Smectite growth is of importance across various fields due to its abundance on the surface of both Earth and Mars. However, the impact of the crystallinity of initial materials on smectite growth processes remains poorly understood. In this study, the kinetic processes of smectite growth were examined via experimental synthesis of trioctahedral Mg-Ni saponites. Mg-Ni saponites were synthesized using mixed precursors, specifically end-member Mg-saponite and Ni-saponite, which exhibit different crystallinities. The crystal chemistry and morphology of samples were analyzed using X-ray diffraction, Fourier-transform infrared spectroscopy, and high-angle annular dark-field scanning transmission electron microscopy. The experimental results converge towards these main conclusions: (i) the formation of Mg-Ni saponite solid solutions are promoted when the precursors are small particles, whereas large-particle precursors limit their own dissolution and do not yield Mg-Ni saponite solid solutions under the experimental conditions; (ii) because Ni exhibits a greater stability within the saponite structure compared to Mg, the Mg-Ni-saponite solid solutions formed more easily from the mixture of Ni-saponite germs and well-crystallized Mg-saponite precursors than from the mixture of Mg-saponite germs and well-crystallized Ni-saponite precursors; (iii) the dissolution extent (DE) of precursor mixtures increases with longer synthesis time, higher synthesis temperature, and larger gap between synthesis temperature of precursors and of samples, and stabilizes once it reaches a certain value. Thus DE can be used to estimate the kinetics of Mg-Ni saponite crystallization from precursor mixtures. These results obtained from the experimental Mg-Ni saponite system are useful for predicting the evolution processes of smectite in natural systems.

Hydrotalcite solid solutions were prepared by coprecipitation followed by hydrothermal treatment between 150° and 250°C. Based on the structural formula [Mg1−xAlx(OH)2]x+[(CO3)x/2 • mH2O]x−, pure solid solutions were formed in the range A1/(A1 + Mg) = 0.2 to 0.33, where m = (1 − 3x/2). Maximum crystallite size was achieved by hydrothermal treatment between 180° and 200°C, with x = 0.337 to 0.429. Crystal strain was also minimized at these values of x. The adsorption capacity for Naphthol Yellow S increased as x increased and reached a maximum (1.56 × 10−6 moles/m2) when x = 0.287, a value eight times larger than that of Mg(OH)2. A weak endothermic DTA peak at about 350°C is probably due to the loss of structural water in the main layer of the structure. On calcination between 400° and 700°C, only periclase was detected, probably containing Al in solid solution. Hydration of the calcined product resulted in the reconstruction of the original hydrotalcite structure.

Electron probe analyses of diagenetic illites in Eocene sandstones from Kettleman North Dome, California, along with analyses of coexisting interstitial waters were used to calculate apparent molal free energies of formation of the illites at the in situ conditions of 100°C and 150 bars. Various triangular and rectangular compositional plots, once contoured for free energy, crudely indicate that the free energy decreases as the potassium content increases, decreases as the Al-for-Si substitution increases, and appears to be a minimum along a narrow composition valley having ~3% Fe2O3 of octahedral cations. Qualitatively, the shape of the free energy surface suggests only small departures from ideality.

The illites and water samples used in the calculations were chosen as close to each other as possible and are in equilibrium. The waters contain mainly Cl−, SO42−, Na+, and Ca2+ and have a total salinity of ~10,000 ppm. The illites, which coexist in the sandstones with diagenetic quartz, albite, and K-feldspar, are phen-gitic and contain 5–9% K2O, 1–3% total FeO, 1% MgO, and minor Na2O, MnO, and TiO2. The Fe2O3/FeO ratio has been set arbitrarily to 7 by comparison with published analyses.

Log K (100°C) was first calculated for the equilibrium illite ⇄ ions from the ionic activities obtained through speciation calculations for each water sample. The apparent free energies of formation of illites at 100°C were then calculated. The effect of instrumental uncertainty on the values of ΔGf was assessed by accepting a 2.5 or 3.5% error in the probe analyses and by using a Monte Carlo method of error propagation. The free energies of formation obtained (on the basis of O10(OH2)) range from -1280 to -1320 kcal/mole at 100°C and 150 bars, and their Monte-Carlo uncertainties are less than ±2 kcal/mole. Combining the compositions of natural, coexisting aqueous solutions and mineral solid solutions through thermodynamic equations constitutes a fast method to determine, at least crudely, the free energy-composition surface for the solid solution. This is in contrast to experimental methods such as calorimetry or hydrothermal equilibrium, which are very time-consuming and difficult.

