0.495Mg + 0.202Fe³⁺ + 0.303Al]_T1[0.248Fe³⁺ + 0.216Al + 1.536Si]_T2 for Ak₄₉FAGeh₄₄Geh₇ (apfu: atoms per formula unit), respectively. The results indicate that Fe³⁺ is distributed at both the T1 and the T2 sites. The mean T1–O distance decreases with the substitution of Fe³⁺ + Al³⁺ for Mg²⁺ at the T1 site, whereas the mean T2–O distance increases with substitution of Fe³⁺ + Al³⁺ for Si⁴⁺ at the T2 site, causing decrease in the a dimension and increase in the c dimension. However, in spite of the successful Rietveld refinements for the X-ray powder diffraction data measured using CuK _α-radiation at room temperature, each Bragg reflection measured using CuK _α1-radiation at room temperature showed weak shoulders, which were not observed in those measured at 200 °C. The Mössbauer spectra of the melilites measured at room temperature consist of two doublets assigned to Fe³⁺ at the T1 site and two or three doublets to Fe³⁺ at the T2 site, implying the existence of multiple T1 and T2 sites with different site distortions. These facts can be interpreted in terms of the IC structure in all synthetic melilites at room temperature, respectively. The results of Mössbauer analysis indicate that the IC structure in melilite is caused by not only known multiple T1 site, but also multiple T2 site at room temperature.     

A multi-analytical study of the crystal structure of unusual Ti–Zr–Cr-rich Andradite from the Maronia skarn, Rhodope massif, western Thrace, Greece

Unusual Ti–Cr–Zr-rich garnet crystals from high-temperature melilitic skarn of the Maronia area, western Thrace, Greece, were investigated by electron-microprobe analysis, powder and single-crystal X-ray diffraction, IR, Raman and Mössbauer spectroscopy. Chemical data showed that the garnets contain up to 8 wt.% TiO₂, 8 wt.% Cr₂O₃ and 4 wt.% ZrO₂, representing a solid solution of andradite (Ca₃Fe³⁺ ₂Si₃O₁₂ ≈46 mol%), uvarovite (Ca₃Cr₂Si₃O₁₂ ≈23 mol%), grossular (Ca₃Al₂Si₃O₁₂ ≈10 mol%), schorlomite (Ca₃Ti₂[Si,(Fe³⁺,Al³⁺)₂]O₁₂ ≈15 mol%), and kimzeyite (Ca₃Zr₂[Si,Al₂]₃O₁₂ ≈6 mol%). The Mössbauer analysis showed that the total Fe is ferric, preferentially located at the octahedral site and to a smaller extent at the tetrahedral site. Single-crystal XRD analysis, Raman and IR spectroscopy verified substitution of Si mainly by Al³⁺, Fe³⁺ and Ti⁴⁺. Cr³⁺ and Zr⁴⁺ are found at the octahedral site along with Fe³⁺, Al³⁺ and Ti⁴⁺. The measured H₂O content is 0.20 wt.%. The analytical data suggest that the structural formula of the Maronia garnet can be given as: (Ca_2.99Mg_0.03)_Σ=3.02(Fe³⁺ _0.67Cr_0.54Al_0.33Ti_0.29Zr_0.15)_Σ=1.98(Si_2.42Ti_0.24Fe_0.18Al_0.14)_Σ=2.98O₁₂OH_0.11. Ti-rich garnets are not common and their crystal chemistry is still under investigation. The present work presents new evidence that will enable the elucidation of the structural chemistry of Ti- and Cr-rich garnets.     

Thermal expansion of gehlenite, Ca2Al[AlSiO7], and the related aluminates LnCaAl[Al2O7] with Ln=Tb, Sm

The thermal expansion of gehlenite, Ca₂Al[AlSiO₇], (up to T=830 K), TbCaAl[Al₂O₇] (up to T=1,100 K) and SmCaAl[Al₂O₇] (up to T=1,024 K) has been determined. All compounds are of the melilite structure type with space group Thermal expansion data was obtained from in situ X-ray powder diffraction experiments in-house and at HASYLAB at the Deutsches Elektronen Synchrotron (DESY) in Hamburg (Germany). The thermal expansion coefficients for gehlenite were found to be: α₁=7.2(4)×10⁻⁶ K⁻¹+3.6(7)×10⁻⁹ΔT K⁻² and α₃=15.0(1)×10⁻⁶ K⁻¹. For TbCaAl[Al₂O₇] the respective values are: α₁=7.0(2)×10⁻⁶ K⁻¹+2.0(2)×10⁻⁹ΔT K⁻² and α₃=8.5(2)×10⁻⁶ K⁻¹+2.0(3)×10⁻⁹ΔT K⁻², and the thermal expansion coefficients for SmCaAl[Al₂O₇] are: α₁=6.9(2)× 10⁻⁶ K⁻¹+1.7(2)×10⁻⁹ΔT K⁻² and α₃=9.344(5)×10⁻⁶ K⁻¹. The expansion-mechanisms of the three compounds are explained in terms of structural trends obtained from Rietveld refinements of the crystal structures of the compounds against the powder diffraction patterns. No structural phase transitions have been observed. While gehlenite behaves like a ’proper’ layer structure, the aluminates show increased framework structure behaviour. This is most probably explained by stronger coulombic interactions between the tetrahedral conformation and the layer-bridging cations due to the coupled substitution (Ca²⁺+Si⁴⁺)-(Ln ³⁺+Al³⁺) in the melilite-type structure. Electronic Supplementary Material Supplementary material is available for this article at     

