High-temperature behaviour of melilite: in situ X-ray diffraction study of gehlenite–åkermanite–Na melilite solid solution (original) (raw)
Related papers
Physics and Chemistry of Minerals, 2003
Low-temperature study of natural melilite (Ca 1.89 Sr 0.01 Na 0.08 K 0.02 ) (Mg 0.92 Al 0.08 )(Si 1.97 Al 0.03 )O 7 : towards a commensurate value of the q vector Abstract 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 ¼ a(a* + b*) and q 2 ¼ a()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 1.89 Sr 0.01 Na 0.08 K 0.02 )(Mg 0.92 Al 0.08 )(Si 1.97 Al 0.03 )O 7 has been studied by X-ray single-crystal diffraction methods in the temperature range 298-100 K. The value of the a 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 Ca 2 MgSi 2 O 7 is given.
Thermal expansion and phase transitions in åkermanite and gehlenite
Physics and Chemistry of Minerals, 2005
Thermal expansion has been measured by laboratory and synchrotron X-ray powder diffraction for end-member åkermanite (ak, Ca2MgSi2O7) and gehlenite (ge, Ca2Al2SiO7) in the range 20–1,500 K. In ak in the range 340–390 K, there is a negative linear thermal expansion in [001] direction. This is related to the phase transition from an incommensurate modulated structure (IC) to a normal one (N). The volumetric mean thermal expansion coefficients for ak and ge, obtained with a linear fit of the experimental data in the temperature range 298–1,400 K, are respectively 32.1×10−6 and 28.3×10−6 K−1 . The variation of the c/a ratio with temperature, due to different thermal expansion along the crystallographic axes, can be related to the different behaviour of the tetrahedral layers in the N and IC phases. Analysis of the variation of the superstructure peaks intensity across the phase transition confirms the tricritical behaviour of the IC/N transition in ak.
Diffusion in single crystal of melilite: interdiffusion of Al + Al vs. Mg + Si
Physics and Chemistry of Minerals, 2001
Interdiusion coecients of Al + Al vs. Mg + Si in the gehlenite±aÊ kermanite system of melilite were determined by coupled annealing of synthesized end-member single crystals. The observed diusion coecients for a couple-annealed sample vary for about 2 orders of magnitude, showing strong dependence on the gehlenite±aÊ kermanite composition: diusion coecient observed at 1350°C, for example, is 3´10 )13 cm 2 s )1 at 5 mol% aÊ kermanite composition (Ak 5 ), increases to 2´10 )11 cm 2 s )1 at Ak 80 , and then decreases to 1´10 )12 cm 2 s )1 at Ak 95 . The diusion coecient± temperature relation indicates high activation energy of diusion of about 420 kJ mol )1 for gehlenite-rich melilite. The observed diusion coecient±composition relation may be explained by a combination of (1) the diusion coecient±melting temperature relation (Flynn's rule) and (2) the feasibility of local charge compensation, which can possibly be maintained more easily in the intermediate chemical composition. The high activation energy value for gehlenitic melilite appears to correspond to the complex diusion mechanism. The observed highly variable diusion coecients suggest that gehlenite±aÊ kermanite zoning in the melilite crystals in Ca, Al-rich inclusions in the carbonaceous meteorites may provide a sensitive indicator for the thermal history of the inclusions.