Transmission electron microscopy has been used to characterize coexisting pyrophyllite and muscovite in low-grade metamorphosed pelites from Witwatersrand and northeastern Pennsylvania. The Witwatersrand sample consisted largely of porphyroblasts of interlayered muscovite and pyrophyllite in a fine-grained matrix of the same phases. In both textures, muscovite and pyrophyllite occurred as interlayered packets (with apparently coherent interfaces) from about 300 Å to a few micrometers in thickness, with no mixed layering. Their compositions were determined with a scanning transmission electron microscope to be

$$(\Box{_{{\rm{0}}{\rm{.11}}}}{\rm{ }}{{\rm{K}}_{{\rm{1}}{\rm{.72}}}}{\rm{Na}_{{\rm{0}}{\rm{.17}}})(A}{{\rm{l}}_{{\rm{3}}{\rm{.91}}}}{\rm{ F}}{{\rm{e}}_{{\rm{0}}{\rm{.03}}}}{\rm{M}}{{\rm{g}}_{{\rm{0}}{\rm{.05}}}}{\rm{T}}{{\rm{i}}_{{\rm{0}}{\rm{.01}}}}{\rm{)(S}}{{\rm{i}}_{{\rm{6}}{\rm{.11}}}}{\rm{ Al}}{{\rm{l}}_{{\rm{1}}{\rm{.89}}}}{\rm{)}}{{\rm{O}}_{{\rm{20}}}}{{\rm{(OH)}}_{\rm{4}}}$$

$$(\Box_{{\rm{1}}{\rm{.90}}}{\rm{N}}{{\rm{a}}_{{\rm{0.06}}}}{{\rm{K}}_{{\rm{0}}{\rm{.04}}}}{\rm{)(A}}{{\rm{l}}_{{\rm{3}}{\rm{.94}}}}{\rm{F}}{{\rm{e}}_{{\rm{0}}{\rm{.01}}}}{\rm{M}}{{\rm{g}}_{{\rm{0}}{\rm{.05}}}}{\rm{)(S}}{{\rm{i}}_{{\rm{7}}{\rm{.94}}}}{\rm{A}}{{\rm{l}}_{{\rm{0}}{\rm{.06}}}}{\rm{)}}{{\rm{O}}_{{\rm{20}}}}{{\rm{(OH)}}_{\rm{4}}},$$

The pyrophyllite and muscovite in the Pennsylvania shale likewise occurred only as coexisting coherent to sub-parallel packets as thin as 200 Å, with compositions of

$$(\Box_{{\rm{1}}{\rm{.89}}}{\rm{N}}{{\rm{a}}_{{\rm{0.04}}}}{\rm{C}}{{\rm{a}}_{{\rm{0}}{\rm{.02}}}}{{\rm{K}}_{{\rm{0}}{\rm{.05}}}}{\rm{)(A}}{{\rm{l}}_{{\rm{3}}{\rm{.93}}}}{\rm{F}}{{\rm{e}}_{{\rm{0}}{\rm{.04}}}}{\rm{M}}{{\rm{g}}_{{\rm{0}}{\rm{.02}}}}{\rm{T}}{{\rm{i}}_{{\rm{0}}{\rm{.01}}}}{\rm{)(S}}{{\rm{i}}_{{\rm{7}}{\rm{.92}}}}{\rm{A}}{{\rm{l}}_{{\rm{0}}{\rm{.08}}}}{\rm{)}}{{\rm{O}}_{{\rm{20}}}}{{\rm{(OH)}}_{\rm{4}}}$$

$$({\rm{N}}{{\rm{a}}_{{\rm{0}}{\rm{.04}}}}{\rm{C}}{{\rm{a}}_{{\rm{0}}{\rm{.02}}}}{{\rm{K}}_{{\rm{2}}{\rm{.03}}}}{\rm{)(A}}{{\rm{l}}_{{\rm{3}}{\rm{.54}}}}{\rm{F}}{{\rm{e}}_{{\rm{0}}{\rm{.24}}}}{\rm{M}}{{\rm{g}}_{{\rm{0}}{\rm{.16}}}}{\rm{T}}{{\rm{i}}_{{\rm{0}}{\rm{.06}}}}{\rm{)(S}}{{\rm{i}}_{{\rm{6}}{\rm{.09}}}}{\rm{A}}{{\rm{l}}_{{\rm{1}}{\rm{.91}}}}{\rm{)}}{{\rm{O}}_{{\rm{20}}}}{{\rm{(OH)}}_{\rm{4}}}.$$

