Volume 106, Issue 3

Information-rich attributes of minerals reveal their physical, chemical, and biological modes of origin in the context of planetary evolution, and thus they provide the basis for an evolutionary system of mineralogy. Part III of this system considers the formation of 43 different primary crystalline and amorphous phases in chondrules, which are diverse igneous droplets that formed in environments with high dust/gas ratios during an interval of planetesimal accretion and differentiation between 4566 and 4561 Ma. Chondrule mineralogy is complex, with several generations of initial droplet formation via various proposed heating mechanisms, followed in many instances by multiple episodes of reheating and partial melting. Primary chondrule mineralogy thus reflects a dynamic stage of mineral evolution, when the diversity and distribution of natural condensed solids expanded significantly.

Raman spectra were collected for an extensive set of well-characterized layer-structure Mn oxide mineral species (phyllomanganates) employing a range of data collection conditions. We show that the application of various laser wavelengths, such as 785, 633, and 532 nm, at low power levels (30–500 μW) in conjunction with the comprehensive database of standard spectra presented here, makes it possible to distinguish and identify the various phyllomanganate minerals. The Raman mode relative intensities can vary significantly as a function of crystal orientation relative to the incident laser light polarization direction as well as incident laser light wavelength. Consequently, phase identification success is enhanced when using a standards database that includes multiple spectra collected for different crystal orientations and with different laser light wavelengths. The position of the highest frequency Raman mode near 630–665 cm –1 shows a strong linear correlation with the fraction of Mn 3+ in the octahedral Mn sites. With the comprehensive Raman database of well-characterized Mn oxide standards provided here (and available online as Online Material Deposit item AM-21-37666, Online Material. Deposit items are free to all readers and found on the MSA website, via the specific issue’s Table of Contents (go to http://www.minsocam.org/MSA/AmMin/TOC/2021/Mar2021_data/Mar2021_data.html). ), and use of appropriate data collection conditions, micro-Raman is a powerful tool for identification and characterization of biotic and abiotic Mn oxide phases from diverse natural settings, including on other planets, as well as for laboratory and industrial materials.

Knowledge of the stability of carbonate minerals at high pressure is essential to better understand the carbon cycle deep inside the Earth. The evolution of Raman modes of carbonates with increasing pressure can straightforwardly illustrate lattice softening and stiffening. Here, we report Raman modes of natural magnesite MgCO 3 up to 75 GPa at room temperature using helium as a pressure-transmitting medium (PTM). Our Raman spectra of MgCO 3 show the splitting of T and ν 4 modes initiated at approximate 30 and 50 GPa, respectively, which may be associated with its lattice distortions. The MgCO 3 structure was referred to as MgCO 3 -Ib at 30–50 GPa and as MgCO 3 -Ic at 50–75 GPa. Intriguingly, at 75.4 GPa some new vibrational signatures appeared around 250–350 and ~800 cm –1 . The emergence of these Raman bands in MgCO 3 under relatively hydrostatic conditions is consistent with the onset pressure of structural transition to MgCO 3 -II revealed by theoretical predictions and high-pressure and high-temperature experiments. This study suggests that hydrostatic conditions may significantly affect the structural evolution of MgCO 3 with increasing pressure, which shall be considered for modeling the carbon cycle in the Earth’s lower mantle.

In the present study, the competitive adsorption geometries for arsenate and phosphate on magnetite surfaces over a pH range of 4–9 were investigated using in situ attenuated total reflectance Fourier-transform infrared spectroscopy (ATR-FTIR) and two-dimensional correlation analysis (2D-COS). The adsorption energies and infrared vibrational frequencies of these surface complexes were also calculated by first-principles simulations. Arsenate and phosphate have different preferences for the magnetite surface in the presence of aqueous solvent at both acid and alkaline pH. For the adsorption of phosphate, mono-protonated monodentate mononuclear (MMM) complexes dominated at acid pH, while non-protonated bidentate binuclear (NBB) complexes were dominant at alkaline pH. Arsenate mainly formed bidentate binuclear (BB) complexes with some outer-sphere species, both of which were more prevalent at acid pH. The pre-absorbed inner-sphere arsenate species were scarcely afected by the introduction of phosphate. However, the pre-absorbed phosphate oxyanions, especially the MMM complexes, were significantly substituted by BB arsenate at the magnetite surfaces. The adsorption afinity of phosphate and arsenate species for magnetite surface was found to increase in the following order: MMM phosphate complex < NBB phosphate complex < BB arsenate complex, which was consistent with the calculated adsorption energies. The simulated infrared vibrational frequencies for the most favorable adsorption modes of each oxyanion display distinctive patterns, and their trends are in excellent agreement with experimental data. The effects of pH, adsorption sequence, and mineral species on the competitive adsorption between arsenate and phosphate oxyanions are also discussed, and their different competing ability and stability on the magnetite surfaces can be ascribed to the variations in adsorption geometry and strength of binding. To the best of our knowledge, this is the first study aiming to distinguish the stability of the different phosphate and arsenate complexes on magnetite by employing a combined approach of in situ spectroscopy and DFT simulations. Our results provide molecular-level insight into the geometries and relative stabilities of the adsorption of phosphate and arsenate on magnetite surfaces, which is useful for interpretation of the mobility and bioavailability of these anions.

