Magnetite is present in most carbonatites, and in the most abundant and best-known form of carbonatite, coarse-grained intrusions, it typically falls in a narrow composition range close to Fe3O4. A fine-grained carbonatite from Zambia contains magnetites with an extraordinary array of compositions (from 18–1% TiO2, 10–2% Al2O3, and 16–4% MgO) outranging previously-reported examples. Zoning trends are from high TiO2 to high Al2O3 and MgO. No signs of exsolution are seen. Checks on similar rocks from Germany, Uganda and Tanzania reveal magnetites with comparable compositions, ranges, and zoning. Magnetites from alkaline and alkaline ultramafic silicate volcanic rocks cover only parts of this array. Magnetite analyses from some other fine-grained carbonatites, reported in the literature, fall in the same composition field, suggesting that this form of carbonatite may be distinctive. The chemistry and zoning would be consonant with rapid high-temperature crystallization in the carbonatite melts, with the lack of exsolution pointing to fast quenching: this contrasts with coarse-grained intrusive carbonatites, in which the magnetite compositions are attributed to slow cooling, with final equilibration at low temperature. In some complexes, both forms of carbonatite, with their different magnetite compositions, are represented.

Post-Variscan vein-type Co-Ni-Bi ores of the Bieber deposit, Spessart mountains, Germany, which are related to the Permian Kupferschiefer, have been investigated by ore microscopy, X-ray powder diffraction and electron-probe microanalysis. The samples contain a variety of ore minerals, notably skutterudite, native bismuth, cobaltite, alloclasite, niccolite, maucherite, gersdorffite, rammelsbergite/pararammelsbergite, safflorite, loellingite and emplectite. The ores display structures indicative of multiple brecciation and complex zoned arsenide assemblages. Three sequential stages of deposition are identified, which are (1) the Cu stage, (2) the main Co-Ni-Bi stage, and (3) the late stage. The arsenide minerals, notably skutterudite, diarsenides and sulpharsenides, show a large range of compositional variation in Co-Ni-Fe space. A relatively limited number of skutterudite and diarsenide compositions lie outside the compositional fields established in the literature. Skutterudite and diarsenides are characterized by a significant substitution of As by S up to 0.44 a.p.f.u. and 0.31 a.p.f.u., respectively, which is larger than the range previously reported for these minerals. Sulpharsenide compositions can be grouped into three populations, which conform to cobaltian arsenopyrite, cobaltite and gersdorffite. They display highly variable As/S ratios between 0.95:1.00 and 1.29:0.73, consistent with experimental data. Estimates of the formation temperatures, based on the presence of dendritic native bismuth and emplectite, are in the range 100–300°C, similar to different post-Variscan mineralization styles widespread in Central Europe. Comparison of the Co-Ni-Bi vein assemblage with the framework of available paragenetic information and radiometric age data for regional mineralization events indicates an age of mineralization of ~150–160 Ma for the Bieber deposit.

Edingtonite and yugawaralite showing sectoral textures were studied by polarized optical microscopy and X-ray analysis. In edingtonite, the m{110} sector (2Vα = 22°) is optically triclinic and the c{001} sector (2Vα = 52°) is orthorhombic. In yugawaralite, the k{011} sector is optically monoclinic, whereas the v{120} sector is triclinic. Their crystal structures were determined. The results of refinement showed that the space groups of the k{011} (Rw = 4.5%) and v{120} (Rw = 5.1%) growth sectors are monoclinic Pc and triclinic P1, respectively. In the v{120} sector, several interatomic distances, bond angles and site occupancies are different with respect to a symmetrical plane of the structure, and therefore the monoclinic c glide is extinct. Thus, the X-ray symmetry correlates with the optical one. From the relationship between the surface and internal texture, the symmetry and sector can be explained by cation (Al/Si) ordering during non-equilibrium crystal growth.

Ilmenites from the least-altered rocks of the Beja-Acebuches Ophiolite Complex (SE Portugal), with low Ti values and excess Fe, despite rare optical evidence of hematite exsolution, were studied by 57Fe Mössbauer spectroscopy and X-ray diffraction. According to single-crystal XRD the sequence of alternate layers characteristic of the ideal ilmenite structure is preserved, the excess Fe being accommodated in the Ti layers. No superparamagnetic oxides were detected by 57Fe Mössbauer spectroscopy. The typical spectra of bulk αFe2O3 and of Fe3+-containing ilmenite, in the paramagnetic state above 49 K and magnetically ordered at 6 K, are observed. The average degree of oxidation of the ilmenites, estimated from the chemical analysis assuming ideally stoichiometric full cation site occupancies, is also confirmed by 57Fe Mössbauer data. Since our crystal chemistry study gave no evidence of crypto-exsolution textures within the ilmenite with the observed compositions, fast cooling from magmatic temperatures and decomposition of ilmenite in supergene conditions is suggested.

The equation of state of synthetic deuterated burtite, CaSn(OD)6, has been determined to 7.25 GPa at 298 K by synchrotron X-ray powder diffraction. Fitting to a third-order Birch-Murnaghan equation of state gives K0 = 44.7(9) GPa and K0′ = 5.3(4). A second-order fit gives K0 = 47.4(4) GPa. Within experimental error the two fits are indistinguishable over the pressure range studied. The decrease in the a parameter with pressure is smooth and no phase transitions were observed. Burtite is much more compressible (by a factor of three or four) than CaSnO3 and CdSnO3 perovskites, indicating that the absence of a cavity cation has a major effect upon the compressibility of the octahedral framework. Burtite is also markedly more compressible than the closely-related mineral stottite FeGe(OH)6 (K0 = 78 GPa). Their different compressibilities correlate with the relative compressibilities of stannate and germanate perovskites. Although different octahedral compressions are likely to be the primary reason for the different compressibilities of burtite and stottite, we also consider the possible secondary role of hydrogen-bonding topology in affecting the compressibilities of protonated octahedral frameworks. Burtite and stottite have different hydrogen-bonding topologies due to their different octahedral-tilt system. Burtite, space group Pn and tilt system a+a+a+, has a hydrogen-bonded network of linked four-membered rings of O-H…O linkages, whereas stottite, space group P42/n and tilt system a+a+c−, has <100> O-H…O crankshafts and isolated four-membered rings. These different hydrogen-bonded configurations lead to different bracing of the empty cavity sites by the O-H…O linkages and very different hydrogen-bonding connectivities in these two minerals that may also enhance the difference between their compressibilities.

