Abstract The origin of the compositional dichotomy between high‐Ti and low‐Ti basalts in continental flood basalt provinces remains debated. Most existing studies attribute this dichotomy directly to variations in primitive magma composition to circumvent the complexities of significant evolutionary processes, such as multiphase fractional crystallization. Here, we evaluate komatiite as a viable parental magma for the late Permian Emeishan Large Igneous Province to constrain its magmatic evolution. Olivine phenocrysts and spinel inclusions in the investigated picrites exhibit close affinities, consistent with derivation from a common komatiitic parental magma. Thermobarometric constraints define an olivine‐spinel‐clinopyroxene crystallization sequence and reveal two pressure regimes: 0–4.4 and 5.7–8.5 kbar. Thermodynamic modeling successfully reproduces the observed mineral assemblages and liquid lines of descent, accounting for two distinct adiabatic and multi‐stage evolutionary paths. We propose that extensive crustal differentiation along the liquid line of descent produced high‐Ti basalts, whereas limited evolution, melt‐cumulate re‐equilibrium, or incomplete crystal‐melt segregation generated low‐Ti basalts. These results demonstrate that contrasting crustal evolutionary pathways acting on a common komatiitic magma govern compositional diversities observed in global continental flood basalt provinces.

The formation of peralkaline silicic magmas and their crustal magmatic processes responsible for the eruptions remain poorly constrained. Here, we report the petrological and geochemical analyses of five-stage pumice samples (C1-C5) from the Tianwenfeng Eruption (TWF; C1-C2) and Millennium Eruption (ME; C3-C4-C5) of the Tianchi volcano, part of the Changbaishan Volcanic Field (CVF) in NE China. We reveal the assembly processes of the pantelleritic and comenditic magmas and the complexity of the magma plumbing system. The magma reservoir of comendite that controlled the formation of the Tianchi volcanic caldera is located at a depth of 3.5-7.5 km in the shallow crust. This reservoir exhibits vertical thermal and compositional zonation, consisting of an upper peralkaline rhyolitic magma body and a lower comenditic crystal mush. The widespread sieve textures observed in alkali feldspar macrocrysts indicate that all the magmas prior to the final eruption underwent significant recharge by deeper magmas, leading to the remobilization of the crystal mush. Crystals from various levels within the mush were entrained into the eruptible magma body, accompanied by magma mixing. These processes collectively contributed to the petrological and mineralogical diversity observed in the five-stage pumices. Furthermore, both the comendites and pantellerites originated from the continuous fractional crystallization of trachytic magma in an open system under low-pressure, reducing conditions, with the crystallization of alkali feldspar macrocrysts playing a critical role in magma evolution. The mature and active magma plumbing system beneath the Tianchi volcano still poses an eruption hazard under frequent geological activity. Future research should focus on magmatic processes at depths of 4-8 km beneath the volcano.

Oceanic large igneous provinces (LIPs), such as Shatsky Rise in the western Pacific, commonly exhibit variable and locally elevated abundances of chalcophile and siderophile elements (CSEs), reflecting the conditions of mantle melting and subsequent magma differentiation. However, the processes and conditions controlling these enrichments remain poorly understood. To unravel these processes, we analyzed major, volatile, and trace elements, including the strongly chalcophile and siderophile trace and ultra-trace elements Cu, Ag, Se, Re, Pt, Au, and Bi, in 48 glass samples from Shatsky Rise and the adjacent Ojin Rise seamounts. We also determined Fe3+/ΣFe ratios in 19 glasses and measured major element and sulfur contents in plagioclase-hosted melt inclusions from the Tamu Massif, the oldest in the Shatsky Rise.Parental Shatsky Rise magmas are strongly enriched in CSEs relative to MORB. They are inferred to have formed at fO2 of FMQ to FMQ − 1 and to have evolved by fractional crystallization, resulting in auto-oxidation to as high as FMQ + 1 in the most evolved melts. The CSE patterns of the normal-type Shatsky Rise magmas are consistent with extensive (22–25%) mantle plume melting that exhausted sulfide from the mantle residue. In contrast, parental Ojin magmas resemble MORB and became sulfide-saturated during the early stages of fractional crystallization. They were likely generated by more moderate degrees of mantle melting, leaving a sulfide-bearing residue at fO2 below FMQ.All Shatsky Rise melts remained sulfide-undersaturated during fractional crystallization down to ~7 wt% MgO. At lower MgO, the glasses define two contrasting evolutionary trends: sulfur-rich, CSE-depleted, sulfide-saturated compositions from the Ori Massif and Ojin Rise seamounts, and sulfur-poor, CSE-enriched, slightly more oxidized, sulfide-undersaturated compositions from the Tamu Massif. Ori Massif magmas record little or no sulfur degassing during crystallization and eruption and remain to be sulfide-saturated. In contrast, Tamu magmas experienced substantial pre-eruptive sulfur loss within an open-vent magmatic system, where prolonged recycling and mixing of degassed and undegassed magmas progressively depleted the magma in sulfur.These results highlight the fundamental role of shallow magmatic processes in controlling CSEs behavior in oceanic plateau magmas and determine their potential for the formation of magmatic sulfide deposits. Sulfide undersaturation during shallow evolution diminishes the ore-forming potential of LIP magmas, whereas late-stage sulfide saturation favors formation of Au–Cu–PGE-rich but Ni-poor magmatic sulfides.

