Beryl crystals from hydrothermal veins in the Colombian emerald mining district, were examined to understand their growth and dissolution processes. Chemical analysis reveals minor substitution of Al by Na, Mg and the colouring elements V, Cr and Fe. Growth features include characteristic indentations on (0001) and a ridge-and-valley structure on the second-order prism, which is also observed on rare conical growth faces. Zoning is the primary internal growth feature, investigated in cross sections parallel to the main growth directions, the a- and c-axes of beryl. All crystals show sector zoning, with enrichment of Na and Mg in the c-sector, whereas Al and most of the trace elements measured are concentrated in the a-sector. H2O and CO2 molecules in the crystal structure were identified by infrared spectroscopy. Two types of H2O are present, with the H–H vector aligned (type I) and perpendicular (type II) to the c-axis, with H2O II predominant in the c-sector. CO2 is present in both sectors, decreasing from core to rim, and is higher in the a-sector. Variable sector boundaries suggest irregular changes in the growth rate in the two growth directions. The substitution (VIMg+channelNa)c-sector = (VIAl+channel□)a-sector creates a pattern of stripes, originating at the steps of the sector boundaries. Etch pits as a result of dissolution are arranged in chains, typically forming etching channels, but the overall amount of dissolution is minor. The arrangement of the etch pits indicates that they formed on dislocation bundles, which originate at the sector boundaries. The observations indicate rapid crystallisation with skeletal growth in [0001] is responsible for the distribution of elements in sectors, and are consistent with a closed-system behaviour in the veins.

Selenolaurite, ideally RuSe2, is a new mineral, the first natural ruthenium selenide. It was discovered in an assemblage with Se-bearing moncheite. Both form xenomorphic inclusions in the crystal aggregates of Os–Ir–Ru minerals found at the Ingul gold placer, Urals, Russia. In addition a mineral with selenolaurite composition was found as a euhedral inclusion within grains of Pt–Fe alloy with isoferroplatinum composition at the Kazan gold placer. These placers are situated in the Chelyabinsk district, South Urals, Russia. The selenolaurite from the Ingul placer forms interstitial grains with maximum size of section of 0.05–0.1 mm. Crystals of the selenolaurite from the Kazan placer reach 20 µm in size. Selenolaurite is grey with metallic lustre and is isotropic. Reflectance values [R (%) for COM approved wavelengths (nm)] are 45.8(470), 44.3(546), 43.8(589) and 43.1(650). The chemical composition of the holotype from the Ingul placer corresponds to the empirical formula (Ru0.99Ir0.05)Σ1.04(Se1.92Te0.03S0.01)Σ1.96. Selenolaurite is the selenium-dominant analogue of laurite, RuS2 with a pyrite-type structure. It is cubic, space group Pa3̅, a = 5.9424(2) Å, V = 209.84 2) Å3, Z = 4 and Dcalc. = 8.415 g·cm–3 (calculated on the basis of empirical formula and unit-cell parameters refined by the Rietveld method). The crystal structure has been refined from the powder data to RB = 0.0067. The strongest lines of the powder X-ray diffraction pattern [d(Å), (I), (hkl)] are: 3.434(41)(111), 2.973(90)(200), 2.6580(100)(210), 2.4264(84)(211) and 1.7913(87)(311). The possible sources of a Ru–Se mineralisation in the South Urals are ophiolitic ultramafic rocks enriched in Ru and depleted with sulfur.

A new oxalate mineral species, edwindavisite, ideally Cu(C2O4)(NH3), was discovered in specimens collected from the Rowley mine, Maricopa County, Arizona, USA. It occurs as fans or sprays of bladed or prismatic crystals (up to 0.50 × 0.08 × 0.06 mm), associated intimately with ammineite, a sampleite-like mineral, baryte, ebnerite, wulfenite and quartz. Edwindavisite is green, transparent with a pale green streak and has a vitreous lustre. It is brittle and has a Mohs hardness of ∼2; cleavage is perfect on {100}. No parting or twinning was observed. The measured and calculated densities are 2.55(2) and 2.53 g/cm3, respectively. Optically, edwindavisite is biaxial (+), with α = 1.550(2), β = 1.559(2), γ = 1.755(5), 2Vmeas. = 26(2)° and 2Vcal. = 26.4°. Electron microprobe analyses yielded the empirical formula (based on Cu = 1 apfu) Cu1.00(C2O4)(NH3)0.99.