Detailed knowledge of the structural evolution of diverse aluminum (oxy)hydroxides with varying temperatures up to ~300 °C provides insights into the dehydration processes involving transitions among metastable phases on Earth’s surface and in the crust. Here, we report the high-resolution solid-state 27 Al NMR spectra for three different types of aluminum (oxy)hydroxides (i.e., boehmite, bayerite, and gibbsite) with varying annealing temperatures up to 300 °C, revealing the effect of distinct precursor minerals on the stability of metastable alumina. 27 Al MAS NMR results allow us to obtain the quantitative fractions and NMR parameters for each phase during transformation. The results demonstrate that each aluminum (oxy)hydroxide phase follows a unique transformation path to metastable alumina. The 27 Al MAS and 3QMAS NMR spectra of boehmite show that a minor but observable [4] Al signal (~2%) is detected at ~50 °C, and the [4] Al fraction gradually increases up to 300 °C (~16%), indicating that the phase transformation from boehmite to γ-/η-Al 2 O 3 occurs at a temperature as low as ~50 °C, significantly lower than earlier estimations based on XRD. Together with the [4] Al fraction, the [5] Al fraction increases from <1% at 50 °C to ~2.3% at 300 °C, whereas the NMR results of bayerite and gibbsite do not show the presence of [5] Al. In addition, the 27 Al 3QMAS NMR spectra resolved the [6] Al site in boehmite and that in γ-/η-Al 2 O 3 , which could not be uniquely determined from 1D NMR spectra. The population of bayerite abruptly decreases from 100% (at 150 °C), through ~47% (at 200 °C), to 0% (at 250 °C), indicating that the phase transition from bayerite to boehmite + γ-/η-Al 2 O 3 occurs within a narrow temperature range. As for gibbsite, while [4] Al is not observed in the spectra up to 200 °C, the [4] Al fraction of ~2% is observed in the spectra for gibbsite annealed at 250 °C, and the [4] Al fraction increases rapidly to ~15% as the annealing temperature increases to 300 °C, suggesting that the phase transformation into γ-/η-Al 2 O 3 occurs at ~250 °C. The results confirm that the phase transformation paths (gradual vs. dramatic) depend on the type of precursor minerals. Particularly, the onset temperature of the phase transformation from boehmite to metastable alumina (~50 °C) is lower than those from other precursor minerals (>150 °C). Furthermore, the phase transformation from boehmite to γ-/η-Al 2 O 3 occurs gradually within broad temperature ranges from ~50 °C. This is due to their configurational disorder as evidenced by the presence of [5] Al. The observed structural evolution in aluminum (oxy)hydroxides in the low-temperature range facilitates our understanding of the nature of phase transformation and dehydration of oxides and hydroxides in the Earth’s surface environments.

We determined for the first time the crystal structures of high-pressure K-cymrite (KAlSi 3 O 8 ⋅H 2 O) and its dehydrated form kokchetavite (KAlSi 3 O 8 ) using single-crystal X‑ray difraction. The K-cymrite structure has been successfully refined in the hexagonal space group P 6/ mmm [ a = 5.3361(3) Å, c = 7.7081(7) Å, V = 190.08(3) Å 3 , R 1 = 0.036 for 127 unique observed reflections], which is in agreement with previous models from powder X‑ray difraction. In contrast, kokchetavite shows superstructural reflections, suggesting a diferent unit cell and a space group of P 6/ mcc [ a = 10.5757(3) Å, c = 15.6404(6) Å, V = 1514.94(10) Å 3 , R 1 = 0.068 for 1455 unique observed reflections]. Upon dehydration, single-crystal grains of K-cymrite transform into single-crystal grains of kokchetavite. The latter questions a previous interpretation of kokchetavite crystals in mineral inclusions as a product of direct crystallization from fluid/melt. The Raman spectrum of K-cymrite shows a strong polarization dependence, which is important in identification of the mineral inclusions.