The genesis of high-grade Fe-Ti-V oxide ores (up to >90 vol%) in layered intrusions remains highly debated. Here, on example of Hongge layered intrusion in China, we show that hydrothermal dissolution and precipitation of Fe-Ti-V oxides played a critical role in forming high-grade massive ore deposits as demonstrated by textural-compositional evidence and in-situ iron isotope data (delta Fe-56), analyzed with femtosecond laser ablation multicollector (LA-MC-) ICP-MS. Hongge is a mafic layered intrusion composed of a Lower olivine clinopyroxenite Zone (LZ), a Middle clinopyroxenite Zone (MZ), where thick massive ore layers (with up to 90 % Fe-Ti-V oxides) formed, and an Upper gabbro Zone (UZ). Magnetite in Hongge exhibits two contrasting generations: 1) Mag1, observed in all lithological zones and formed at the magmatic stage, has extensive ilmenite exsolution lamellae and high Ti and Cr content. The delta Fe-56 of Mag1 shows considerable variations from -0.23 to 0.63 parts per thousand and strikingly an offset of similar to 0.3 parts per thousand towards lower values in the massive ore zone compared to the zones below and above; 2) Mag2, concentrated mainly in thick massive ore layers in MZ without exsolution lamellae, is almost pure magnetite (with low Ti, Al content) and has extremely low delta Fe-56 values (-1.24 to -0.09 parts per thousand), indicating precipitation from Fe-enriched hydrothermal fluids. Similarly, the delta Fe-56 of ilmenite shows significant variations from -1.08 to -0.27 parts per thousand and is significantly lower than typical values for igneous ilmenite (-0.4-0 parts per thousand). Ilmenite displays a similar Fe isotope variation pattern to Mag1 along the stratigraphic position, i.e., with significantly lower delta Fe-56 in the massive ore zone. As magnetite and ilmenite together contain essentially all Fe, the isotopic shift of these minerals in the ore zone translates to a bulk isotopic offset of similar to-0.3 parts per thousand compared to the zones below and above. This requires a bulk flux of isotopically light Fe resulting in Fe enrichment in this zone to form massive or even monomineralic ores. The very light isotopic values, particularly hydrothermal magnetite (Mag2) and petrologic evidence, strongly indicate that the Fe flux into the massive ore layers occurred during hydrothermal reworking. This scenario is furthermore supported by magnetite-ilmenite elemental and isotopic thermometry, according to which Fe-Ti oxides experienced hydrothermal re-equilibration in a temperature range of 400-300 degrees C. Iron isotopic mass balance calculations imply that similar to 20-30 % of the Fe in the thick massive ore layers may result from secondary enrichment through hydrothermal precipitation, significantly increasing the ore tonnages and grades. Potentially, other layered intrusions experienced similar mechanisms of hydrothermal Fe enrichment, which will have to be proven in future investigations.