The compositions of these coexisting pyrophyllite and muscovite define a solvus with steep limbs and extremely limited solid solution. Illite is a white mica, intermediate in composition between pyrophyllite and muscovite, formed at much lower temperatures than muscovite. These relations show that illite is metastable relative to pyrophyllite + muscovite in all of its diagenetic and low-grade metamorphic occurrences. This further implies that illite precursor phases, such as smectite, are also metastable. The prograde reactions involving smectite, illite, and muscovite are therefore inferred to represent Ostwald-step-rule-like advances through a series of metastable phases toward the equilibrium states attained in the greenschist facies. “Illite crystallinity” can therefore be a measure of reaction progress, for which temperature is only one of several determining factors.

Activity expressions are presented for a six end-member, regular solution, site-mixing model for chlorites. The end members are ideal 14-Å chlorites for which experimental stability data are lacking. Estimates were made of the standard state 25°C and 1 bar molar 3rd law entropies, volumes, and the Maier-Kelley heat capacity coefficients. Experimental stability data from the literature for 14-Å chlorites were used with different sets of exchange energies, for cations on adjacent sites, to compute estimates of the standard state chemical potentials of the end members at 25°C and 1 bar. More experimental data are needed for an adequate definition of the exchange energies and the standard state chemical potentials. The model was applied to diagenesis in clastic reservoirs. Aqueous activity ratios of Mg2+:Fe2+ were computed as an equilibrium function of the corresponding molar ratios in authigenic chlorites. The aqueous activity ratio was independent of the chlorite Al content at a constant molar ratio of Mg2+:Fe2+ in the chlorite. The model predicts a wide range of molar Mg2+:Fe2+ ratios in authigenic chlorites in equilibrium with reservoir fluids. Trends in these molar ratios should be independent of the Al content in the chlorites. The model can be applied directly to 7-Å chlorites when experimental data become available to estimate the various thermodynamic parameters of the 7-Å end members.

Chemical activity diagrarns, prepared to illustrate the properties expected if mixed-layer mont-morillonite/illite is regarded as a solid solution, are compared to those derived from a treatment of these materials as a mixture of two phases. If the system is a solid solution, the coexisting aqueous solution should range from higher dissolved silica contents in the presence of kaolinite and a montmorillonite end member to lower dissolved silica in the presence of kaolinite and an illitic end member. Silica concentration in the aqueous solution might vary by a factor of as much as six. If the system is two phase, the silica content of a solution in equilibrium with kaolinite and both phases would be fixed at a given T and P, as would a solution equilibrated with both phases and K-feldspar. In the absence of a third phase, silica in equilibrium with both phases should be nearly constant, but increase with increasing ratio of K+/H+ in solution. Available data on coexisting aqueous solutions apparently are more nearly consistent with two phases than with a solid solution.

Mn-substituted iron oxides were synthesized by coprecipitating Fe(NO3)3 and Mn(SO4) solutions with NH4OH, adjusting the suspensions to pH 4 or 6, and then keeping the suspensions at 55°C for 62 days. The Mn mole fraction of the final products ranged from 0 to 0.3. X-ray powder diffraction patterns showed that goethite and hematite formed in each Fe-containing system. Groutite formed in systems having initial Mn mole fractions ≥0.35. Only manganite and hausmannite formed in the pure Mn systems. The oxalate-soluble Fe in the samples increased as the Mn mole fraction increased and was slightly larger for the pH 6 series.

For samples that contained the largest Mn mole fraction, the b and c dimensions of the goethite unit cell were shifted toward those of groutite, and the b and c dimensions of the groutite unit cell were shifted toward those of goethite. Assuming the Vegard rule holds for the unit-cell c dimension, the goethite accommodated a maximum Mn mole fraction of 0.34, and the groutite accommodated a maximum Fe mole fraction of 0.31. The unit-cell dimensions of hematite did not vary systematically with the mole fraction of Mn in solution, probably because little Mn substituted into the hematite structure.