The Canadian Mineralogist, 2006
Members of the pyrophanite-geikielite solid-solution series, Mn 1-x Mg x TiO 3 (0 < x < 0.7 apfu Mg), were obtained by solidstate synthesis at 1000°C at ambient pressure in air. In common with ilmenite, ternary Mn-Mg titanates adopt ordered R3 structures. The maximum solubility of Mg in MnTiO 3 under the given conditions is considered to be ~0.6 apfu Mg, as compounds with greater Mg content could not be synthesized. The structures of these titanates were refi ned by the Rietveld method from powder X-ray-diffraction data. Within the solid-solution series, unit-cell parameters and unit-cell volumes decrease with increase in Mg content. All compounds consist of distorted TiO 6 and AO 6 (A = Mn,Mg) octahedra, and in common with geikielite, pyrophanite, and ilmenite (sensu lato), the TiO 6 octahedra are distorted to a greater degree than (Mn,Mg)O 6. The extent of displacement of (Mn,Mg) and Ti from the centers of their coordination polyhedra varies irregularly with increasing Mg content, reaching a maximum for x = 0.1 and 0.2 apfu Mg. Entry of Mg 2+ into the VI A site results in "puckering" of layers consisting of TiO 6 octahedra (less distorted) above and below planes parallel to (001), and decreased "puckering" of the AO 6 octahedra (more distorted). The interlayer distance across the vacant octahedral site in the TiO 6 layer decreases regularly with entry of the smaller Mg 2+ cation into the VI A site. The absence of natural solid-solutions between geikielite and pyrophanite seems to be due to the contrasting geochemistry of Mn and Mg rather than to crystallochemical reasons.
Five-Dimensional Incommensurate Structure of the Melilite Electrolyte [CaNd]2[Ga]2[Ga2O7]2
Journal of the American Chemical Society, 2011
Solid oxide fuel cells (SOFCs) are an efficient and low-polluting energy conversion technology. A focus of current research is to reduce SOFC operating temperatures to ∼500°C through deployment of highly conductive electrolytes. Usually, oxygen transport is mediated via anionic vacancies, typically, in yttria-stabilized zirconia (YSZ) and gadolinium-doped ceria. 1 However, in some materials oxide ions conduct via an interstitial mechanism at "intermediate" temperatures. 2 Of particular interest are the apatite, 3 fergusonite, 4 and melilite 5,6 families, which contain mobile superstoichiometric oxygen. Those structures often show nonrational crystallographic translations, not accommodated within standard three-dimensional crystallographic space groups, but rather superspace groups, which describe periodicity in four, five, and six dimensions. 7 Numerous materials are now classified as higher dimensional structures, including Na 2 CO 3 , 8 intermetallic quasicrystals, 9 and cuprate hightemperature superconductors, 10 along with certain apatites, 11 fergusonites, 12 and melilites. 13 Of the latter structure types, the melilite family is currently the most broadly understood in terms of incommensurability and ion conduction, but a direct causal link between crystal chemistry and electrolyte performance remains to be established. To a first approximation, melilites adopt tetragonal P42 1 m symmetry and have the general formula [A 2 ] 2 [B I ] 2 [B II 2 O 7 ] 2 , where A is a large divalent or trivalent ion (e.g., Ln, Ca, Sr, and Ba) and B I and B II are symmetrically distinct small cations (e.g., Si and Ge). Two-dimensional extended oxygen networks of the type [(3.5.4.5) 2 , 3.5.3.5] are created by corner connection of B I O 4 and B II 2 O 7 units, with these layers separated by AO 8 polyhedra. 14 This structural framework can accommodate a large excess of oxygen, up to 0.32 per formula unit (e.g., [Ca 0.36 La 1.64 ] 2-[Ga] 2 [Ga 2 O 7.32 ] 2), 15 believed responsible for the high ion conductivity. Misfit of the tetrahedral layers with the interlayer cations is accommodated by atomic displacements, leading to incommensuration, which has been extensively studied in natural and synthetic silicate melilites, 13 but not so deeply for electrolyte compositions. Seifert et al. 16 analyzed akermanite ([Ca 2 ] 2 [Mg] 2-[Si 2 O 7 ] 2) using 29 Si magic angle spinning nuclear magnetic resonance (MAS NMR) spectroscopy and found an incommensurate to commensurate phase transition at 78°C. Synthetic [Ca 2 ] 2 [Co] 2 [Si 2 O 7 ] 2 (Co-akermanite) was found to be modulated using single-crystal X-ray diffraction, 17 and soon after, the solid solutions of [(Sr/Ca)] 2 [Co/Zn/Mg] 2 [Si 2 O 7 ] 2 were described, 18À21 with specific combinations of A and/or B I site substitutions resulting in modulation. Bindi et al. 22 reported a natural melilite ([Ca 1.89 Sr 0.01 Na 0.08 K 0.02 ] 2 [Mg 0.92 Al 0.08 ] 2 [Si 1.97 Al 0.03 O 7 ] 2) was also modulated, while transmission electron microscopy (TEM) defined the incommensurate to commensurate phase transition temperatures of [(Sr/Ca)] 2 [(Co/Mg/Zn/Fe/Cu)] 2-[(Si/Ge) 2 O 7 ] 2 compositions. 13 Almost all previous work has focused on substitutions at the A (4e) and B I (2a) cation sites.