We introduce three new synthetic basalt reference materials and a new high‐precision set‐up for stable carbon isotope measurement in basaltic glasses using a large‐geometry secondary ion mass spectrometry (SIMS) instrument. The new reference materials, characterised for carbon mass fraction and isotope composition, show homogeneity for in situ analysis for the reported set‐up. Their bulk hydrogen mass fraction and isotope ratios are reported. Our SIMS protocol uses multi‐collection, cycling between concurrent measurements of 12 C and 13 C on electron multipliers, and either 30 Si or 18 O, as a reference mass, on a 10 11 Ω resistor Faraday cup. This set‐up achieves high measurement repeatability for δ 13 C down to ± 0.35‰ 1RSE at 1706 +89 / ‐88 μg g ‐1 CO 2 , with ± 1.00‰ 1RSE or better between 163 +5.1 / ‐5.2 and 267 +8.9 / ‐8.9 μg g ‐1 CO 2 , using a 10 nA primary beam current and a 40 μm analytical pit over a 100 cycle analysis. Carbon blanks were characterised by measuring carbon‐free olivines, allowing for blank corrections on δ 13 C measurements. After blank and instrument mass fractionation corrections, we measure δ 13 C in glasses down to 26.16 +0.85 / ‐0.86 μg g ‐1 CO 2 with a final measurement standard sample deviation of ± 2.97‰ 1 s . We report in situ measurements on an ocean floor basaltic glass and a set of synthetic basaltic glasses to demonstrate our approach. Reference materials and the SIMS set‐up improve the accuracy and precision of δ 13 C measurements in natural basaltic glasses across a wide range of geologically relevant carbon contents.

The origin of intraplate basalts remains controversial, despite numerous isotope and seismic studies. Here we employ thermodynamic modeling with MAGEMin to integrate the Sr-Nd-Hf isotopes of basaltic volcanics with the seismic velocities of their mantle source. The results indicate that residual garnets generate high P-wave velocity anomalies in the mantle, and impart distinctive time-integrated εNd-εHf isotope signatures to the partial melts (i.e., the erupted basaltic magmas). Since the garnet-spinel transition (~42-60 km) is always shallower than the lithosphere-asthenosphere boundary (~70-100 km), garnet effects may serve as an indicator of thermal magmatic events. These findings suggest that the widely distributed Cenozoic intraplate basalts in Eastern China originated from the lithospheric mantle (<~60 km) that didn’t retain garnet. Two notable exceptions are the Wudalianchi and Nuominhe basalts, which originate from the asthenosphere (~100-140 km) and the lithosphere-asthenosphere transition, respectively. The integrated radiogenic isotope-seismic velocity framework can be applied to intraplate basalts globally. Most Eastern China basalts originate or re-equilibrate with the lithospheric mantle, except for Wudalianchi and Nuominhe basalts, which originate from asthenospheric or lithosphere-asthenosphere transition sources, according to analysis of thermodynamic modeling, radiogenic isotopes, and seismic tomography.

Beryl can be coloured by Fe. The presence of Fe can be determined by chemical analysis, but also by UV-Vis-NIR spectroscopy. The UV-Vis-NIR spectra of untreated blue to greenish blue beryl and Fe-containing emerald were analysed with curve-fitting techniques. For spectra showing Fe2+ absorption in the infrared, as well as Fe2+-Fe3+ intervalence charge transfer absorption and Fe3+ absorptions near the ultraviolet, the background-corrected integrated areas under Fe-related absorption bands were plotted against the Fe contents determined with electron probe microanalysis (EPMA). Correlations between the Fe contents and three sets of Gaussian curves were evaluated: (1) the sum of the areas under the Gaussian curves fitted to all the Fe-related absorption bands (excluding the oxygen-to-metal charge-transfer signal), (2) the sum of the areas under the Gaussian curves fitted to the broad Fe-related absorption in the near-infrared and (3) the area of the fitted Gaussian curve centred at 13761 cm-1. An evaluation of seven test samples of blue to greenish blue beryl showed that for six of them the Fe content could be predicted within a 10% range of the Fe measured with EPMA, assuming a linear correlation between the Fe contents and the resulting Gaussian curve areas. The linear correlation using the area under the Gaussian curve centred at 13761 cm-1 and the corresponding equation Fe (mol%) = (Gaussian curve area + 28808) showed 83643 the most accurate results, with the lowest overall deviation from the EPMA Fe value. This evaluation model is a practical approach which makes possible the calculation/prediction of the Fe content in a beryl sample of known sample thickness by UV-Vis-NIR spectral analysis, and thereby allows a basic chemical characterisation of beryl samples using minimal analytical equipment.