Iron-chromium hydroxides are important solid phases governing the aqueous concentrations of Cr(III) in soils and fly ashes. Although direct identification of noncrystalline (Fe,Cr)(OH)3 is difficult, the infrared spectra of noncrystalline Fe(OH)3 and Cr(OH)3, coprecipitated (Fe,Cr)(OH)3, and physical mixtures of Fe(OH)3 and Cr(OH)3 can be distinguished on the basis of the asymmetric stretching doublet (v3) of structural carbonate anions. As the Cr mole fraction of the coprecipitated (Fe,Cr)(OH)3 increases, the position of the low-frequency v3 peak (v3″) changes progressively to higher frequencies, and the carbonate v3 splitting decreases. No change in carbonate v3 splitting or v3″ location was observed for physical mixtures of Fe(OH)3 and Cr(OH)3. The changes in v3 splitting are believed to be caused by different degrees of polarization of the carbonate ligand by the Fe and Cr cations.

Pure Cr(OH)3 exhibits a strong affinity for carbonate and H2O and tends to remain noncrystalline even at very high pHs. In contrast, pure Fe(OH)3 gradually converts to crystalline goethite at high pH, to the exclusion of much of the H2O and carbonate. The presence of Cr in (Fe,Cr)(OH)3 solid solutions seems to inhibit the transformation to crystalline goethite. The strong association of carbonate with Cr and the kinetic inertness of Cr(III) inner-sphere complexes in general may account for the maintenance of noncrystalline solid-solution materials in lieu of transformation to a crystalline end product.

Double hydroxide solids precipitated homogeneously from three laboratory-synthesized aqueous solutions that simulated mildly contaminated surface or groundwater. Over a limited pH range, precipitates formed rapidly from dissolved ions, and more slowly by incorporating ions dissolving from other solids, including highly soluble aluminous solids. The precipitates were characterized by size and shape via transmission electron microscopy (TEM), by composition via inductively coupled plasma-mass spectrometry (ICP-MS) of mother solutions and analytical electron microscopy (AEM) of precipitates, and by structure via powder X-ray diffraction (XRD), TEM, and extended X-ray absorption fine structure (EXAFS) spectroscopy. They were identified as nanocrystalline cobalt hydrotalcite (CoHT) of the form [Co(II)1-xAl(III)x(OH)2]x+(An−x/n)·mH2O, with x = 0.17–0.25, A = CO32−, NO3−, or H3SiO4−n = anion charge and m undetermined. Complete solid solution may exist at the macroscopic level for the range of stoichiometrics reported, but clustering of Co atoms within hydroxide layers indicates a degree of immiscibility at the molecular scale. Composition evolved toward the Co-rich endmember with time for at least one precipitate. The small layer charge in the x = 0.17 precipitate caused anionic interlayers to be incomplete, producing interstratification of hydrotalcite and brucite-like layers. Solubility products estimated from solution measurements for the observed final CoHT stoichiometries suggest that CoHT is less soluble than the inactive forms of Co(OH)2 and CoCO3 near neutral pH. Low solubility and rapid formation suggest that CoHT solids may be important sinks for Co in contact with near neutral pH waters. Because hydrotalcite can incorporate a range of transition metals, precipitation of hydrotalcite may be similarly effective for removing other trace metals from natural waters.

Genthelvite from pegmatite veins hosted by the albite-enriched granite (ca.1.8 Ga) corresponding to the Sn–Nb–Ta (F, REE, Li, Zr, U, Th) Madeira deposit, Amazonas, Brazil was studied. Genthelvite, the exclusive Be-bearing mineral within the deposit, occurs as massive crystals of up to 4.7 cm in size. Compositions are homogeneous within individual crystals, although there is moderate variation in the overall composition reflecting relatively limited substitutions within the helvine–genthelvite–danalite solid-solution series, with relatively high Zn contents (36.96 to 49.45 wt.% ZnO), lower Mn contents (0.61 to 3.03 wt.% MnO), and variable Fe contents (2.10 to 10.94 wt.% FeO), completing an existing compositional gap in this system. Genthelvite formed in an alkaline and subaluminous environment, under stable conditions within the late-evolved fluids, at relatively high temperature (>400°C), in a reducing environment. The extremely high concentration of fluorine in the magma and the crystallisation of magmatic galena resulted in an effective reduction of H2S fugacity. This resulted in the stabilisation of genthelvite during the transition from the late magmatic to early hydrothermal stages of the albite-enriched granite evolution. The variability in Fe content within genthelvite is associated primarily with localised variations in the mineral assemblage (e.g. the presence of riebeckite and polylithionite). Genthelvite was altered by low-temperature aqueous fluids rich in F which resulted in the incorporation of Fe, Mn, Mg, Pb, Ba, Na, K, U and REE into the Zn2+ structural site and the allocation of excess Si, Al, Ti and P in the IVSi and IVBe structural sites. The substantial content of U and REE substituting for Zn, together with Si substituting for Be, is charge balanced by the presence of vacancies at the A site.

