Helium irradiation study on zircon

Synthetic ZrSiO₄ and (mildly to strongly radiation-damaged) natural zircon samples were irradiated with 8.8 MeV ⁴He²⁺ ions (fluences in the range 1 × 10¹³–5 × 10¹⁶ ions/cm²). For comparison, an additional irradiation experiment was done with 30 MeV ¹⁶O⁶⁺ ions (fluence 1 × 10¹⁵ ions/cm²). The light-ion irradiation resulted in the generation of new (synthetic ZrSiO₄) or additional (mildly to strongly metamict natural samples) damage. The maximum extent of the damage is observed in a shallow depth range approximately 32–33 μm (8.8 MeV He) and ~12 μm (30 MeV O) below the sample surface, i.e. near the end of the ion trajectories. These depth values, and the observed damage distribution, correspond well to defect distribution patterns as predicted by Monte Carlo simulations. The irradiation damage is recognised from the notable broadening of Raman-active vibrational modes, lowered interference colours (i.e. decreased birefringence), and changes in the optical activity (i.e. luminescence emission). At very low damage levels, a broad-band yellow emission centre is generated whereas at elevated damage levels, this centre is suppressed and samples experience a general decrease in their emission intensity. Most remarkably, there is no indication of notable structural recovery in pre-damaged natural zircon as induced by the light-ion irradiation, which questions the relevance of alpha-assisted annealing of radiation damage in natural zircon.     

Alteration of crystalline zircon solid solutions: a case study on zircon from an alkaline pegmatite from Zomba–Malosa, Malawi

A natural, altered zircon crystal from an alkaline pegmatite from the Zomba–Malosa Complex of the Chilwa Alkaline Province in Malawi has been studied by a wide range of analytical techniques to understand the alteration process. The investigated zircon shows two texturally and chemically different domains. Whereas the central parts of the grain (zircon I) appear homogeneous in backscattered electron images and are characterised by high concentrations of trace elements, particularly Th, U, and Y, the outer regions (zircon II) contain significantly less trace elements, numerous pores, and inclusions of thorite, ytttrialite, and fergusonite. Zircon II contains very low or undetectable concentrations of non-formula elements such as Ca, Al, and Fe, which are commonly observed in high concentrations in altered radiation-damaged zircon. U–Pb dating of both zircon domains by LA-ICPMS and SHRIMP yielded statistically indistinguishable U–Pb weighted average ages of 119.3 ± 2.1 (2σ) and 118 ± 1.2 (2σ) Ma, respectively, demonstrating that the zircon had not accumulated a significant amount of self-irradiation damage at the time of the alteration event. Electron microprobe dating of thorite inclusions in zircon II yielded a Th–U-total Pb model age of 122 ± 5 (2σ) Ma, supporting the age relationship between both zircon domains. The hydrothermal solution responsible for the alteration of the investigated zircon was alkaline and rich in CO₃ ²⁻, as suggested by the occurrence of REE carbonates and CO₂-bearing fluid inclusions. The alteration of the crystalline, trace element-rich zircon is explained by an interface-coupled dissolution-reprecipitation mechanism. During such a process, the congruent dissolution of the trace element-rich parent zircon I was spatially and temporally coupled to the precipitation of the trace element-poor zircon II at an inward moving dissolution-precipitation front. The driving force for such a process was merely the difference between the solubility of the trace element-rich and -poor zircon in the hydrothermal solution. The replacement process and the occurrence of mineral inclusions and porosity in the product zircon II is explained by the thermodynamics of solid solution-aqueous solution systems.     

A borehole investigation of zircon radiation damage annealing

This contribution reports Raman radiation damage measurements of zircons from the Kontinentale Tiefbohrung, on the western border of the Bohemian Massif. The mean wavenumbers (ω₃) and widths (Γ₃) of the ν₃(SiO₄) Raman band are constant down to 3 km, decrease (ω₃) resp. increase (Γ₃) between 3 and 5 km, and are again constant between 5 and 7 km. Uniform high Γ₃ values associated with ω₃ values close to those of undamaged zircon between 5 and 7 km are interpreted as due to residual damage predating the exhumation of the Bohemian Massif. A superimposed post‐exhumation signal indicates full damage retention down to 3 km depth, partial annealing between 3 and 5 km, and zero retention at greater depth. An attempt to calculate radiation damage ages gives results of a meaningful order of magnitude but also exposes difficulties associated with dating basement samples with complex damage accumulation and annealing histories.     