Edwindavisite is the natural counterpart of synthetic catena-μ-oxalato-ammine-copper(II), Cu(C2O4)(NH3). It is orthorhombic with space group Pbca and unit-cell parameters a = 11.1998(10), b = 9.4307(9), c = 8.3977(7) Å, V = 886.98(14) Å3 and Z = 8. In the edwindavisite structure, each Cu2+ cation is coordinated by (5O + N), forming a rather distorted and elongated octahedron. The Cu-octahedra share corners with one another to form chains extending along [001], which are joined together by oxalate (C2O4)2– groups, giving rise to layers parallel to (100). These layers are linked together by N–H···O hydrogen bonds. Among 37 oxalate minerals documented to date, edwindavisite is the first one that contains ammonia (NH3).

The new mineral zincostottite (IMA2024-024), ZnGe(OH)6, was found on specimens from the Tsumeb mine, Tsumeb, Namibia, where it is a secondary oxidation-zone mineral. It occurs as heavily etched remnants of equant or tabular crystals, up to ∼1 mm in diameter. Crystals are colourless and transparent, with vitreous to subadamantine lustre and a white streak. The mineral is brittle with irregular stepped fracture. The Mohs hardness is ∼4.5. Cleavage is good on {100} and poor on {001}. The calculated density is 3.834 g·cm–3. Optically, zincostottite is uniaxial (–) with ω = 1.785(5) and ε = 1.765(5) (white light). The empirical formula is (Zn0.77Fe3+0.23)Σ1.00Ge1.00O6H5.77. Zincostottite is tetragonal, space group P42/n, with cell parameters: a = 7.4522(18), c = 7.4000(8) Å, V = 411.0(2) Å3 and Z = 4. The crystal structure (R1 = 2.65% for 452 I > 2σI reflections) is the same as that of stottite with Zn in place of Fe2+

The metamorphic history of the eastern Beartooth Mountains of Montana in the northern Archean Wyoming Province is challenging to establish given the multi-staged nature of metamorphism in the area and the prevalence of metamorphic overprinting. A petrologically illustrative cordierite-orthopyroxene xenolith in ∼2.80 Ga calc-alkaline granitic igneous rocks exhibits a mineral assemblage responsive to multiple metamorphic overprints. Cordierite-orthopyroxene and/or -orthoamphibole rocks (CORs) have mineral assemblages that provide instructive information on the metamorphic development of the terranes and the processes that form their protoliths. This investigation combines detailed petrographic and mineral-chemical investigations with petrological modelling of this COR locality, to establish important aspects of the local pressure-temperature (P-T) history.

The COR lithology, found in the Quad Creek locality of the study area, preserves textural evidence for three discrete metamorphic events consisting of a granulite-facies event (M1) that is overprinted by two, lower-grade hydration events: an upper amphibolite-facies event (M2) and a greenschist-facies event (M3). The M1 event is characterised by the mineral assemblage rutile-biotite-orthopyroxene-cordierite-quartz-melt ± sillimanite that equilibrated at ∼0.60 GPa and 775–815°C. M2 overprinting is marked by partial hydration of orthopyroxene to anthophyllite, crystallisation of plagioclase from a putative melt and ilmenite–rutile co-existence. The final metamorphic stage, M3, is marked by local hydration of orthorhombic inosilicates to talc and cordierite to pinite (chlorite). Estimates of M2 P-T conditions suggest development below ∼0.40 GPa at 615–725°C, whereas M3 probably developed below 0.40 GPa and 500–600°C. The resultant clockwise cordierite-orthopyroxene P-T path resembles modelled metamorphic P-T paths in proximal xenolithic lithologies. This interpretive P-T history is consistent with metamorphism in an Archean continental arc subduction zone with M1 metamorphism due to compression and tectonic burial, M2 due to hydration following, or coeval, with emplacement of voluminous ∼2.80 Ga plutons and M3 due to later low-T fluids. Although not conclusive, the Quad Creek COR was probably derived from a metasomatised basaltic protolith from the Archean seafloor, a comparable mode of origin to many COR occurrences worldwide.