Crystal chemistry of a Mg-vesuvianite and implications of phase equilibria in the system CaO-MgO-Al2O3-SiO2-H2O-CO21

Abstract Chemical analysis (including H₂, F₂, FeO, Fe₂O₃) of a Mg-vesuvianite from Georgetown, Calif., USA, yields a formula, Ca_18.92Mg_1.88Fe³⁺_0.40Al_10.97Si_17.81- O_69.0.1(OH)_8.84F_0.14, in good agreement on a cation basis with the analysis reported by Pabst (1936). X-ray and electron diffraction reveal sharp reflections violating the space group P4/nnc as consistent with domains having space groups P4/n and P4nc. Refinement of the average crystal structure in space group P4/nnc is consistent with occupancy of the A site with Al, of the half-occupied B site by 0.8 Mg and 0.2 Fe, of the half-occupied C site by Ca, of the Ca (1,2,3) sites by Ca, and the OH and O(10) sites by OH and O. We infer an idealized formula for Mg-vesuvianite to be Ca₁₉Mg(MgAl₇)Al₄Si₁₈O₆₉(OH)₉, which is related to Fe^3+-vesuvianite by the substitutions Mg + OH = Fe³⁺+ O in the B and O(10) sites and Fe³⁺= Al in the AlFe site. Thermodynamic calculations using this formula for Mg-vesuvianite are consistent with the phase equilibria of Hochella, Liou, Keskinen & Kim (1982) but inconsistent with those of Olesch (1978). Further work is needed in determining the composition and entropy of synthetic vs natural vesuvianite before quantitative phase equilibria can be dependably generated. A qualitative analysis of reactions in the system CaO-MgO-Al₂O₃-SiO₂-H₂O-CO₂ shows that assemblages with Mg-vesuvianite are stable to high T in the absence of quartz and require water-rich conditions (XH₂O > 0.8). In the presence of wollastonite, Mg-vesuvianite requires very water-rich conditions (XH₂O > 0.97).     

Local structure of (Ca, Sr)2 (Mg, Co, Zn) Si2O7 melilite solid-solution with modulated structure

The local structure around Co, Zn and Sr atoms in incommensurately modulated, melilite-type X₂T¹ T ₂ ² O₇ (X=Ca and Sr, T¹=Mg, Co and Zn, T²=Si) solid-solutions has been investigated by EXAFS analyses. The modulated structure was confirmed in Ca_2-xSr_xCoSi₂O₇ solid-solutions with X=0.0 to 0.6 and for both Ca₂Mg_1-YCo_YSi₂O₇ and Ca₂Mg_1-YZn_YSi₂O₇ solid-solutions over the whole compositional range at room temperature. The actual bond-distances determined by the EXAFS method for the T¹ site (Co-, Zn-O) in the modulated structure are longer than the mean bond-distances obtained from the X-ray diffraction method. This is attributable to the libration of the T¹ tetrahedra. In the Ca_1-XSr_XCoSi₂O₇ solid-solution both the Sr-O and Co-O distances by the EXAFS method for the X-site increase from Ca end-member to Sr end-member. These increases are respectively 0.8% and 0.6%. This means the local expansions of the tetrahedral sheets and of the XO polyhedra are well matched. In the modulated Ca₂Co_1-YMg_YSi₂O₇ and Ca₂Zn_1-YMg_YSi₂O₇ solid-solutions, the actual Co-O and Zn-O distances for the T¹-sites are nearly constant in the whole compositional range. The compositional variations of the local structure around the cations in the solid-solution are different for the X and T¹ sites. It is concluded that the local geometric restriction for the size of substituted cation in X site is larger than that in T¹ site. The dimension of the tetrahedral sheet puts restriction on the size of the cations situated at the interlayer X sites. In other words, the different behavior of the local geometric restriction between the X and T¹ sites is an important feature of the melilite structure and is also related to the modulated structure.     