Partial melts subducting oceanic crust (slab melts) have been proposed to play an important role in ore-forming processes and in the generation of continental crust, but the scarcity of erupted slab melts hampers direct evaluation of their role. Primitive Mg-rhyodacites with molar Mg/Mg+Fe-T >0.60 from the Western Aleutians represent nearly unmodified melts from eclogitized mid-ocean-ridge basalt (MORB) of the subducting Pacific plate. Here, we show that the presence of anhydrite phenocrysts and hornblende compositions indicate highly oxidizing conditions (Delta NNO >=+1) of crystallization. Despite these conditions and the absence of sulfides in the mineral assemblage, the rocks exhibit strongly fractionated chalcophile element ratios (Ag/Cu similar to 10 x MORB and Bi/Cu similar to 100 x MORB), suggesting their geochemical signatures were imposed by the presence of monosulfide solid solution (MSS) phase during slab melting. The chalcophile element abundances in the Mg-rhyodacites can be quantitatively explained by partial melting of MORB eclogite at 2 GPa and 900 degrees C in the presence of MSS. With the exception of Ag and Bi, slab melts have low contents of strongly chalcophile elements, such as Cu, Se, Pt, and Au, and therefore a limited capacity to transport these elements under upper mantle conditions. The distinctively high Ag/Cu and Bi/Cu ratios in the Mg-rhyodacites further suggest that slab-derived magmas may have been significant contributors to the chalcophile element signatures of the continental crust throughout Earth's history.

High-Ti basalts are commonly believed to represent parental magmas leading to the formation of mafic-ultramafic layered intrusions, such as Late Permian Panzhihua and Hongge in the Emeishan Large Igneous Province (SW China). Consequently, elucidation of the crystallisation and crustal differentiation of high-Ti basalts is critical for our understanding of the petrogenesis of these layered intrusions and the associated oxide ore mineralisation. Here, we present the results of crystallisation experiments carried out in internally heated pressure vessels using a primitive high-Ti basaltic composition. The experiments were conducted at 100 and 300 MPa, in the temperature interval of 950-1200 degrees C and with water activities (aH(2)O) from 0 to 1. The oxygen fugacity (fO(2)) was controlled and varied from FMQ -1 to FMQ +3.3 log units (FMQ corresponds to fayalite-quartz-magnetite buffer). The main mineral phases are olivine, clinopyroxene and plagioclase, accompanied by Cr-Fe-Ti-oxides, orthopyroxene, apatite and amphibole, depending on the conditions. Redox conditions primarily influence the stability fields of Cr-Fe-Ti oxides. Clinopyroxene, orthopyroxene and amphibole are pressure-dependent and have larger stability fields under high pressure conditions. The olivine -> orthopyroxene and olivine -> amphibole peritectic reactions are observed. Comparisons of phase equilibria between this study and experiments conducted with parental magma of Skaergaard layered intrusion demonstrate the effect of bulk system composition. For instance, ilmenite crystallisation is determined not only by intrinsic parameters such as fO(2) but also by additional compositional parameters (e.g. melt Ti, Fe, Al and Mg content). Although COMAGMAT and MELTS modelling results generally reproduce the crystallisation sequence, only the stability field of clinopyroxene and its composition are perfectly modelled. The comparison of experimental results with the rocks from the lowest units of Panzhihua and Hongge layered intrusions are applied to constrain storage conditions in the magma reservoirs. Compared to Hongge, we conclude that the Panzhihua magma chamber was probably located at a shallower depth (similar to 3-6 km), that magma crystallisation started at lower temperatures (similar to 1125-1100 degrees C), higher fO(2) (similar to FMQ + 1 to FMQ + 2) and that its initial melt H2O content was lower (similar to 0.5-1 wt. %).

Carbonatites and their comagmatic silicate rocks related deposit provide significant resources of rare earth elements (REEs), niobium (Nb) and other elements such as U, Th, Mo, V, Ba, Sr, etc. However, the genesis of mineralization, especially for REEs and Nb, in carbonatite remains enigmatic. Previous liquid immiscibility experiments have demonstrated that both REEs and Nb are preferentially enriched in the silicate conjugate instead of carbonate melts under anhydrous conditions. Nevertheless, ligands other than carbonate ion appear to be abundant due to ubiquity of apatite, baryte, celestine, fluorite and sodalite in carbonate-silicate magmatic systems. Here, we experimentally investigate the trace element partitioning between natrocarbonate and silicate (nephelinite) melts in systems doped with varying amounts of additional F-, PO43-, Cl-, and SO42- ligands (0, 2, 4 and 6 wt%) to understand and constrain the role of ligands. The experiments were conducted at 850 degrees C and 0.1 GPa using rapid quench cold-seal pressure vessels (CSPVs). A comparison of experimental partition coefficients in this study reveals that the significant amounts F- and PO43- incorporated in the silicate melts can increase the D values for REE by influencing melt structure (D-La(CM/SM) = 0.85-7.42). In contrast, irrespective of the amount of added Cl- and SO42-, D-CM/SM is not affected significantly by these species and the D-REE(CM/SM) values remain always lower than 1 (D-La(CM/SM) = 0.12-0.40). Notably, the D-Nb(CM/SM) values are all <1, with only one exception containing 6 wt% F. Besides, in all the investigated systems, Ba, Sr, Mo, V, Cs, Rb and Li preferentially partition into the conjugate carbonate melt. All the high field strength elements (Pb, Th, U, Zr, Hf, Nb, Ta), transition metals (Mn, Co, Cu, Zn) and common network formers (Ga, Ge) essentially partition into the silicate melt.