The crystal structure of cummingtonite-(P21/m) was characterised by single-crystal structure-refinement, infrared spectroscopy and 57Fe Mössbauer spectroscopy. Previous cummingtonite-(P21/m) specimens characterised have Mn2+ as the dominant constituent at M(4) but this amphibole has Fe2+ dominant at M(4). The formula of the amphibole was corrected for minor exsolved calcium-amphibole and is (Mg5.66Fe2+1.28Mn0.06)Σ7.00Si8.00O22(OH)2. The crystal structure (a = 9.4885(19), b = 18.040(4), c = 5.2891(11) Å, β = 102.06(3)°, V = 885.4(3) Å3, space group P21/m and Z = 2), was refined to an R1-index of 3.34% for 2338 observed reflections. Site-occupancy refinement gave the following site-populations: M(1) = 1.972(8) Mg + 0.028 Fe2+, M(2) = 2.000 Mg, M(3) = 0.989(6) Mg + 0.011 Fe2+, M(4) = 0.815(8) Mg + 1.125 Fe2+ + 0.060 Mn2+ apfu. Infrared spectroscopy in the principal (OH)-stretching region shows two peaks, at 3367 and 3652 cm–1, that were assigned to the local arrangements M(1)MgM(1)MgM(3)Mg–OH and M(1)MgM(1)Fe2+ M(3)Mg–OH (≈ M(1)MgM(1)MgM(3)Fe2+–OH) with relative intensities in accord with the refined site-populations. 57Fe Mössbauer spectrum shows three quadrupole-split doublets with parameters indicative of octahedrally coordinated Fe2+ at M(4) and M(1,2,3), and octahedrally coordinated Fe3+ that occurs in exsolved calcium amphibole. All three techniques indicate a small amount of Fe2+ at M(1,2,3) despite the fact that there is more than sufficient CMg to completely fill the M(1,2,3) sites: 5.66 Mg pfu. Issues involving the current and possible future nomenclature and classification of the magnesium-iron-manganese amphiboles are discussed in detail.

The new mineral barronite (IMA 2024-053), (□1.5Ba0.5)2(UO2)2Si5O12(OH)·2H2O, was found in the material from the Menzenschwand uranium deposit, Black Forest Mts., Germany, where it occurs as globular/acicular aggregates, consisting of long-prismatic crystals, up to 0.3 mm in length, in baryte and quartz-based gangue. Barronite is not associated with any other supergene minerals. Crystals are pale yellow with a colourless to pale yellow streak. Nevertheless, some of the crystals have a brown-orange tint, caused by Fe–Si-gels. The tenacity is brittle, the Mohs hardness is 1–2. The mineral has distinct cleavage on {100}; the fracture is uneven. Barronite is biaxial (+), with α = 1.599(2), β = 1.607(2), γ = 1.617(3); and 2V (meas.) = 86°. Optical orientation is X = b, Y ˄ a ≈ 3° in the obtuse angle β. Dispersion is distinct r>v. Pleochroism is distinct in hues of pale yellow, X<Y<Z. Electron microprobe analyses provided (based on 19 O atoms) (□1.369Ba0.345K0.165Ca0.086Pb0.024Fe0.011)Σ2.000(U0.996O2)2Si4.989O12(OH)·2H2O. Barronite is monoclinic, C2/m, a = 14.2115(11) Å, b = 14.0169(19) Å, c = 9.6545(8) Å, β = 111.59(6)°, with V = 1788.2(8) Å3 (Z = 4), refined from the corrected 3D ED data at 94K. The crystal structure refinement (R1 = 0.0791 for 6596 [I > 3σ(I)] reflections) refined from the 3D ED data confirmed that barronite has the same structural architecture as weeksite; however, it contains less H2O in the channels of the uranyl-silicate framework structure.