Natural occurrence of a (Fe,Mn, Mg) tetrasilicic potassium mica

A mica whose structural formula: (K_1.76Na_0.31)(Fe_2.22Mn_1.29Mg_0.99Ti_0.28Al_0.240.98) ·(Si_7.33Al_0.67)O_20.26(F_2.16OH_1.58) closely approximates that of tetrasilicic potassium mica K₂(M ₅ ²⁺ )Si₈O₂₀(OH,F)₄ where M²⁺ represents Mg²⁺, Fe²⁺, Mn²⁺, ..., has been discovered in the matrix of a peralkaline rhyolite (comendite) of the Mont-Dore massif (France). These micas had been obtained previously by synthesis only. In the groundmass of the rock, the micaceous phase is accompanied by a manganoan arfvedsonite, pyrophanite, magnetite, apatite, sphene, zircon and fluorite. The crystallographic properties of the mica are typically that of a tetrasilicic mica, with d ₀₆₀ = 1.533Å and space group C2/m. There is a regular decrease of d ₀₆₀ (parameter b) with the ionic radius of the octahedral cation in synthetic micas containing Fe²⁺, Co²⁺, Mg²⁺, Ni²⁺. The purely Mn²⁺ end-member could not be synthesised; its instability is discussed on the basis of structural considerations. The conditions of crystallization of the micaceous phase are estimated to be 760 ° C, 800 bars with a f _{o 2}=10^–14.7 bar. This mica has crystallized from a residual liquid, with high activity of silica and low activity of alumina, whose origin is discussed. The name MONT-DORITE is proposed for this natural tetrasilicic mica having Fe/Fe+Mg >1/2 and Fe/Fe+Mn >1/2. This name is from the stratovolcano Mont-Dore.     

Thermodynamic study of calcic amphiboles

The paper reports original thermochemical data on six natural amphibole samples of different composition. The data were obtained by high-temperature melt solution calorimetry in a Tian–Calvet microcalorometer and include the enthalpies of formation from elements for actinolite Ca_1.95(Mg_4.4Fe _0.5 ²⁺ Al₀₁)[Si_8.0O₂₂](OH)₂(–12024 ± 13 kJ/mol) and Ca_2.0(Mg_2.9Fe _1.9 ²⁺ Fe _0.2 ³⁺ )[Si_7.8Al_0.2O₂₂](OH)₂, (–11462 ± 18 kJ/mol), and Na_0.1Ca_2.0(Mg_3.2Fe _1.6 ²⁺ Fe _0.2 ³⁺ )[Si_7.7Al_0.3O₂₂](OH)₂ (–11588 ± 14 kJ/mol); for pargasite Na_0.5K_0.5Ca_2.0-(Mg_3.4Fe _1.8 ²⁺ Al_0.8)[Si_6.2Al_1.8O₂₂](OH)₂ (–12316 ± 10 kJ/mol) and Na_0.8K_0.2Ca_2.0(Mg_2.8Fe _1.3 ³⁺ Al_0.9) [Si_6.1Al_1.9O₂₂](OH)₂ (–12 223 ± 9 kJ/mol); and for hastingsite Na_0.3K_0.2Ca_2.0(Mg_0.4Fe _1.3 ²⁺ Fe _0.9 ³⁺ Al_0.2) [Si_6.4Al_1.6O₂₂](OH)₂ (?10909 ± 11 kJ/mol). The standard entropy, enthalpy, and Gibbs free energy of formation are estimated for amphiboles of theoretical composition: end members and intermediate members of the isomorphic series tremolite–ferroactinolite, edenite–ferroedenite, pargasite–ferropargasite, and hastingsite.     

Crystal chemistry of volcanic allanites indicative of naturally induced oxidation-dehydrogenation