Applications of near-infrared FT-Raman spectroscopy in metamict and annealed zircon: oxidation state of U ions

We report a near-infrared Fourier-transform (FT) Raman spectroscopic method to characterize the electronic transitions of U ions and the alpha-decay damage in natural zircon. The application is demonstrated by analyzing metamict and annealed zircons from Sri Lanka. The data from crystalline zircon reveal a relatively sharp spectral feature appearing near 2733 cm^–1 in Stokes spectra with a laser excitation of 1064 nm. The feature is assigned as signals related to the previously reported U⁵⁺ absorption near 6668 cm^–1. With increasing self-irradiation dose, the feature shows a systematic decrease in intensity, accompanied by a gradual development of a broad feature between 3000 and 3400 cm^–1. On heating for 1 h, the U⁵⁺ feature shows an increase in intensity starting near ~700 K for partially metamict zircon, whereas for highly damaged zircon the first recovery of the feature takes place near 1000 K, accompanied by a decrease in the radiation-induced broad band. The changes observed in the present study reflect the variations of local environments of U ions in natural zircon during metamictization and thermal annealing.     

Phase equilibria in the MgO–SiO<Subscript>2</Subscript>–ZrO<Subscript>2</Subscript> system

The results of study of phase equilibria in the MgO–SiO₂–ZrO₂ system at 1450–1550°C are reported. The studied system contains two eutectic points and six fields: (I) MgSiO₃ + SiO₂; (II) MgSiO₃ + ZrO₂; (III) ZrSiO₄ + SiO₂; (IV) MgSiO₃ + Mg₂SiO₄; (V) ZrO₂ + MgO; (VI) ZrSiO₄ + ZrO₂. The presence of fields (II) and (III) on the diagram shows that zircon in equilibrium with olivine and pyroxene crystallizes at very low concentrations of ZrO₂ in the system. This provides a solution for one of the most important problems in zirconology of dunites: the probability of the formation and preservation of zircon in the course of the formation and evolution of dunite.     

Crystallization environment of Kazakhstan microdiamond: evidence from nanometric inclusions and mineral associations

Nanometric solid inclusions in diamond incorporated in garnet and zircon from felsic gneiss of the Kokchetav massif, Kazakhstan, have been examined utilizing electron microscopy and focused ion beam techniques. Host garnet and zircon contain numerous pockets of multiple inclusions, which consist of 1–3 diamond crystals intergrown with quartz, phengite, phlogopite, albite, K‐feldspar, rutile, apatite, titanite, biotite, chlorite and graphite in various combinations. Recalculation of the average chemical composition of the entrapped fluid represented by multiple inclusion pockets indicates that such fluid contained a low wt% of SiO₂, suggesting a relatively low‐temperature fluid rather than a melt. Transmission electron microscopy revealed that the diamond contains abundant nanocrystalline inclusions of oxides, rare carbonates and silicates. Within the 15 diamond crystals studied, abundant inclusions were found of SiO₂, TiO₂, Fe_xO_y, Cr₂O₃, ZrSiO₄, and single grains of Th_xO_y, BaSO₄, MgCO₃, FeCr₂O₄ and a stoichiometric Fe‐rich pyroxene. The diversity of trace elements within inclusions of essentially the same stoichiometry suggests that the Kokchetav diamond crystallized from a fluid containing variable amounts of Si, Fe, Ti, Cr, Zr, Ba, Mg and Th and other minor components such as K, Na, P, S, Pb, Zn, Nb, Al, Ca, Cl. Most of the components in crystals included in diamond appear to have their origin in the subducted metasediments, but some of them probably originate from the mantle. It is concluded that Kokchetav diamond most likely crystallized from a COH‐rich multicomponent supercritical fluid at a relatively low temperature (hence the apparently low content of rock‐forming elements), and that the diversity of major and minor components suggests interactions between subducted metasediments and mantle components.     