Gehlenite From Three Occurrences of High-Temperature Skarns, Romania: New Mineralogical Data
The Canadian Mineralogist, 2011
Gehlenite is a common mineral in three high-temperature calcic skarns in Romania: Măgureaua Vaţei and Cornet Hill in Apuseni Mountains and Oraviţa in Banat. In all three occurrences, a gehlenite zone occurs within the contact between dioritic or monzodioritic bodies of Upper Cretaceous age and carbonaceous sequences of Mesozoic age. The melilite solid-solutions vary from Ak 34.1 to Ak 51.2 (mean Ak 41.2 ) at Oraviţa, from Ak 30.4 to Ak 42.9 (mean Ak 38.3 ) at Măgureaua Vaţei, and from Ak 24.3 to Ak 41.7 (mean Ak 32.9 ) at Cornet Hill, respectively. The content of "Na-melilite" is low, from up to 1.96 mol.% at Cornet Hill to up to 3.60 mol.% at Oraviţa. The cell parameter a ranges from 7.679(3) to 7.734(3) Å at Oraviţa, from 7.683(4) to 7.735(1) Å at Măgureaua Vaţei, and from 7.684(3) to 7.733(1) Å at Cornet Hill, whereas c varies from 5.043(3) to 5.065(3) Å at Oraviţa, from 5.040(1) to 5.070(3) Å at Măgureaua Vaţei, and from 5.044(1) to 5.067(4) Å at Cornet Hill. It is not possible to quantify the variations in the crystallographic parameters by considering only the åkermanitic substitution because the "Na-melilite" and "Fe-åkermanite" substitutions also play an important role. The cell parameter a remains practically constant within experimental errors, in spite of the increasing åkermanite-for-gehlenite substitution, due to the opposite influence of the "Na-melilite" component. The åkermanite substitution prevails over the opposite influence of "Na-melilite" in increasing the cell volume, and is additive to the influence of "Na-melilite" in decreasing the cell parameter c. The record in the infrared-absorption spectra of bands located at ~855 cm -1 and ~670 cm -1 , respectively, which may be tentatively assigned to Al-O-Al stretching, is indicative of the presence of gehlenite-rich members of the melilite group in all the three occurrences.
Journal of Mineralogical and Petrological Sciences, 2005
Crystal structure and the 57 Fe Mössbauer spectrum of synthetic (Ca 1.53 Na 0.47)(Mg 0.52 Fe 3+ 0.49)Si 1.99 O 7 melilite have been investigated to confirm the location of Fe 3+ at the T1 site, and to evaluate the effect of an incommensurate structure on 57 Fe Mössbauer hyperfine parameters. A conventional Rietveld refinement with occupancies of Ca(W), Na(W), Mg(T1), Fe 3+ (T1) and Si(T2) fixed to 0.765, 0.235, 0.515, 0.485 and 2.00, respectively, converged well with an R wp (R weighted pattern) of 13.14% and a goodness of fit of 1.337, indicating exclusive Fe 3+ distribution at the T1 site. However, the Mössbauer spectrum consists of two doublets with an isomer shift of 0.19 mm/s and quadrupole splitting of 0.75 and 1.04 mm/s, which can be attributed to Fe 3+ at the T1 site. There are two sorts of T1 site with distinguishable distortion in the synthetic Na Fe 3+ melilite, which might be attributed to the existence of an incommensurate phase at room temperature.