The crystal chemistry of volcanic allanites from both the Youngest Toba Tuff (YTT), Sumatra, Indonesia and SK100 volcanic ash beds (SK100-VAB), Niigata, Japan has been examined by electron microprobe analysis (EMPA), Fourier-transform infrared spectroscopic analysis (FTIR), and single-crystal structure analysis. In the FTIR study, based on the Diamond ATR accessory, YTT and SK100- VAB allanites were observed to have different OH contents, respectively: the former has 0.64 wt% H₂O (OH: 0.40 apfu.), while the latter has 1.65 wt% H₂O (OH: 1.00 apfu.). The crystal structures of these two allanites have been refined to individual R indices (3.64 and 4.25) based on 1350 observed reflections (|Fo| > 4sig|Fo|) measured using a single-crystal diffractometer with MoKα X-radiation. The OH-poor YTT allanite has a shorter b axis, a longer c axis, and larger β value than the relatively OH-rich SK100-VAB one. The bond valence sums of O4 (accepter oxygen for H atom) and O10 (donor oxygen for H atom) are 1.962 and 1.709 v.u. for YTT allanite (valence sum: 3.671 v.u.) and 1.754 and 1.271 v.u. for SK100-VAB one (valence sum: 3.025 v.u.). The difference from the ideal total bond valence value (4.00 v.u.) of O4 and O10 in YTT allanite (0.33 v.u.) is smaller than that in SK100-VAB (0.98 v.u.). These difference values are also broadly consistent with the corresponding differences in OH content between the YTT (OH: 0.40 apfu.) and SK100-VAB allanites (OH: 1.00 apfu.) determined by FTIR- ATR. Chemical analyses, FTIR-ATR and crystal structure refinement of YTT and SK100-VAB allanites yielded the following crystal chemical formula: YTT: (Ca_0.83Mn²⁺ _0.06Fe²⁺ _0.11)(La_0.24Ce_0.32Pr_0.04Nd_0.11Sm_0.02Th_0.04Ca_0.21)(Al_0.73Fe³⁺ _0.19Ti_0.08)(Al_0.89Fe³⁺ _0.11)(Fe²⁺ _0.22Fe³⁺ _0.62Mg_0.16)(SiO₄)Si₂O₇O_1.6(OH)_0.4, SK100-VAB: (Ca_0.81Fe²⁺ _0.13Mn²⁺ _0.06)(La_0.22Ce_0.34Pr_0.05Nd_0.13Sm_0.02Th_0.02Ca_0.22)(Al_0.76Fe³⁺ _0.19Ti_0.05)Al_1.00(Fe²⁺ _0.73Fe³⁺ _0.17Mg_0.10)(Si_0.96Al_0.04O₄)Si₂O₇O(OH). Therefore, it is concluded that welding of the Youngest Toba Tuff caused the following post-crystallization changes to occur in YTT allanite: oxidation of Fe²⁺ to Fe³⁺, release of H₂, and the concomitant replacement of OH^? by O^2?. These oxidation and dehydrogenation processes advanced during the welding to thereby produce oxyallanite. Oxyallanite had been reported only in laboratory studies where it was produced by heating natural allanite. Our report on natural oxyallanite suggests that it may be present in other welded silicic volcanic rocks as well.     

Synthetic dumortierite: its PTX-dependent compositional variations in the system Al2O3-B2O3-SiO2-H2O

Dumortierite, generally simplified as Al₇BSi₃O₁₈, was synthesized in the pure system Al₂O₃–B₂O₃–SiO₂–H₂O (ABSH) using gels with variable Al/Si ratios mixed with H₃BO₃ and H₂O in known proportions as starting materials. Synthesis conditions ranged from 3 to 5 and 15 to 20 kbar fluid pressure at 650° to 880°C. On the basis of analyses, synthetic dumortierite shows relatively narrow homogeneity ranges with regard to Al/Si which, however, vary as a function of pressure: at low pressures (3–5 kbar) Al/Si is 2.77–2.94 versus 2.33–2.55 at high pressures (15–20 kbar). Outside of these homogeneity limits, dumortierite was found to coexist with quartz or corundum, depending on the starting composition. Whereas synthetic dumortierite invaribly contains 1.0 boron atom per formula unit (p.f.u.) based on 18 oxygens, the water contents vary drastically as a function of pressure and temperature (1.32–2.30 wt.% H₂O or 0.85–1.47 H p.f.u.). H₂O is an essential component in dumortierite. Structural formulae based on complete chemical analyses of the dumortierites synthesized reveal that there is invariably an Si-deficiency against the ideal number of 3.0 p.f.u. In the calculation procedure used here, this deficiency is balanced by assuming tetrahedral Al. The remaining Al, taken to occupy the octahedral sites, is always below the ideal number of 7.0 p.f.u. Charge-balancing the structure with the hydrogen found analytically leads to two different mechanisms of H incorporation: (1) 3H⁺ + octahedral vacancy for Al^[6]; (2) H⁺ + tetrahedral Al for Si^[4]. Dumortierite synthesized at high fluid pressure contains little Al^[4] and, thus, little H⁺ of type 2; its hydrogen is predominantly present as type 1. Conversely, dumortierite formed at low fluid pressures is high in Al^[4] and hydrogen type 2. The amounts of hydrogen type 1 in low-pressure dumortierites decrease with rising temperatures of synthesis. Typical structural formulae are: (Al_6.670.33)[Al_0.49Si_2.51–O_13.53(OH)_1.47](BO₃) for a low-pressure product, and (Al_6.680.32)[Al_0.09Si_2.91O_13.94(OH)_1.06](BO₃) for a high-pressure product. Independently of the synthesis conditions, dumortierite was found always to be orthorhombic, with b₀/a₀ deviating slightly, but significantly from the valid for hexagonal lattice geometry. As a function of increasing Al/Si in the synthetic crystals, their a₀, c₀, and V₀ rise, whereas b₀ decreases. Thus b₀/a₀ decreases most sensitively with rising Al/Si and also with growing Al^[4]. More experimentation is required before the compositional variations of dumortierite found here can be applied successfully to geothermobarometry of natural rocks.     