Here we present a confocal Fe K-edge μ-XANES method (where XANES stands for X-ray absorption near-edge spectroscopy) for the analysis of Fe oxidation state in heterogeneous and one-side-polished samples. The new technique allows for an analysis of small volumes with high spatial 3D resolution of <100 µm3. The probed volume is restricted to that just beneath the surface of the exposed object. This protocol avoids contamination of the signal by the host material and minimizes self-absorption effects. This technique has been tested on a set of experimental glasses with a wide range of Fe3+ / ΣFe ratios. The method was applied to the analysis of natural melt inclusions trapped in forsteritic to fayalitic olivine crystals of the Hekla volcano, Iceland. Our measurements reveal changes in Fe3+ / ΣFe from 0.17 in basaltic up to 0.45 in dacitic melts, whereas the magnetite–ilmenite equilibrium shows redox conditions with Fe3+ / ΣFe ≤0.20 (close to FMQ, fayalite–magnetite–quartz redox equilibrium) along the entire range of Hekla melt compositions. This discrepancy indicates that the oxidized nature of glasses in the melt inclusions could be related to the post-entrapment process of diffusive hydrogen loss from inclusions and associated oxidation of Fe in the melt. The Fe3+ / ΣFe ratio in silicic melts is particularly susceptible to this process due to their low FeO content, and it should be critically evaluated before petrological interpretation.

Platinum and Pd-arsenides, antimonides, tellurides, bismuthinides, and sulfides are the major hosts of Pt and Pd in magmatic and hydrothermal Cu-Ni-sulfide ores. Textural relationships among such minerals in nature often provide contradictory messages about the mechanism and timing of their formation. To know how and when Pt and Pd mineral phases crystallize from sulfide-saturated liquids and how their compositions and textures evolve during cooling, we undertook controlled cooling experiments in evacuated silica tubes. A Cu-Ni-Fe sulfide mixture, similar in composition to the average Merensky Reef sulfide magma, was charged (in a 9:1 wt.% proportion) with one of the PtAs2, PtTe2, PtSb2, PtBi2, PdAs2, PdTe2, PdSb, and PdBi2 compounds, heated to 1,100°C, and slowly cooled (15°C/day) to room temperature. The run products were sampled at 950°C, 750°C, and 25°C by quenching the silica tubes in water. The results show that Pt formed stable phases with As, Te, and Sb above 950°C and with Bi and S below 750°C. The Pt phases were found mostly in the decomposed intermediate solid solution (ISS), and all the phases survived to 25°C with no compositional changes. At 950°C, Pd formed arsenide, telluride, and antimonide immiscible melts that coexisted with monosulfide solid solution (MSS) and sulfide melt. The composition of the Pd-semimetal melt droplets changed during cooling by equilibration with the host sulfide. Palladium mineral phases formed at temperatures below 750°C directly from the immiscible semimetal-rich melts where they kept the rounded shapes of the melt droplets. Ni-sulfarsenides preceded the formation of Pd arsenides. No Ni tellurides, antimonides, or bismuthinides formed in any of the systems, implying that their formation requires higher semimetal/Pt + Pd ratios. The decomposition of MSS and ISS to base metal sulfides led to significant textural changes in Pt and Pd mineral grains. The compositions of Pt and Pd phases are inherited from the magmatic stage, and their textures are low-temperature (<450°C) features.