The source of fluids and their mechanism of metal precipitation in sediment-hosted, disseminated orogenic gold deposits are ambiguous. Pyrite texture, trace element, S, Pb, and He-Ar isotope compositions of sulfides and C-O isotope data of calcite from Chang’an orogenic gold deposit in the Ailaoshan orogenic belt, southwest (SW) China, were studied to provide a new genetic model for the sediment-hosted orogenic gold deposit, furthering knowledge of the source of fluids and their mechanism of metal precipitation. Orebodies at Chang’an are mainly hosted by Ordovician turbidite with a few in Oligocene syenite. Two stages of mineralization have been identified in the deposit: stage I disseminated quartz-arsenopyrite-pyrite and stage II veined quartz-calcite-polymetallic sulfides. Five generations of pyrite have been identified in turbidite: pre-ore syn-sedimentary pyrite, py I-1 and py I-2 in stage I, and py II-1 and py II-2 in stage II, and an unzoned pyrite population developed in syenite. Py I-1 commonly overgrows syn-sedimentary pyrite with irregular boundaries and contains arsenopyrite, galena, chalcopyrite, and electrum inclusions along the boundaries. Py I-1 is overgrown by thin and inclusion-free py I-2 , and crosscut by py II-1 , which is rimmed by py II-2 . The syn-sedimentary pyrite is distributed parallel to the sedimentary bedding and contains As (620.8 ppm), Pb (61.6 ppm), Ni (59.8 ppm), Mo (54.4 ppm), Co (23.4 ppm), and Cu (13.0 ppm) with low-Au content of 0.06 ppm. This pyrite has δ 34 S values of –18.1 to +30.4‰ and high-radiogenic Pb isotope ratios (average 206 Pb/ 204 Pb, 207 Pb/ 204 Pb, and 208 Pb/ 204 Pb of 19.05, 15.86, and 39.87, respectively). Py I-1 and coexisting arsenopyrite are enriched in invisible Au (up to 227.1 and 353.3 ppm, respectively), As, Ni, Cu, and Pb, while py I-2 contain much lower trace element abundances relative to py I-1 and arsenopyrite. Partial replacement of syn-sedimentary pyrite by py I-1 plus arsenopyrite, galena, chalcopyrite, and electrum, and similar Pb isotope ratios between syn-sedimentary pyrite and py I-1 indicate that reaction of external deep Au-rich fluids with syn-sedimentary pyrite is responsible for gold precipitation in stage I. Py I-1 , arsenopyrite, and py I-2 show a narrower δ 34 S range of –3.2 to 7.1‰ relative to syn-sedimentary pyrite, demonstrating that the fluid-pyrite interaction has homogenized the sulfur. The unzoned pyrite in syenite has similar mineral inclusions (arsenopyrite, galena, etc.), δ 34 S values (+0.6 to 6.3‰) and Pb isotope ratios to py I-1 , but much lower trace element abundances relative to py I-1 . It may be attributed to different reactions of similar fluids with different wall-rocks. Py II-1 and py II-2 in stage II contain elevated As, Pb, Cu, Sb, Zn, and Ag with low mean Au content (3.3 ppm) and have δ 34 S ranges of –2.8 to +1.2‰ and –6.2 and –0.8‰, respectively. Galena in stage II has lower radiogenic Pb isotope ratios than stage I pyrites, indicative of a diferent Pb source or fluid evolution. The gases released from a mixture of py II-1 -py II-2 have R/Ra of 0.38 to 0.98 and 40 Ar*/ 4 He of 0.50 to 1.34, falling between the fields of mantle-derived and crustal fluids. Late ore calcites have δ 13 C PDB of –8.7 to 2.7‰ and δ 18 O SMOW of 8.05 to 25.58‰, also plotting between sedimentary carbonate and mantle fields. These signatures indicate that ore fluids in stage II are base metal-rich fluids with a small amount of contribution from the mantle. Different ore assemblages, trace element composition and isotope data between stages I and II at Chang’an suggest that the deposit experienced an evolution from early Au-rich fluids to late base metal-rich ones. This study highlights that ore metals in sediment-hosted disseminated orogenic gold deposits may be sourced from both deep fluids and local wall-rock, and that fluid-rock interaction behaved as a key control on ore precipitation.