Mendigite,Mn2Mn2MnCa(Si3O9)2, a new mineral species of the bustamite group from the Eifel volcanic region,Germany

A new mineral, mendigite (IMA no. 2014-007), isostructural with bustamite, has been found in the In den Dellen pumice quarry near Mendig, Laacher Lake area, Eifel Mountains, Rhineland-Palatinate (Rheinland-Pfalz), Germany. Associated minerals are sanidine, nosean, rhodonite, tephroite, magnetite, and a pyrochlore-group mineral. Mendigite occurs as clusters of long-prismatic crystals (up to 0.1 × 0.2 × 2.5 mm in size) in cavities within sanidinite. The color is dark brown with a brown streak. Perfect cleavage is parallel to (001). D _calc = 3.56 g/cm³. The IR spectrum shows the absence of H₂O and OH groups. Mendigite is biaxial (–), α = 1.722 (calc), β = 1.782(5), γ = 1.796(5), 2V _meas = 50(10)°. The chemical composition (electron microprobe, mean of 4 point analyses, the Mn²⁺/Mn³⁺ ratio determined from structural data and charge-balance constraints) is as follows (wt %): 0.36 MgO, 10.78 CaO, 37.47 MnO, 2.91 Mn₂O₃, 4.42 Fe₂O₃, 1.08 Al₂O₃, 43.80 SiO₂, total 100.82. The empirical formula is Mn_2.00(Mn_1.33Ca_0.67) (Mn_0.50 ²⁺ Mn_0.28 ³⁺ Fe_0.15 ³⁺ Mg_0.07)(Ca_0.80 (Mn_0.20 ²⁺)(Si_5.57 Fe_0.27 ³⁺ Al_0.16O₁₈). The idealized formula is Mn₂Mn₂MnCa(Si₃O₉)₂. The crystal structure has been refined for a single crystal. Mendigite is triclinic, space group \(P\bar 1\); the unit-cell parameters are a = 7.0993(4), b = 7.6370(5), c = 7.7037(4) Å, α = 79.58(1)°, β = 62.62(1)°, γ = 76.47(1)°; V = 359.29(4) ų, Z = 1. The strongest reflections on the X-ray powder diffraction pattern [d, Å (I, %) (hkl)] are: 3.72 (32) (020), 3.40 (20) (002, 021), 3.199 (25) (012), 3.000 (26), (\(01\bar 2\), \(1\bar 20\)), 2.885 (100) (221, \(2\bar 11\), \(1\bar 21\)), 2.691 (21) (222, \(2\bar 10\)), 2.397 (21) (\(02\bar 2\), \(21\bar 1\), 203, 031), 1.774 (37) (412, \(3\bar 21\)). The type specimen is deposited in the Fersman Mineralogical Museum, Russian Academy of Sciences, Moscow, registration number 4420/1.     

Mayenite-supergroup minerals from burned dump of the Chelyabinsk Coal Basin

Three minerals of the mayenite supergroup have been found in fluorellestadite-bearing metacarbonate rock (former fragment of petrified wood of ankeritic composition) from the dump at the Baturinskaya-Vostochnaya-1-2 mine. These are eltyubyuite Ca₁₂Fe1°Si₄O3₂Cl6, its fluorine analog Ca₁₂Fe₁₀³⁺Si₄O₃₀F₁₀, and chlormayenite-wadalite Ca₁₂(Al,Fe)₁₄O₃₂Cl₂-Ca₁₂(Al,Fe)₁₀Si₄O₃₂Cl₂. The first two phases occur in the reaction mantle around hematite, magnesioferrite, and Ca-ferrite aggregates (“calciohexaferrite” CaFe₁₂O₁₉, “grandiferrite” CaFe₄O₇, and “dorrite phase” Ca₂(Fe₅^3 +Mn0^0.5Mg_0.5)(Si_0.5Fe_5.5³⁺)O₂₀) and, rarely, as individuals in grained aggregates of fluorellestadite-cuspidine (± lar- nite ± rusinovite Ca₁₀(Si₂O₇)₃Cl₂). Assemblages of zoned chlormayenite-wadalite crystals are found in grained aggregates of fluorellestadite- cuspidine, which lack Ca-ferrite. Also, harmunite CaFe₂O₄, chlorellestadite, fluorapatite, anhydrite, rondorfite CasMg(SiO₄)₄Cl₂, fluorine analog of rondorfite CasMg(SiO₄)₄F₂, “Mg-cuspidine” Ca_3.5(Mg,Fe)_0.5(Si₂O₇)F₂, fluorite, barioferrite BaFe₁₂O₁9, zhangpeishanite BaFCl, and other rare phases are identified in this rock. Data on the chemical composition and Raman spectroscopy of the mayenite-supergroup minerals are given. The genesis of metacarbonate rock is considered in detail: “oxidizing calcination” of Ca-Fe-carbonates with the formation of hematite and lime; reaction between hematite and lime with the formation of different Ca-ferrites; formation of larnite as a result of reaction between SiO₂ and lime or CaCO₃; and reactionary impact of hot Cl-F-S-bearing gases on early assemblages. Eltyubyuite and its fluorine analog crystallized at the stages of gas impact. It is presumed that the maximum temperature during the formation of rock reached 1200–1230 °C. © 2015, V.S. Sobolev IGM, Siberian Branch of the RAS. Published by Elsevier B.V. All rights reserved.     