Solution behavior of +4 cations in high silica melts: petrologic and geochemical implications

The saturation surfaces of rutile (TiO₂), zircon (ZrSiO₄), and hafnon (HfSiO₄) were determined in anhydrous, peraluminous, high silica liquids of the system SiO₂-Al₂O₃-Na₂O-K₂O as functions of silica concentration at 1,400° C in air. The saturation concentrations of TiO₂, ZrO₂, and HfO₂ in rutile, zircon, and hafnon-saturated liquids, respectively, decrease smoothly and gradually as functions of increasing silica concentration. Thermodynamic analyses of the data demonstrate that the activity coefficients of TiO₂, ZrO₂, and HfO₂ increase smoothly and gradually as silica concentration is increased from 67 wt-% to 80 wt-%, and that changes in SiO₂ of 1 or 2 wt-% result in small changes in the saturation concentrations and activity coefficients of +4 cations. Because the solution behavior of +4 cations in highly siliceous liquids (>75 wt-% SiO₂) is predictably different than in less siliceous liquids (70 to 75 wt-% SiO₂), classification of highly-siliceous igneous rocks on the basis of silica concentration alone should not be interpreted to mean that their solution chemistry differs significantly from that of less siliceous rocks. The results of this study are compared with other studies of +4 cation solution behavior. From this it is concluded that variations in liquid compositions observed in cogenetic suites of high silica rhyolites cannot cause the observed changes in +4 cation concentrations. Thus, even if a large change in solution behavior of +4 cations is inferred from the large variations in their concentrations, it cannot be due to changes in bulk composition of the parental liquid. In addition, the similarity in the solution behavior of Zr and Hf seen in this study suggests that their solution mechanisms are similar. It is thus unlikely that liquid-state processes can fractionate one with respect to another, and variations in Zr/Hf ratios in suites of extrusive rocks are likely due to crystal-liquid equilibria, e.g., zircon fractionation.     

Prograde evolution of Sulu UHP metamorphic rock in Yangzhuang,Junan region,deduced by combined Raman and petrological studies

The prograde metamorphic history of the Sulu ultrahigh‐pressure metamorphic terrane has been revealed using Raman‐based barometry of the SiO₂ phases and other mineral inclusions in garnet porphyroblasts of a coesite eclogite from Yangzhuang, Junan region, eastern China. Garnet porphyroblasts have inner and outer segments with the boundary being marked by discontinuous changes in the grossular content. In the inner segment, the SiO₂ phase inclusions are α‐quartz with no coesite or relict features such as radial cracks. The residual pressures retained by the quartz inclusions systematically increase from the crystal centre to the margin of the inner segment. The metamorphic conditions estimated by calculation from the residual pressure and conventional thermodynamic calculation range from 500 to 630 °C and 1.3 to 2.3 GPa for the stage of the inner segment. Coesite and its pseudomorph occur as inclusions in the outer segment of the garnet and matrix omphacite. This occurrence of coesite is consistent with the pressure and temperature conditions of 660–725 °C and 3.1 GPa estimated by conventional geothermobarometry. Our results suggest that the quartz inclusions in the inner segment were trapped by garnet under α‐quartz‐stable conditions and survived phase transition to coesite at the peak metamorphic stage. The SiO₂ phases and other inclusions in the garnet have retained evidence of the pre‐eclogite prograde stage even during exhumation stage. The combined Raman spectroscopic and petrological approaches used here offers a powerful means for obtaining more robust constraints prograde stages involving garnet growth where different SiO₂ phases are present as inclusions.     

High-pressure Raman spectroscopic study of Co- and Ni-olivines

 The Raman spectra of synthetic α-Co₂SiO₄ and α-Ni₂SiO₄ olivines have been studied at room temperature and various pressures. All the Raman frequencies of the two olivines increase with increasing pressure, and most of the frequency–pressure plots obtained under both quasi- and nonhydrostatic conditions are nonlinear. It has been found that the average pressure derivative of Raman frequencies of the lattice modes in both Co- and Ni-olivines is smaller than that of the internal modes of SiO₄, indicating that the distortion of SiO₄ tetrahedra under static compression may be more severe than that of MO₆ octahedra. In addition, four new Raman bands were observed in Ni-olivine under nonhydrostatic compression and above 30 GPa. This result suggests that a new phase of Ni-olivine should be formed at 30 GPa or amorphization may occur at still higher pressure. Received: 11 July 2000 / Accepted: 19 December 2000     

Response of Precambrian zircon to the chemical abrasion (CA-TIMS) method and implications for improvement of age determinations