We study three ultrabasic-alkaline carbonatite complexes (UACCs: Afrikanda, Vuorijarvi, Kovdor) from the Kola Alkaline Province to focus on the comparison of REE contents from pyroxenites and carbonatites and the fate of perovskite. Pyroxenites and carbonatites have large REE variations and similar contents in all three complexes. Afrikanda and Vuorijarvi pyroxenites and carbonatites have higher REE contents compared to all Kovdor rocks. Despite their comparable LREE contents, pyroxenites and carbonatites have different REE-carrier minerals. In most studied pyroxenites, perovskite is the main REE-bearing mineral and its abundance controls the LREE enrichment of the whole rock. When perovskite is absent or in low abundance, apatite is the main REE carrier. Instead, in carbonatites and phoscorites, apatite and calcite control REE contents. REE-carbonates - found in several carbonatite samples - do not play a substantial role in the overall REE budget because of their very low abundance and tiny grain sizes. We found large LREE variations at very low local scales (< 1 mm) in several thin sections. These large variations are related to carbonatite infiltration and associated antiskarn reactions. Perovskite near calcite is replaced by titanite that has much lower LREE contents. Most LREE released from the breakdown of perovskite were probably dissolved in the carbonatite melt because apatite and calcite that crystallized from this melt are highly enriched in LREE. Newly formed apatites and calcites show large LREE variations that probably are controlled by local factors (e.g. variations of Si, Ca, Mg activities). Y/Ho ratios are strongly fractionated in most minerals and often show large variations at local scales (mm) in these antiskarn reactions. The potential of REE remobilization from such perovskite replacement reactions is very high as perovskite contributes about 70% to the REE budget for these three UACCs. Therefore, more attention should be given to perovskite replacement reactions and its role for LREE enrichment of carbonatite melts.

To improve our understanding of the formation of sedimentary copper deposits, the reaction of cuprite with 0.2 m HAc-KAc or pure H2O solutions is studied systematically at 100-250 degrees C and 5-30 MPa. The experiments were carried out for periods of up to 72 h in a Parr autoclave, allowing for the in situ sampling of the fluid phase. The experiments conducted in this study demonstrate that cuprite (Cu2O) underwent a series of changes: (i) simple dissolution, (ii) Cu(I) disproportionation to native Cu and Cu(II), and (iii) subsequent oxidation into tenorite (CuO). In pure water, only (i) and (ii) steps can be discerned, whereas all three processes have been observed in an acetate-bearing system. In HAc-KAc solutions, the maximum dissolved Cu content correlates inversely with temperature, i.e., 378 to 168 mu g/g at 100 and 200 degrees C, respectively. However, equilibrium has not been reached in our experiments and these values may be treated as minimum cuprite solubility. In situ Cu isotope analyses have been carried out by laser ablation combined with a multicollector inductively coupled plasma-mass spectrometer. The data imply that copper isotope fractionation during cuprite replacement reactions is small. Both the microscopic observations on cross sections and the analytical data support the idea that the mineral replacement reaction is controlled by a coupled dissolution-reprecipitation (CDR) mechanism. This applies to both the deposition of metallic copper and the formation of tenorite. As suggested by the formation of pore spaces in the deposited layers, only a portion of the dissolved copper is redeposited directly in situ. The isotopic analyses of the solution and solid phases show that the partial transfer of copper into the surrounding solution is not associated with a significant isotopic effect, e.g., a measured difference between Cu and Cu2O is within 0.32 +/- 0.06 parts per thousand. Our study indicates that acetate plays a dual role in copper transport and deposition. On one hand, the presence of acetate strongly enhances the Cu content in solution up to 400 mu g/g, implying that acetate complexation can be responsible for metal transport in hydrothermal fluids. On the other hand, decarboxylation of acetate substantially decreases the dissolved Cu and aids the precipitation of tenorite. This may lead to the co-occurrence of Cu-bearing minerals with different oxidation valence states at low temperatures in a variety of geological settings such as supergene hydrothermal systems.