Iron isotope fractionation in hydrothermal systems is a useful diagnostic tool for tracing ore-forming processes. Here, we report on the Fe isotopic compositions of a suite of hydrothermal minerals from ores (pyrite, pyrrhotite, and quartz) from the Wulong gold deposit, Liaodong Peninsula, East China. Pyrites from quartz sulfide ores show a δ 56 Fe ( 56 Fe/ 54 Fe in the sample relative to IRMM-14) range from +0.11 ± 0.03‰ to +0.78 ± 0.03‰ (2SD), and pyrrhotites from the same vein are isotopically lighter than the pyrites, varying between –0.85 ± 0.01‰ and –0.07 ± 0.00‰. This result is consistent with theoretical predictions of equilibrium fractionation and published mineral compositions. For the first time, to our knowledge, we report the Fe isotopes of hydrothermal quartz that records the isotopic compositions of the ore-forming fluids. Two quartz separates in the quartz-sulfide vein yield δ 56 Fe values of –0.02 ± 0.02‰ and +0.07 ± 0.07‰. Our Fe isotope fractionation calculations show that pyrrhotite with light Fe isotopes crystallized first from the ore-forming fluids, which indicates a relatively reduced condition for the initial ore-forming fluids. Then the remaining fluids with heavy Fe isotopes precipitated pyrites with positive δ 56 Fe values, and their mineral crystallization sequence records an increase of oxygen fugacity during mineralization. Gold deposits in Wulong and Jiaodong share many similar geological characteristics. The pyrites from the Wulong deposit have higher δ 34 S values (+0.5 to +4.1‰) than the pyrrhotites (–1.2 to +1.2‰). Pyrites from the Jiaodong gold deposits show a wide range of both positive and negative δ 56 Fe values as well as high δ 34 S values, whereas those from the Wulong deposit have a relatively narrow range of positive δ 56 Fe values and near-zero δ 34 S values. The differences in Fe isotopes may be due to early precipitation of pyrrhotite with light Fe isotopes under a relatively low oxygen fugacity environment in the Wulong deposit, resulting in pyrite precipitated from the remaining fluids with heavy isotopes. The sulfur isotope variations between Wulong and Jiaodong gold deposits reflect differences in their source regions rather than oxygen fugacities. In addition, we have compiled Fe isotopic compositions of pyrites and pyrrhotites from different types of ore deposits to investigate their Fe isotopic behavior in magmatic-hydrothermal systems. Pyrite grains show a wide range of δ 56 Fe values. Positive pyrite δ 56 Fe values reflect an equilibrium isotope effect, whereas negative pyrite δ 56 Fe values may be due to kinetic isotope effects or to a mixture of sedimentary host rocks. Pyrrhotite grains show similar negative δ 56 Fe values, and they have a strong influence on the Fe isotope systematics in magmatic and hydrothermal systems. Our data show that Fe isotopes can be used to trace precipitation orders of pyrite and pyrrhotite and oxygen fugacity evolution in relatively reduced hydrothermal deposits.

Tungsten deposits are usually associated with granitic intrusions that record a long and complex evolution of the magmatic-hydrothermal system. However, the genetic link between magmatic-hydrothermal evolution and tungsten mineralization remains unclear. The Xihuashan tungsten deposit in South China, an important vein-type wolframite deposit, is closely associated with greisen and multiphase intrusive activity that produced biotite granite, two-mica granite, and muscovite granite. From the biotite granite to the two-mica granite to the muscovite granite, micas vary from siderophyllite to lithian siderophyllite, with decreasing K/Rb and Nb/Ta ratios and increasing Rb and Cs contents. The zoned micas in the muscovite granite and greisen display fluorine-depleted rims, reflecting subsolidus replacement by external aqueous fluids. The presence of siderite indicates a Fe-, Mn-, and CO 2 -rich fluid under reducing conditions. The micas in the greisen have higher-F contents and lower Fe 3+ /Fe 2+ ratios than those in the muscovite granite, suggesting that the fluids contributing to greisen formation had a relatively high-fluorine content and were reduced. The increase of CO 2 in the fluid enhanced its ability to unlock W from melts/rocks into fluids. The reducing environment also facilitated the tungsten mineralization. During greisenization, the pH value of the fluid increased, which destabilized the polymeric tungstates to form WO 4 2 – . The mixture of W-rich solution and Fe-, Mn-rich external fluid eventually precipitated as vein-type wolframite in favorable locations. An empirical equation (Li 2 O = 0.0748 × F 2 + 0.0893 × F) was introduced for estimating the Li 2 O contents of hydrothermal micas using the F contents determined by EPMA.