Crystal chemistry of selenates with mineral-like structures: V. Crystal structures of (H3O)2[(UO2)(SeO4)2(H2O)](H2O)2 and (H3O)2[(UO2)(SeO4)2(H2O)](H2O), new compounds with rhomboclase and goldichite topology

The crystal structures of two new compounds (H₃O)₂[(UO₂)(SeO₄)₂(H₂O)](H₂O)₂ (1, orthorhombic, Pnma, a = 14.0328(18), b = 11.6412(13), c = 8.2146(13) Å, V = 134.9(3) ų) and (H₃O)₂[(UO₂)(SeO₄)₂(H₂O)](H₂O) (2, monoclinic, P2₁/c, a = 7.8670(12), b = 7.5357(7), c = 21.386(3) Å, β = 101.484(12)°, V = 1242.5(3) ų) have been solved by direct methods and refined to R ₁ = 0.076 and 0.080, respectively. The structures of both compounds contain sheet complexes [(UO₂)(SeO₄)₂]^2? formed by cornershared [(UO₂)O₄(H₂O)] bipyramids and SeO₄ tetrahedrons. The sheets are parallel to the (100) plane in structure 1 and to (?102) in structure 2. The [(UO₂)(SeO₄)₂(H₂O)]^2? layers are linked by hydrogen bonds via interlayer groups H₂O and H₃O⁺. The sheet topologies in structures 1 and 2 are different and correspond to the topologies of octahedral and tetrahedral complexes in rhomboclase (H₂O₂)⁺[Fe(SO₄)₂(H₂O)₂] and goldichite K[Fe(SO₄)₂(H₂O)₂](H₂O)₂, respectively.     

Synthesis of Ca-tourmaline in the system CaO-MgO-Al2O3-SiO2-B2O3-H2O-HCl

Summary. ?Ca-tourmaline has been synthesized hydrothermally in the presence of Ca(OH)₂ and CaCl₂-bearing solutions of different concentration at T = 300–700 °C at a constant fluid pressure of 200 MPa in the system CaO-MgO-Al₂O₃-SiO₂-B₂O₃-H₂O-HCl. Synthesis of tourmaline was possible at 400 °C, but only above 500 °C considerable amounts of tourmaline formed. Electron microprobe analysis and X-ray powder data indicate that the synthetic tourmalines are essentially solid solutions between oxy-uvite, CaMg₃- Al₆(Si₆O₁₈)(BO₃)₃(OH)₃O, and oxy-Mg-foitite, □(MgAl₂)Al₆(Si₆O₁₈)(BO₃)₃(OH)₃O. The amount of Ca ranges from 0.36 to 0.88 Ca pfu and increases with synthesis temperature as well as with bulk Ca-concentration in the starting mixture. No hydroxy-uvite, CaMg₃(MgAl₅)(Si₆O₁₈)(BO₃)₃(OH)₃(OH), could be synthesized. All tourmalines have < 3 Mg and > 6 Al pfu. The Al/(Al + Mg)-ratio decreases from 0.80 to 0.70 with increasing Ca content. Al is coupled with Mg and Ca via the substitutions Al₂□Mg₋₂Ca₋₁ and AlMg₋₁H₋₁. No single phase tourmaline could be synthesized. Anorthite ( + quartz in most runs) has been found coexisting with tourmaline. Other phases are chlorite, tremolite, enstatite or cordierite. Between solid and fluid, Ca is strongly fractionated into tourmaline ( + anorthite). The concentration ratio D = Ca(fluid)/Ca(tur) increases from 0.20 at 500 °C up to 0.31 at 700 °C. For the assemblage turmaline + anorthite + quartz + chlorite or tremolite or cordierite, the relationship between Ca content in tourmaline and in fluid with temperature can be described by the equation (whereby T = temperature in °C, Ca(tur) = amount of Ca on the X-site in tourmaline, Ca( fluid) = concentration of Ca²⁺ in the fluid in mol/l). The investigations may serve as a first guideline to evaluate the possibility to use tourmaline as an indicator for the fluid composition.