The causes of colour in beryl have been a research topic for decades. For some varieties, such as emerald (green, coloured by Cr3+ and/or V3+), the main cause of colour is substitutions by metal atoms within the framework. However, the causes for the yellow and blue colours in heliodor, golden beryl and aquamarine are still debated. It is generally agreed that Fe ions are responsible for the colour, but there are differing conclusions about the valence states of these ions, the occupied positions and the colour-inducing processes involved. The colour of aquamarine is commonly attributed to intervalence charge transfer (IVCT) between Fe3+ and Fe2+. Various combinations of sites have been proposed to host the Fe ions engaging in this IVCT. Here we present a new approach to address the topic of colour generation: atomic resolution scanning transmission electron microscopy (STEM). For the first time, atomic resolution images of a beryl (natural aquamarine) are presented in the three crystallographic directions [0001], [1-210] and [1-100]. Ions are clearly resolved in the channels. From the ratio of channel occupation and the correlation of the atoms per formula unit (apfu) calculations we conclude that Fe resides in the framework, not in the channels. The projections in the [1-210] direction directly show that the cavity channel site 2a is occupied, most likely by Cs, in agreement with recent results in the literature.

Lascar volcano, located in northern Chile, is among the most active volcanoes of the Andean Central Volcanic Zone (CVZ). Its activity culminated in the last major explosive eruption in April 1993. Lascar andesites which erupted in April 1993 have a phase assemblage composed of plagioclase, clinopyroxene, orthopyroxene, Fe–Ti oxides, and rhyolitic glass. To better constrain storage conditions and mechanisms of magmatic differentiation for andesitic magmas in a thick continental crust, crystallization experiments were performed in internally heated pressure vessels at 300 and 500 MPa, in the temperature (T) range of 900–1050 °C, at various water activities (aH2O) and oxygen fugacities (logfO2 between QFM+1.5 and QFM+3.3 at aH2O =1; QFM is quartz–fayalite–magnetite oxygen buffer). The comparison of experimental products with natural phase assemblages, phase compositions, and whole-rock compositions was used to estimate magma storage conditions and to reconstruct the magma plumbing system. We estimate that Lascar two-pyroxene andesitic magmas were stored at 975±25 °C, 300±50 MPa, and logfO2 of QFM+1.5±0.5, under H2O-undersaturated conditions with 2.5 wt % to 4.5 wt % H2O in the melt. The geochemical characteristics of the entire suite of Lascar volcanics indicates that a fractionating magmatic system located at a depth of 10–13 km is periodically replenished with less evolved magma. Some eruptive stages were dominated by volcanic products resulting most probably from the mixing of a mafic andesitic magma with a felsic component, whereas compositional variations in other eruptive stages are better explained by crystal fractionation processes. The relative importance of these two mechanisms (mixing vs. crystal fractionation) may be related to the amount and frequency of magma recharge in the reservoir.

Incorporation of ions into the crystal structure of beryl (Be3Al2[Si6O18]) can take place by direct ion-to-ion substitution of the framework components Al3+, Be2+ and Si4+ or by occupation of interstitial or structural channel sites. The most common impurities in beryl include transition metals, alkalis and H2O. It is accepted that the transition metals Mn, Cr and V directly substitute for Al at the octahedral site and induce colour. Similarly, the octahedral site can host Fe instead of Al. Nevertheless, it is shown that it remains disputed whether Fe can also be present at the tetrahedral, interstitial, or channel sites, and opposing hypotheses exist regarding these possibilities. However, in the case of Fe, not only the possible occupation of these sites remains under debate, but also their influence on the subsequent colour of beryl. Similarly, the residence of Li in the channels and at the Be tetrahedral or interstitial tetrahedral sites is still under debate. The presence of more than two types of H2O (type I and type II) in the structural channels of beryl is also unclear. This article aims to give an overview on the consensus and on the current debates found in the literature regarding these aspects. It mainly concentrates on the substitution by and the role of Fe ions and on channel occupancy by H2O.