Electron backscatter imaging, Raman spectroscopy and U-Pb geochronology have been applied to Precambrian zircon grains that were annealed at 1000 and 1450 °C for various times, then leached with HF to constrain the conditions for healing radiation damage and attaining primary U-Pb zircon ages using the chemical abrasion (CA-TIMS) method. SEM images reveal a variety of textures for ZrO₂ overgrowths on 1450 °C annealed and leached zircon surfaces that depend on the degree of radiation damage and annealing history. Highly damaged zircon produces finer textures than zircon with less damage.Raman spectroscopy indicates that crystals with different levels of radiation damage are only partially restored by annealing at 1000 °C for 2-3 days. Longer annealing periods of 20 days are not noticeably more effective. Annealing at 1450 °C for 1 h results in partial breakdown of zircon but restores Raman peak widths and wave numbers to values characteristic of undamaged zircon after ZrO₂ overgrowths are removed by HF. Raman spectra are much less sensitive to polarization angle for annealed highly damaged grains than for weakly damaged zircon.U-Pb isotopic analyses of low to moderately damaged zircon (alpha fluence ranging up to 10¹⁹/g corresponding to an amorphization volume fraction of 80% or more) yield almost concordant data (0.3-0.5% discordance) after high-temperature annealing at 1450 °C followed by HF leaching at 195 °C. Analyses of cracked zircon annealed at 1450 °C and leached may remain discordant but those of uncracked grains are concordant. Most analyses show primary ²⁰⁷Pb/²⁰⁶Pb ages although cracked grains annealed at 1450 °C may produce discordant data with ²⁰⁷Pb/²⁰⁶Pb ages that are too young after leaching. The solubility of highly damaged, very disordered zircon (amorphization level of 99%) is only slightly reduced by annealing, and analyses of leach residues are strongly discordant although primary ²⁰⁷Pb/²⁰⁶Pb ages are obtained.Annealing of highly damaged zircon under any conditions apparently results in a mass of randomly oriented micro-crystals that pseudomorph the original grain. This could explain the fine-scale pattern observed on etched crystal surfaces, reduced anisotropy at the 5 μm scale of the Raman laser beam and high solubility in HF. It may be impossible to restore primary U-Pb isotopic ages in such cases but precise ages can still potentially be determined from ²⁰⁷Pb/²⁰⁶Pb ratios or by application of the air abrasion method.     

Experimental determination of the Raman CH4 symmetric stretching (ν1) band position from 1-650 bar and 0.3-22 °C: Application to fluid inclusion studies

The position of the Raman methane (CH₄) symmetric stretching band (ν₁) over the range 1-650 bar and 0.3-22 °C has been determined using a high-pressure optical cell mounted on a Raman microprobe. Two neon emission lines that closely bracket the CH₄ band were collected simultaneously with each CH₄ spectrum. The peak position was determined after least squares fitting using a summed Gaussian-Lorentzian method, resulting in a precision of ≈±0.02 cm⁻¹ in peak position determination. The CH₄ν₁ band position shifts to lower wave number with increasing pressure. At a given pressure, the band shifts to lower wave number with decreasing temperature, and the magnitude of the temperature shift increases with increasing pressure. The relationship between the Raman CH₄ν₁ band position and temperature and pressure determined here may be used to estimate the internal pressure in natural or synthetic CH₄-bearing fluid inclusions. This information, in turn, may be used to determine the density of pure CH₄ fluid inclusions and the salinity of CH₄-bearing aqueous inclusions.     

Theoretical study of point defects in crystalline zircon

We present a numerical study of point defects in crystalline zircon (ZrSiO₄). Vacancies and interstitials of all the constituents of zircon have been considered. For each defect, the structure and the formation energies have been calculated. Calculations, using the supercell method, are based on the Density Functional Theory in the Local Density Approximation. Empirical potentials have also been considered for comparison with electronic structure results. We find a formation energy for the oxygen interstitial of 1.7 eV. This value is compatible with the experimental activation energy for oxygen diffusion in zircon, which proves an interstitial mechanism for the diffusion of oxygen in zircon. For all other defects the calculated formation energies lead to negligible thermal concentration at equilibrium. Received: 8 January 1999 / Revised, accepted: 14 May 1999     

A hitherto unrecognised band in the Raman spectra of silica rocks: influence of hydroxylated Si–O bonds (silanole) on the Raman moganite band in chalcedony and flint (SiO<Subscript>2</Subscript>)