---

Zusammenfassung. ?Synthese von Ca-Turmelin im System CaO-MgO-Al ₂ O ₃ -SiO ₂ -B ₂ O ₃ -H ₂ O-HCl Im System CaO-MgO-Al₂O₃-SiO₂-B₂O₃-H₂O-HCl wurde Ca-Turmalin hydrothermal aus Ca(OH)₂ and CaCl₂-haltigen L?sungen bei T = 300–700 °C und einem konstanten Fluiddruck von 200 MPa synthetisiert. Die Synthese von Turmalin war m?glich ab 400 °C, aber nur oberhalb von 500 °C bildeten sich deutliche Mengen an Turmalin. Elektronenstrahl-Mikrosondenanalysen und R?ntgenpulveraufnahmen zeigen, da? Mischkristalle der Reihe Oxy-Uvit, CaMg₃Al₆(Si₆O₁₈)(BO₃)₃(OH)₃O, und Oxy-Mg-Foitit, □(MgAl₂)Al₆(Si₆O₁₈)(BO₃)₃(OH)₃O gebildet wurden. Der Anteil an Ca variiert zwischen 0.36 und 0.88 Ca pfu und nimmt mit zunehmender Synthesetemperatur und zunehmender Ca-Konzentration im System zu. Hydroxy-Uvit, CaMg₃(MgAl₅) (Si₆O₁₈)(BO₃)₃(OH)₃(OH), konnte nicht synthetisiert werden. Alle Turmaline haben < 3 Mg und > 6 Al pfu. Dabei nimmt das Al/(Al + Mg)- Verh?ltnis mit zunehmendem Ca-Gehalt von 0.80 auf 0.70 ab. Al ist gekoppelt mit Mg und Ca über die Substitutionen Al₂□Mg₋₂Ca₋₁ und AlMg₋₁H₋₁. Einphasiger Turmalin konnte nicht synthetisiert werden. Anorthit (+ Quarz in den meisten F?llen) koexistiert mit Turmalin. Andere Phasen sind Chlorit, Tremolit, Enstatit oder Cordierit. Ca zeigt eine deutliche Fraktionierung in den Festk?rpern Turmalin (+ Anorthit). Das Konzentrationsverh?ltnis D = Ca(fluid)/Ca(tur) nimmt von 0.20 bei 500 °C auf 0.31 bei 700 °C zu. Für die Paragenese Turmalin + Anorthit + Quarz mit Chlorit oder Tremolit oder Cordierit gilt folgende Beziehung zwischen Ca-Gehalt in Turmalin und Fluid und der Temperatur: (wobei T = Temperatur in °C, Ca(tur) = Anteil an Ca auf der X-Position in Turmalin, Ca(fluid) = Konzentration von Ca²⁺ im Fluid in mol/l). Die Untersuchungen dienen zur ersten Absch?tzung, ob Turmalin als Fluidindikator petrologisch nutzbar ist.

---

---

Received July 24, 1998;/revised version accepted October 21, 1999     

Viscosity-temperature relationships in the system Na2Si2O5-Na4Al2O5

The viscosity-temperature relationships of five melts on the join Na₂Si₂O₂-Na₄Al₂O₅ (5, 10, 20, 30 and 40 mole percent Na₄Al₂O₅) have been measured in air, at 1 atm and 1000–1350°C with a concentric cylinder viscometer. All the melts on this join of constant bulk polymerization behave as Newtonian fluids, in the range of shear rates investigated, and the melts exhibit Arrhenian viscosity-temperature relationships.Isothermal viscosities on this join initially decrease and then increase with increasing mole percent Na₄Al₂O₅. The minimum viscosity occurs near 20 mole percent Na₄Al₂O₅ at 1000°C and moves to higher Na₄Al₂O₅ content with increasing temperature.The observation of a viscosity minimum along the join Na₂Si₂-O₅-Na₄Al₂O₅ is not predicted based on earlier viscosity data for the system Na₂O-Al₂O₃-SiO₂ (RlEBLlNG, 1966) or based on calculation methods derived from this and other data (Bottinga and Weill, 1972). This unexpected behavior in melt viscosity-temperature relations emphasizes the need for a more complete data set in simple silicate systems.Previous spectroscopic investigation of melts on the join Na₂2Si₂O₅-Na₄Al₂O₅ offer a structural explanation for the observed viscosity data in terms of a disproportionation reaction involving polyanionic units. Macroscopically, the viscosity data may be qualitatively reconciled with the configurational entropy model for viscous flow (Richet, 1984).     