Magma mixing is a widespread magmagenic process. However, its significance in the formation of ultrapotassic magmas has been largely overlooked so far as they are commonly thought to originate directly from the mantle and ascend rapidly through the crust. The Hezhong ultrapotassic lavas in Western Yunnan (SW China) are (basaltic) trachy-andesitic in composition. These rocks display porphyritic textures with olivine, clinopyroxene (Cpx), spinel, and phlogopite occurring as both phenocryst and glomerocryst. Disequilibrium textures and complex zonation of crystals are commonly observed. Specifically, based on the textural and compositional characteristics, olivines can be classified into three different populations: two populations are characterized by highly to moderately magnesian olivines with normal chemical core-rim zonation (Fo ~94–86 to Fo ~89–79 and Fo ~91–89 to Fo ~86–84 , respectively). The third population lacks obvious crystal zonation, but individual crystals exhibit some compositional variety at lower Fo contents (Fo 83–76 ). Similarly, four populations of Cpx and two populations of spinel phenocrysts are recognized in terms of texture and composition. Notably, Cpx with reverse zoning contains a ‘green-core’ surrounded by a colourless mantle and rim. Hence, based on the variations of mineral assemblage, types of inclusions, and chemical compositions, all phenocryst/glomerocryst minerals can be divided into three groups. Mineral Group I (MG I) consists of high Fo cores of olivine, cores of the zoned spinel, and phlogopite. MG II only includes the green cores of reversed zoned Cpx (green-core Cpx), and MG III is composed of micro phenocrysts without obvious zoning and rims of large phenocrysts. Comparing these mineral groups with relevant minerals occurring in typical temporally and spatially associated igneous rocks, we suggest that the MG I and II could have been derived from magmas with compositions resembling an olivine lamproite and a trachyte, respectively. The overall bulk-rock geochemical and isotopic features of Hezhong lavas also agree with a mixing process between these two endmembers. Hence, we infer that mixing between these two magmas played a key role in the petrogenesis of the ultrapotassic Hezhong lavas and that the MG III crystallized from the mixed magmas. Our study highlights the complex formation of ultrapotassic magmas inferring that caution must be taken when using bulk chemical magma compositions are to deduce source signatures.

&lt;p&gt;Besides H&lt;sub&gt;2&lt;/sub&gt;O, CO&lt;sub&gt;2&lt;/sub&gt; and sulphur, halogens (F, Cl, Br, I) are important volatile components in magmas. The extremely high chemical activity of halogens in melts and liquids leads to a significant influence on (a) magmatic properties, (b) the degassing of magma, (c) the extraction, transport and deposition of metals, (d) the chemistry of volcanic emissions and (e) the composition of the atmosphere. Indeed, their geochemical behaviour can be used as a key indicator of the genetic conditions and evolution of magma.&lt;/p&gt;&lt;p&gt;In July 2021 a joint interdisciplinary campaign of petrologists, chemists and atmospheric physicists took place at Mt Etna volcano, Italy. Due to the favourable volcanic activity at Mt Etna (frequent paroxystic activity characterized by lava fountaining) we were able to collect &amp;#160;a unique dataset of simultaneously sampled fresh tephra fallout, in-situ gas samples (multiGAS, alkaline traps, 1,3,5-Trimethoxybenzene impregnated denuders) and spectral data with remote sensing techniques (DOAS, FTS, IFPICS) of the volcanic plume. The halogen and sulphur content was analysed in the volcanic plume as well as in the melt inclusion and glasses of the deposits. Results of the various applied techniques are presented. They allow us a direct comparison of degassing signatures (e.g., Cl/F, Br/Cl, and S/Cl) from the pre-eruptive melt to the volcanic plume.&lt;/p&gt;

Demantoid is a rare Cr-bearing variety of andradite garnet historically mined in the Middle Urals of Russia, where its mode and time of formation remain controversial. A first direct in situ U-Pb LA-ICP-MS dating of demantoid was carried out on samples with extremely low U content obtained from the Karkodinskoe and Poldnevskoe deposits, both of which are located in the Karkodin meta-ultramafic massif of the Middle Urals. In both locations, demantoid mineralization occurred within cracks in ca. 400 Ma serpentinite host rocks composed of chrysotile and antigorite, calcite and aragonite, and magnesite. Magnetite, chromite, and pyrite occur as accessory minerals. The results yielded ages of 343 +/- 4 Ma and 292 +/- 1 Ma for the Karkodinskoe and Pold-nevskoe deposits, respectively. These ages correspond to the times of continental collision and subsequent exhumation in the Middle Urals during the Uralian orogeny. Geochronology combined with trace element data of Uralian demantoid suggests that mineralization was induced by hydrothermal processes at T < 300 degrees C. But trace-element signatures point to different origins for the two hydrothermal fluids. Rare-earth element (REE) patterns in garnet from Karkodinskoe deposit appear to be inherited from syenitic rocks, whereas REEs in demantoid from Poldnevskoe show affinities with skarn environments. During the collision stage, magmatic fluids that were probably associated with syenitic rocks of the Kozlinogorsky complex were re-activated leading to the re-working of the meta-ultramafic protolith of the Karkodin massif. Metamorphic processes during the exhumation stage subsequently resulted in the mobilization of hydrothermal fluids likely linked to skarns. Their migration into fractures within meta-ultramafic rocks developed the demantoid mineralization in Poldnevskoe occurrence.