Aluminium-rich tourmaline can contain significant amounts of Li. Until now, syntheses have not been successful in producing Li-rich tourmalines. Because it is still not clear how Li enters the Y site in tourmaline, possible short-range orders, including Li, are discussed using bond valence calculations. Structural arrangement graphs of the Y-site neighbourhood in the structure of elbaitic tourmalines were investigated to determine more information about their stability. Possible short-range ordering including Al and Li at the Y sites, with Na, Ca or □ (vacancy) at the X site, with Si (and Al or B) at the T sites and either (OH) or F at the W site were investigated. Tourmaline with varying amounts of Na but no Ca can only contain ≤1 apfu (atoms per formula unit) YLi. This is consistent with the composition of synthetic tourmaline. Tourmaline with higher Li content (Li >1 apfu) can form when Ca is included. Such tourmaline requires fluorine because more Li results in O1 underbonding, whereas more Al at the Y site leads to O1 overbonding. Underbonding of O1 is preferable for F because OH at the O1 site usually has a bond valence sum (BVS) higher than 1.00 vu (valence units) due to hydrogen bonding of H to ring oxygen atoms. Therefore, liddicoatitic tourmaline is enriched in F and is usually F-dominant. If no fluorine is available in the starting material of a tourmaline synthesis, it is likely that significant proportions of cations such as B and Al will be incorporated into the tetrahedral site. Ultimately, however, only a smaller proportion of Li can be incorporated. Tourmalines with such tetrahedral cations, which also contain a significant Ca content, could theoretically also have significant Y-site vacancies. In order to synthesise Li-rich tourmalines, we would therefore recommend that the starting material also contains both Ca and F.

Annivite-(Zn), Cu6(Cu4Zn2)Σ6Bi4S13, is a new IMA-approved mineral species from the Geister vein, Jáchymov ore district, Czech Republic. It occurs as anhedral grains, up to 50 μm in size, and growth zones, up to 100 μm in thickness, hosted by oscillatory zoned annivite-(Zn)/tennantite-(Zn) grains, and associated with Bi-rich tennantite-(Zn), tennantite-(Fe), tetrahedrite-(Zn), the not-yet approved ‘annivite-(Fe)’, bismuth, emplectite, wittichenite and supergene bismite, walpurgite and metazeunerite. In reflected light, annivite-(Zn) is isotropic, pale grey with a brownish shade and very rare pale brown internal reflections. Reflectance data for the four COM wavelengths in air are [λ (nm): R (%)]: 470: 32.3; 546: 32.0; 589: 32.0; 650: 31.6. Electron microprobe analysis gave (in wt.% – average of 5 spot analyses): Cu 36.29, Ag 0.14, Fe 0.08, Zn 7.11, Pb 0.19, As 6.07, Sb 4.50, Bi 21.08, S 23.68, total 99.14. On the basis of ΣMe =16 atoms per formula unit, the empirical formula of annivite-(Zn) is Cu10.13Ag0.02Zn1.93Fe0.03Pb0.02Bi1.79As1.43Sb0.66S13.10. Annivite-(Zn) is cubic, I$\bar 4$3m, with unit-cell parameters a = 10.3545(6) Å, V = 1110.16(19) Å3 and Z = 2. Its crystal structure was refined by single-crystal X-ray diffraction data to a final R1 = 0.0493 on the basis of 278 unique reflections with Fo > 4σ(Fo) and 23 refined parameters. Annivite-(Zn) is isotypic with other tetrahedrite-group minerals. Its crystal chemistry is discussed, and previous findings of Bi-rich tetrahedrite-group minerals are briefly reviewed, along with the description of a second finding of annivite-(Zn) from the abandoned Mauritius tin mine, Hřebečná, Krušné hory Mountains, Czech Republic.

A calcium-silicate xenolith (no. 11) from the ignimbrite of the Upper Chegem Caldera in Kabardino-Balkaria, Russia, has revealed a diverse mineral assemblage with As- and B-bearing phases from the apatite supergroup such as the svabite and johnbaumite–hydroxylellestadite series, in addition to cahnite and datolite. Three distinct zones of variable arsenic content have been investigated. Notably, the outermost altered zone adjacent to the ignimbrite hosts the highest concentration of arsenic and arsenate minerals. A detailed structural analysis using Raman spectroscopy was carried out to investigate the distribution of boron and arsenic in tetrahedral coordination. This has provided the basis for describing a solid-solution system between hydroxylellestadite, svabite and johnbaumite and can be used as a novel technique for identifying apatite-supergroup minerals. One aim of the analysis was to elucidate the origin of various elements and content levels, particularly in relation to the distance from the xenolith–ignimbrite contact. The presence of boron and arsenic, probably derived from ignimbrites, highlights the important role of volcanic rocks as potential contributors of these elements in mineral formation processes.