Comparative compressibilities of majorite-type garnets

Relative compressibilities of five silicate garnets were determined by single-crystal x-ray diffraction on crystals grouped in the same high-pressure mount. The specimens include a natural pyrope [(Mg_2.84Fe_0.10Ca_0,06) Al₂Si₃O₁₂], and four synthetic specimens with octahedrally-coordinated silicon: majorite [Mg₃(MgSi)Si₃O₁₂], calcium-bearing majorite [(Ca_0.49Mg_2.51)(MgSi)Si₃0₁₂], sodium majorite [(Na_1.88Mgp_0.12)(Mg_0.06Si_1.94)Si₃O₁₂], and an intermediate composition [(Na_0.37Mg_2.48)(Mg_0.13Al_1.07 Si₀₈₀) Si₃O₁₂]. Small differences in the compressibilities of these crystals are revealed because they are subjected simultaneously to the same pressure. Bulk-moduli of the garnets range from 164.8 ± 2.3 GPa for calcium majorite to 191.5 ± 2.5 GPa for sodium majorite, assuming K′=4. Two factors, molar volume and octahedral cation valence, appear to control garnet compression.     

Low-temperature study of natural melilite (Ca1.89Sr0.01Na0.08K0.02) (Mg0.92Al0.08)(Si1.97Al0.03)O7: towards a commensurate value of the q vector

As is usual for peculiar chemical compositions, melilite-type compounds exhibit a two-dimensional incommensurately modulated structure which can be described with two wave vectors: q ₁ =(a* + b*) and q ₂ =(–a* + b*), where a* and b* are the tetragonal reciprocal axes of the basic cell. The low-temperature dependence of the modulation wave vector of a natural melilite crystal with chemical composition (Ca_1.89Sr_0.01 Na_0.08K_0.02)(Mg_0.92Al_0.08)(Si_1.97Al_0.03)O₇ has been studied by X-ray single-crystal diffraction methods in the temperature range 298–100 K. The value of the coefficient shows a continuous linear increase, ranging from 0.281(1) at 298 K to 0.299(1) at 100 K. No plateau-like temperature dependence was observed throughout the temperature studied, thus indicating that no independent phase with a specific q stabilizes in this natural crystal. A comparison with the low-temperature behaviour of synthetic Ca₂MgSi₂O₇ is given.     

Thermodynamic properties of chlorite CCa-2. Heat capacities, heat contents and entropies

The heat capacities of the international reference clay mineral chlorite CCa-2 from Flagstaff Hill, California, were measured by low temperature adiabatic calorimetry and differential scanning calorimetry, from 5 to 520 K (at 1 bar). The studied chlorite is a Fe-bearing trioctahedral chlorite with an intermediary composition between ideal clinochlore (Si₃Al)(Mg₅Al)O₁₀(OH)₈ and chamosite (Si₃Al)(Fe₅Al)O₁₀(OH)₈. Only few TiO₂ impurities were detected in the natural chlorite sample CCa-2. Its structural formula, obtained after subtracting the remaining TiO₂ impurities, is (Si_2.633Al_1.367)(Al_1.116Mg_2.952Mn_0.012Ca_0.011)O₁₀(OH)₈. From the heat capacity results, the entropy, standard entropy of formation and heat content of the chlorite were deduced. At 298.15 K, the heat capacity of the chlorite is 547.02 (±0.27) J mol⁻¹ K⁻¹ and the molar entropy is 469.4 (±2.9) J mol⁻¹ K⁻¹. The standard molar entropy of formation of the clay mineral from the elements is −2169.4 (±4.0) J mol⁻¹ K⁻¹.     

Distance least squares modelling of the cubic sodalite structure and of the thermal expansion of Na8(Al6Si6O24)I2

The Distance Least Squares (DLS) structure modelling technique is used to determine the room-temperature structures of the sodalites Li₈(Al₆Si₆O₂₄)Cl₂, Na₈(Al₆Si₆O₂₄)Cl₂, K₈(Al₆Si₆O₂₄)Cl₂, Na₈(Al₆Si₆O₂₄)Br₂, and Na₈(Al₆Si₆O₂₄)I₂. The technique is also used to calculate the thermal expansion behaviour of Na₈(Al₆Si₆O₂₄)I₂ assuming that the discontinuity in its thermal expansion curve occurred either when the ideal fully-expanded state was achieved (case 1) or when the x-coordinate of the sodium atom became 0.25 (case 2). The results are given as plots of bond lengths and bond angles as a function of temperature. Case 2 was preferred and analysis of the results implied that the driving force for the untwisting of the partially-collapsed sodalite framework was in the framework bonds with the cavity ion bonds resisting the untwisting. Best estimates indicate that the expansion of the Na-O and Na-I bonds are 9% and 27.4% respectively, between room temperature and 810° C, and there is an apparent shortening of the framework bond distances of about 1.5%.