Since granitic rocks in high-grade terranes commonly undergo amphibolite-granulite facies metamorphic overprint, recovering magmatic records from the metamorphic modification remains a major challenge. Here, we report an early Paleozoic, garnet-bearing Yunlu charnockite that outcropped in the Yunkai terrane of the Cathaysia block from South China and underwent amphibole- grade metamorphic overprint in the late Devonian. Field observation, micro-texture, and mineral geochemistry combined with diffusion modeling constrain that the metamorphic overprint with an extremely short duration of ~0.2–0.5 Ma only influences a narrow rim of <100 μm for most minerals. The magmatic information can be retrieved by combining rhyolite-MELTS modeling with mineral thermobarometry using mineral core compositions to quantitatively estimate magmatic pressure, temperature, and melt H 2 O contents. Rhyolite-MELTS modeling results are evaluated by comparison with experimentally determined phase relations for a peraluminous granite with ~69.83 wt% SiO 2 at a pressure of ~500 MPa. The comparison suggests that the modeling reproduces phase relationships of feldspars and quartz within 20–60 °C when the melt H 2 O contents are below 7.0 wt%, but fails to account properly for all the phases when the melt H 2 O contents are higher than 7.0 wt%. The modeling results using reconstructed primary magma composition of the Yunlu charnockite combined with the orthopyroxene-garnet-plagioclase-quartz thermobarometry and fluid inclusion analyses suggest that the magma was emplaced at a pressure of ~600 MPa, a temperature of >900 °C, and an initial H 2 O content of ~4.0 wt% with rare CO 2 components. The orthopyroxene-garnet, biotite-garnet, and biotite-orthopyroxene thermometers yield a consistent temperature range of 770–820 ± 60 °C, which is significantly higher than the H 2 O-saturated solidus temperature of ~630 °C estimated from experimental results and two-feldspar thermometry. These results indicate that the early crystallized minerals (e.g., garnet, orthopyroxene, and some euhedral biotite) of the Yunlu charnockite equilibrate at higher temperatures with crystallinities of ~30–45%, rather than the H 2 O-saturated solidus conditions. We thus propose a hypothesis of melt extraction at 780–820 °C in a deep-seated, slowly cooling, partially crystalline magma reservoir. The melt extraction physically segregates the early crystallized minerals from residual interstitial melts, which inhibits element difusion equilibration between these minerals and interstitial melts. Granite thermometry commonly yields a large range of temperature estimations, which may be related to melt extraction events. Our study shows that melt extraction recorded in granites can be identified by combining micro-texture, mineral thermometry and rhyolite-MELTS modeling, which further provides quantitative insights into the fractionation process of silicic magmas.

Modeling of ionic diffusion in natural crystals has been developed over the last three decades to calculate timescales of geological processes. As the number of studies and the size of data sets have expanded, improvements in the precision of the general technique are needed to resolve temporal patterns that would otherwise be masked by large uncertainties. This contribution examines fundamental aspects of timescale calculation uncertainty using Mg-Fe zonation in olivine crystals from a Piton de la Fournaise oceanite erupted in 2002 CE. First, we quantitatively consider the role of geometric uncertainty in data sets from the perspectives of sectioning angle, crystal shape, and crystal agglomeration. Second, we assess how crystal growth and changing boundary conditions during diffusion pose problems for simplistic, 1D, diffusion-only modeling. An initial database of 104 timescales (7–45 days) was generated using typical, 1D, isothermal diffusion-only methods for profiles taken from 30 compositionally and texturally zoned crystals of olivine. This simplistic modeling yields poor model fits and imprecise timescales; prior to this work, we would have rejected >60% of these data. Universal-stage measurements of crystal boundary angles and three-dimensional (3D) X‑ray microcomputed tomography observations of crystal shape address geometric uncertainties. U-stage measurements show that, contrary to expectations of random sectioning, most boundaries modeled initially were close to the ideal sectioning plane. Assessment of crystal morphology from 2D thin sections suggests olivine crystals are dominantly euhedral; however, 3D imaging reveals that they are significantly subhedral and often exist as agglomerates, an observation that underscores both the potential for diverse crystal interactions through time in the magma (Wieser et al. 2019) and out-of-plane effects capable of influencing calculations of diffusion profiles. Refinements to timescale determination can be made using a dynamic 1D modeling code to resolve growth and changing boundary effects simultaneous with diffusion. We incorporated temperature-dependent crystal growth rates (both linear growth and quadratically increasing, with a peak growth rate ~1.9 ×10 –11 ms –1 ) and temperature-dependent boundary conditions (controlled using a cooling rate of –0.5 ± 0.1 °C/h) to remodel 13 timescales, resulting in significantly improved fits of the diffusion model to the initial data, better agreement between different faces of the same crystal, and less scatter within the whole data set.The use of 3D imaging and the inclusion of changing boundary conditions and crystal growth for diffusion calculations will enable more robust conclusions to be drawn from similar data in the future. Accurately retrieving timescale information from these crystals expands the pool of data available and reduces sampling bias toward “well-behaved” crystals.