Hydrotalcite minerals are layered double hydroxides (LDH) which play an important role in immobilizing hazardous compounds to decontaminate industrial wastewaters. The stability of an LDH is mostly evaluated in terms of its low solubility in water. However, the solubility of divalent trace metals immobilized by Mg-Al-type LDHs is not well known. Hydrotalcites containing Zn in solid solution, (Mg+Zn)3-Al-LDH, were synthesized by alkaline co-precipitation. A series of eleven LDH phases with Zn mole fractions XZn = Zn/(Mg+Zn) of 0–1 were characterized by powder X-ray diffractometry (XRD), Fourier-transform infrared (FTIR) spectroscopy, thermogravimetry (TGA), scanning and transmission electron microscopy (SEM/TEM), Brunauer-Emmett-Teller (BET) surface area analysis, and inductively coupled plasma mass (ICP-MS) spectrometry. The XRD analysis provided sharp characteristic spacings for d003 and d006 which occurred for all samples, confirming a layered LDH structure. Cell parameters (a, c) obeyed Vegard’s law and confirmed the formation of a regular solid-solution series without a mixing gap. An aqueous equilibrium time was determined by kinetic dissolution experiments. Steady-state solubility occurred after 120 days, but the experiments continued up to 240 days. The XRD and SEM/TEM analyses indicated no phase changes during the long-term dissolution experiments; neither were phase impurities detected after 240 days. The solubility products of the Mg- and Zn-bearing endmember compositions were calculated from experimentally determined total cation and anion concentrations using the Visual Minteq code for considering element speciation and ion pairing. The solubility product decreased as the Zn mole fraction increased, suggesting that the Zn-bearing LDH phases were more stable than the pure Mg3-Al-LDHs. Solid-solution aqueous-solution thermodynamic equilibrium modeling using the Lippmann “total solubility product” approach and applying Lippmann diagrams with logarithmic x-axes revealed a log-linear decrease in aqueous Zn solubility. The results are promising for remediation of metal-bearing liquid wastes because the metals that co-precipitated with the LDH were more strongly retained and, therefore, less soluble than the hydroxides or carbonates of the trace metal.

Low-temperature chlorites formed in diagenetic to low-grade metamorphic environments generally have greater Si contents and larger numbers of octahedral vacancies, and smaller Fe+Mg contents than higher-grade metamorphic chlorites. The compositional variations are characterized approximately by four end-member components: Al-free trioctahedral chlorite, chamosite, corundophilite, and sudoite. The solid solution is considered to be a random mix of cations and vacancies in the octahedral sites. Using the compositions of chlorites from Niger, Rouez, and Saint Martin diagenetic-hydrothermal series, a new, more convenient geothermometer, applicable to low-T chlorites is proposed and comparison made with geothermometers proposed previously. The chlorites studied contain appreciable amounts of Fe(III) (>14% of the total Fe), determined by Mössbauer spectroscopy. The calculations under which all Fe was regarded as ferrous gave considerable overestimates for the formation temperature, irrespective of the geothermometer used. This problem was reduced by taking into account the presence of Fe(III) in the octahedral sites. The geothermometer from this study gave more reasonable estimates than the geothermometers proposed by Walshe (1986) and Vidal et al. (2001), particularly in the case of the Niger chlorites which crystallized in the lowest-temperature conditions. The ordered-site substitution model of solid solution developed by Vidal et al. (2001) predicted satisfactorily the formation temperature of the Rouez chlorites and of some of the Saint Martin chlorites, suggesting that the chlorite compositions are controlled by the exchange at low-T conditions while they are controlled by Tschermak exchange at higher temperatures. The decreasing number of vacancies with temperature are poorer in Fe-rich than in Fe-poor chlorites. Furthermore, the ordered-site occupation of cations and vacancies in trioctahedral chlorite occurs concomitantly with the compositional changes ruled by increasing temperature conditions.