Titania slag, produced from smelting placer ilmenite concentrates and used as a feedstock for TiO2 pigment production, contains low levels of radioactivity due to thorium and uranium. This study investigated the distribution and speciation of thorium in Rio Tinto Chloride Slag (RTCS), which contains an average of 170 ppm Th and 16 ppm U, using a variety of analytical methods from powder X-ray diffraction (PXRD) analysis to bulk and laser ablation inductively coupled plasma mass spectrometry (ICP-MS and LA-ICP-MS), electron microprobe analysis (EMPA), quantitative evaluation of materials by scanning electron microscopy (QEMSCAN), Raman spectroscopy, microbeam synchrotron X-ray fluorescence (µsXRF) mapping, synchrotron Laue X-ray diffraction (LXRD) and synchrotron X-ray absorption spectroscopy (XAS). Our data demonstrate that ∼99.4% of Th in the RTCS is hosted by a chevkinite-like Th–REE–Ti aluminosilicate containing an average of 8.05±0.64 wt.% ThO2. The Th–REE–Ti aluminosilicate occurs as acicular (∼0.3×12 µm) or tabular (∼5×15 µm) crystals in association with a Th-bearing aluminosilicate glass (0.41±0.35 wt.% ThO2) as infillings either in interstitials or along the fractures of the main Ti–Fe oxides of the sassite–ferropseudobrookite solid-solution series. The Th–REE–Ti aluminosilicate and associated Th-bearing aluminosilicate glass formed probably during the quenching stage of the titania slag production. LA-ICP-MS analyses and µsXRF mapping show that the main Ti–Fe oxides in the RTCS contain an average of only 0.32±0.60 ppm Th. Future pyrometallurgy operations that utilise Th- and U-bearing heavy mineral sands must consider their environmental effects and mitigate radioactivity. In addition, preferential acid dissolution of the Th–REE–Ti aluminosilicate in RTCS and other titania slags may be used to recover Th and REE for dual environmental and economic benefits.

Ferroinnelite, Ba4Ti2Na(NaFe2+)Ti(Si2O7)2[(SO4)(PO4)]O2[O(OH)], is a new mineral from the Phlogopite deposit, Kovdor alkaline massif, Kola Peninsula, Russia. In an agpaitic pegmatite, ferroinnelite occurs as transparent elongated platy to tabular crystals up to 0.15 mm long. Associated minerals are cancrinite, orthoclase, aegirine–augite, magnesio-arfvedsonite, golyshevite and fluorapatite. The mineral is yellow to yellow brown with a vitreous lustre and a white streak, Dcalc. is 4.088 g/cm3. Ferroinnelite is triclinic, space group P$\bar 1$, a = 5.3994(8), b = 7.09239(13), c = 14.7345(4) Å, α = 98.4086(19), β = 94.3275(18), γ = 90.0133(13)°, V = 556.56(8) Å3. The chemical composition of ferroinnelite is SO3 5.47, Nb2O5 0.45, P2O5 4.59, ZrO2 0.13, TiO2 16.91, SiO2 17.55, Al2O3 0.06, BaO 42.83, SrO 1.01, FeO 3.34, MnO 0.97, CaO 0.09, MgO 0.64, K2O 0.01, Na2O 4.47, H2O 1.11, F 0.15, O = F –0.06, total 99.72 wt.%, with H2O calculated from structure refinement. The empirical formula calculated on 26 (O + F) apfu is (Na1.95Fe2+0.64Mg0.21Mn2+0.19Ca0.02)Σ3.01(Ba3.84Sr0.13Na0.03)Σ4.00(Ti2.91Nb0.05Al0.02Zr0.01Mg0.01)Σ3.00Si4.02S0.94P0.89H1.70O25.89F0.11, Z = 1. The crystal structure was refined to an R1 index of 7.45% from 4485 unique reflections (Fo > 4σF). The structure is a combination of a TS (Titanium Silicate) and an I (intermediate) blocks. The TS block consists of HOH sheets (H – heteropolyhedral and O – octahedral). The O sheet is composed of Ti-, Na- and (Na,Fe2+)-octahedra, the H sheet, of [5]Ti-polyhedra and Si2O7 groups. The I block contains T sites, statistically occupied by S and P, and Ba atoms. Ideal compositions of the TS and I blocks are {Ti2Na(NaFe2+)Ti(Si2O7)2O2[O(OH)]}3– and {Ba4[(SO4)(PO4)]}3+. Ferroinnelite is a new member of the lamprophyllite group of the seidozerite supergroup. It is isostructural with innelite-1A, Ba4Ti2Na(NaMn2+)Ti(Si2O7)2[(SO4)(PO4)]O2[O(OH)]. Ferroinnelite and innelite are related by the following cation substitution at the MO3 site in the O sheet: Fe2+fer ↔ Mn2+inn. IR and Raman spectroscopies confirm presence of OH and H2O groups in ferroinnelite and innelite.