The structure of a natural sample of chondrodite (Mg4.89Fe0.07Si2.04O8F1.54(OH)0.46) was refined using powder neutron diffraction data and the Rietveld technique (P21/b; Z = 2; a = 4.7204(1)Å; b = 10.2360(3)Å; c = 7.8252(2)Å; α = 109.11(1)°; V = 357.26(2)Å3). Hydrogen was found to occupy the H1 site. The significance of hydrogen at this site is discussed in terms of hydrogen-bond stabilization of humite structures containing varying amounts of OH, F and Ti. Arguments are proposed as to why the F and Ti contents of natural humites usually result in only one H per formula unit when there is no crystal-chemical reason why fully hydrated samples should not be favoured.

The results of a HRTEM and electron diffraction investigation of a Cu- and Ag-rich cosalite, Cu1.2Ag0.9Pb6.7Bi8.2S20, from Ivigtut, Greenland, are presented. Electron diffraction patterns down [001] show pairs of satellite reflections at Gp±~1/8[140]* and Gp±~1/8[40]* in addition to the strong Bragg reflections of the underlying average structure. The superlattice corresponds to a compositional and displacive modulation G±qprim where qprim is the primary modulation vector 1/8[140]*. The superlattice is attributed to coupled Ag/Pb/Bi and Cu/□ ordering and associated structural relaxation. The HRTEM images show a complex moire texture due to the structural modulations. Cosalite from Ivigtut, Greenland, also forms coherent intergrowths with members of the lillianite–gustavite series (Pb3Bi2S6–PbAgBi3S6). A model for the intergrowth between the structures is presented based on sharing both S and Pb atoms in the PbS-like slabs in both structures along [010]cosalite.

Hoganite, copper(II) acetate monohydrate, and paceite (pronounced ‘pace-ite’), calcium(II) copper(II) tetraacetate hexahydrate, occur as isolated crystals embedded in ferruginous gossan from the Potosi Pit, Broken Hill, New South Wales, Australia. They are associated with goethite, hematite, quartz, linarite, malachite, azurite, cerussite and cuprian smithsonite. Hoganite is bluish green with a pale blue streak and a Mohs hardness of 1½; it possesses perfect {001} and distinct {110} cleavages and has a conchoidal fracture. Chemical analysis of hoganite gave (wt.%) C 23.85; H 3.95; Cu 31.6; Fe 0.4; O (by difference) 40.2, yielding an empirical formula of C4H7.89O5.07Cu1.00Fe0.01. The simplified formula is C4H8O5Cu or Cu(CH3COO)2.H2O, the mineral being identical to the synthetic compound of the same formula. Single-crystal X-ray data for hoganite are: monoclinic, space group C2/c, a = 13.162(3), b = 8.555(2), c = 13.850(3)Å, β = 117.08(3)°, Z = 8. The density, calculated from single-crystal data, is 1.910 g cm−3. The strongest lines in the X-ray powder pattern are [dobs (Iobs) (hkl)] 6.921 (100) (011); 3.532 (28) (202); 6.176 (14) (200); 3.592 (11) (22); 5.382 (10) (11); 2.278 (10) (204); 5.872 (9) (002). Hoganite (orientation presently unknown) is biaxial positive with α = 1.533(2), β = 1.541(3), γ = 1.554(2), 2V(meas.) = 85(5)°, 2V(calc.) = 76.8°, dispersion is r < v, medium (white light); it is strongly pleochroic with X = blue, Y = pale bluish, Z = pale bluish green and absorption X > Y > Z. The mineral is named after Graham P. Hogan of Broken Hill, New South Wales, Australia, a miner and well-known collector of Broken Hill minerals.

Paceite is dark blue with a pale blue streak and a Mohs hardness of 1½; it possesses perfect {100} and {110} cleavages and has an uneven fracture. Chemical analysis of paceite gave (wt.%) C 21.25; H 5.3; Ca 9.0; Cu 14.1; O (by difference) 50.35, yielding an empirical formula of C8H23.77O14.23Ca1.02-Cu1.00. The simplified formula is C8H24O14CaCu or CaCu(CH3COO)4.6H2O, the mineral being identical to the synthetic compound of the same formula. Unit-cell data (refined from X-ray powder diffraction data) for paceite are: tetragonal, space group I4/m, a = 11.155(4), c = 16.236(17)Å, Z = 4. The density, calculated from refined cell data, is 1.472 g cm−3. The strongest lines in the X-ray powder pattern are [dobs (Iobs) (hkl)] 7.896 (100) (110); 3.530 (20) (310); 5.586 (15) (200); 8.132 (8) (002); 9.297 (6) (101); 2.497 (4) (420); 3.042 (3) (321). Paceite is uniaxial positive with ω = 1.439(2) and ɛ = 1.482(3) (white light); pleochroism is bluish with a greenish tint (O), pale bluish with a greyish tint (E), and absorption O ⩾ E. The mineral is named after Frank L. Pace of Broken Hill, New South Wales, Australia, an ex-miner and well-known collector of Broken Hill minerals.