A high-grade ore sample from the Cu–Zn–Au Photo Lake volcanogenic massive sulfide deposit (Flin Flon–Snow Lake greenstone belt, Manitoba, Canada) contains a Bi–Ag sulfo-selenide with a composition situated approximately in the middle of the S–Se substitution range (Se ≈ 0.86 apfu and S ≈ 1.05 apfu). These new data, combined with a literature compilation of all publicly available matildite and bohdanowiczite compositional data, reveal a nearly complete range of S–Se substitution between these two minerals, with only the section between BiAgSe0.78S1.18 and BiAgSe0.25S1.75 – about a quarter of the complete S–Se range – not yet documented. These observations suggest that a complete solid-solution series between matildite and bohdanowiczite, as previously suspected, might exist and in a manner similar to the galena–clausthalite complete solid-solution series.

The crystal chemistry and crystal structure of the rare earth element phosphates, monazite-(Ce), Ce(PO4), and xenotime-(Y), Y(PO4), as well as the arsenates, gasparite-(Ce), Ce(AsO4), and chernovite-(Y), Y(AsO4), from the hydrothermal quartz-bearing fissures, related to pegmatites overprinted by amphibolite facies, cropping out at Mt. Cervandone, Western Alps, Piedmont, Italy, have been investigated by means of electron microprobe analysis in wavelength dispersion mode and single-crystal X-ray diffraction. The chemical data reveal the occurrence of a full solid solution among the isostructural chernovite-(Y) and xenotime-(Y) with tetragonal symmetry, whereas a wide miscibility gap is observed for the isostructural gasparite-(Ce) and monazite-(Ce) of Mt. Cervandone, with monoclinic symmetry. A significant chemical heterogeneity has been observed for several investigated samples, especially related to the Th content, which is locally enriched in ThSiO4 grains. The analysis of the refined structural models demonstrates the significant control played by the composition of the tetrahedrally-coordinated (As,P)-bearing sites on the bulk unit-cell volume, and on the size and shape of the (REE)-coordination polyhedra.

Pyrrhotite Fe1–xS is the main sulfide component of platinum-group element (PGE) ores and commonly contains from a few tenths to a few dozen ppm of disseminated Pt. Here we report an X-ray absorption spectroscopy investigation into the state of Pt in synthetic pyrrhotite in combination with theoretical spectra modelling. The pyrrhotite crystals were obtained by means of the salt flux technique, using a eutectic mixture of alkali metal halides as the transport media. Analysis of the chemical composition of synthesised crystals showed that an increase of the temperature and sulfur fugacity yields higher concentrations of Pt in pyrrhotite. The Pt content reached 0.6 wt.% at the maximum temperature and sulfur fugacity (t = 720°C, log $f_{{\rm S}_ 2}$ = –0.1) achieved in a Pt-saturated system. X-ray absorption near-edge structure (XANES) analysis of Pt L3-edge spectra revealed that Pt is present in pyrrhotite in the 4+ and 2+ ‘formal’ oxidation states. Theoretical modelling of XANES and interpretation of extended X-ray absorption fine structure (EXAFS) spectra showed that Pt4+ substitutes for Fe in the crystal lattice of pyrrhotite, whereas Pt2+ forms PtS-like clusters disseminated in the pyrrhotite matrix. Atoms of isomorphous Pt4+ are surrounded by 6 S atoms at a distance of 2.39 ± 0.02 Å. According to theoretical simulations using the FDMNES program, the second coordination sphere of the solid-solution Pt contains one vacancy in the Fe sublattice within the Fe-layer. The Pt2+S-like clusters can be considered as a quench product. High sulfur fugacity stabilises the solid-solution Pt and prevents the formation of the PtS-like clusters during cooling. The maximum content of the solid-solution Pt in pyrrhotite is ca. 50 times lower than in pyrite and can be approximated by a straight line in the log C(Pt) vs. 1/T plot, it increases from 1 ppm at 350°C to 3 wt.% at 900°C.

The crystal structure of Ni-rich gordaite–thérèsemagnanite has been determined from a sample collected at pillar 80 in the North mine, Cap Garonne, Var, France. The structure was refined to R1 = 0.0693 for 935 reflections with I > 2σ(I). The mineral is isostructural with gordaite, forming a layered structure with an extensive hydrogen-bonding network. The possible polytypic relationship between gordaite, thérèsemagnanite and guarinoite is also discussed. The guarinoite formula (Zn,Co,Ni)6(SO4)(OH,Cl)10·5H2O is also likely to be incorrect and is more likely to be Na(Zn,Co)4(SO4)(OH)6Cl·5–6H2O, meaning that guarinoite is equivalent to Co-rich gordaite-2H and would not be a distinct mineral species.