Electrum dendrites and associated minerals from the Ametistovoe Au–Ag epithermal deposit (Kamchatka) were investigated by optical and electron microscopy, X-ray computed tomography, electron microprobe and electron back-scattered diffraction techniques. Electrum forms dendrites up to 1 mm in size within sphalerite, or, rarely, within galena and chalcopyrite and random particles within quartz. The electrum has a very homogeneous composition (fineness 603 to 615) which correlates neither with the host mineral nor with the electrum morphology. X-ray computed microtomography showed electrum dendrites are 2-dimensional and consist of an acicular ‘trunk’ hosting numerous ‘branches’. 3-dimensional and ‘bush-like’ dendrites are rare. Angles between ‘trunk’ and ‘branches’ are usually 90° and less commonly, 60° suggesting twinning along (100) and (111) planes, respectively. Electron back-scattered diffraction showed the dendrites are single crystals with no defects detected, whereas electrum within quartz has a mosaic texture with low pattern quality. A model proposed for the observed phenomena includes: (1) growth of single-crystal electrum dendrites within silica gel; (2) overgrowth of dendrites by sphalerite crystals; (3) recrystallisation of the opaline silica to crystalline quartz which resulted in destruction of electrum dendrites located within silica and preservation of those hosted by sphalerite. This model proposes crystallisation of electrum dendrites from a hydrothermal solution and implies a critical role for amorphous silica gel as a precursor for dendrite formation.

Minerals incorporate 72 different essential elements, many of which are redox sensitive. We compiled oxidation states of ions in 4834 IMA-approved mineral species with oxygen and/or halogens as anions and have identified 87 essential mineral-forming ions. We compiled data on the coexistence of these ions as recorded in their minerals’ chemical formulas, and applied methods of network analysis with community detection and heatmap analysis with agglomerative clustering to reveal patterns of ion coexistence.

Unipartite networks illustrate the most common coexisting ion pairs, whereas Louvain and Walktrap methods reveal distinct ion groups—patterns that both reinforce and refine the Goldschmidt geochemical classification of elements. Key findings include: (1) that mineral-forming ions group into two major communities with a number of subcommunities; (2) that different ion communities primarily reflect contrasting geochemical and paragenetic processes such as primary igneous mineralisation, hydrothermal precipitation, and near-surface oxidation and weathering, rather than crystal chemical constraints; and (3) that different oxidation states of some redox-sensitive elements fall into two or more of these communities, underscoring how ions of the same elements commonly display contrasting geochemical and/or paragenetic affinities.

Heatmap analysis reveals groupings of co-occurring ions that mimic many aspects of community detection methods, as well as significant patterns of ion antipathies—groups of ions that are seldom if ever paired. For example, alkali metals commonly associated with late-stage igneous fluids (Cs+, Li+ and Rb+) rarely co-occur with low field strength ions found concentrated in brines (Ag+, Br–, Cu+, Hg+ and I–) or high field strength ions from weathered primary oxide or sulfide deposits (Cr6+, Pb4+, Mo4+, Te4+ and Te6+). Such ion pairs are well known in synthetic oxides. Therefore, with the exceptions of cations having very different redox potentials, unobserved ion pairs are principally the consequences of element rarity coupled with natural geochemical and paragenetic antipathies rather than crystal chemical constraints.