Chalcedony is a spatial arrangement of hydroxylated nanometre-sized α-quartz (SiO₂) crystallites that are often found in association with the silica mineral moganite (SiO₂). A supplementary Raman band at 501 cm⁻¹ in the chalcedony spectrum, attributed to moganite, has been used for the evaluation of the quartz/moganite ratio in silica rocks. Its frequency lies at 503 cm⁻¹ in sedimentary chalcedony, representing a 2 cm⁻¹ difference with its position in pure moganite. We present a study of the 503 cm⁻¹ band’s behaviour upon heat treatment, showing its gradual disappearance upon heating to temperatures above 300 °C. Infrared spectroscopic measurements of the silanole (SiOH) content in the samples as a function of annealing temperature show a good correlation between the disappearance of the 503 cm⁻¹ Raman band and the decrease of structural hydroxyl. Thermogravimetric analyses reveal a significant weight loss that can be correlated with the decreasing of this Raman band. X-ray powder diffraction data suggest the moganite content in the samples to remain stable. We propose therefore the existence of a hitherto unknown Raman band at 503 cm⁻¹ in chalcedony, assigned to ‘free’ Si–O vibrations of non-bridging Si–OH that oscillate with a higher natural frequency than bridging Si–O–Si (at 464 cm⁻¹). A similar phenomenon was recently observed in the infrared spectra of chalcedony. The position of this Si–OH-related band is nearly the same as the Raman moganite band and the two bands may interfere. The actually observed Raman band in silica rocks might therefore be a convolution of a silanole and a moganite vibration. These findings have broad implications for future Raman spectroscopic studies of moganite, for the assessment of the quartz/moganite ratio, using this band, must take into account the contribution from silanole that are present in chalcedony and moganite.     

New thermodynamic models and revised calibrations for the Ti-in-zircon and Zr-in-rutile thermometers

The models recognize that ZrSiO₄, ZrTiO₄, and TiSiO₄, but not ZrO₂ or TiO₂, are independently variable phase components in zircon. Accordingly, the equilibrium controlling the Zr content of rutile coexisting with zircon is ZrSiO₄ = ZrO₂ (in rutile) + SiO₂. The equilibrium controlling the Ti content of zircon is either ZrSiO₄ + TiO₂ = ZrTiO₄ + SiO₂ or TiO₂ + SiO₂ = TiSiO₄, depending whether Ti substitutes for Si or Zr. The Zr content of rutile thus depends on the activity of SiO₂ as well as T, and the Ti content of zircon depends on and as well as T. New and published experimental data confirm the predicted increase in the Zr content of rutile with decreasing and unequivocally demonstrate that the Ti content of zircon increases with decreasing . The substitution of Ti in zircon therefore is primarily for Si. Assuming a constant effect of P, unit and that and are proportional to ppm Zr in rutile and ppm Ti in zircon, [log(ppm Zr-in-rutile) + log] = A₁ + B₁/T(K) and [log(ppm Ti-in-zircon) + log − log] = A₂ + B₂/T, where the A and B are constants. The constants were derived from published and new data from experiments with buffered by either quartz or zircon + zirconia, from experiments with defined by the Zr content of rutile, and from well-characterized natural samples. Results are A₁ = 7.420 ± 0.105; B₁ = −4,530 ± 111; A₂ = 5.711 ± 0.072; B₂ = −4,800 ± 86 with activity referenced to α-quartz and rutile at P and T of interest. The zircon thermometer may now be applied to rocks without quartz and/or rutile, and the rutile thermometer applied to rocks without quartz, provided that and are estimated. Maximum uncertainties introduced to zircon and rutile thermometry by unconstrained and can be quantitatively assessed and are ≈60 to 70°C at 750°C. A preliminary assessment of the dependence of the two thermometers on P predicts that an uncertainty of ±1 GPa introduces an additional uncertainty at 750°C of ≈50°C for the Ti-in-zircon thermometer and of ≈70 to 80°C for the Zr-in-rutile thermometer.     

Nouvelles synthéses hydrothermales du zircon: substitutions isomorphiques; relation morphologie-milieu de croissance

Zircon has been synthesized from a large number of mineral salts and also from one organic compound of zirconium. Its crystallization requires an acid reactive environment.The obtained crystals have been studied morphologically and the crystallographic forms obtained are linked to the presence of specific ions.The study of substitutions in the zircon network shows that hydro-zircon αZr(SiO₄)_1?x(OH)_4x will form only when fluorine is present. Other varieties of the α and β phases ZrSiO₄ have been obtained, substituting Hf, U, for Zr, and Ge for Si and an immiscible α phase ThSiO₄.Germanium enters the zircon network only at low temperature. Uranium and Hf, but not Th, can substitute for Zr in zircon.     

Microstructural constraints on the mechanisms of the transformation to reidite in naturally shocked zircon

Zircon (ZrSiO₄) is used to study impact structures because it responds to shock loading and unloading in unique, crystallographically controlled manners. One such phenomenon is the transformation of zircon to the high-pressure polymorph, reidite. This study quantifies the geometric and crystallographic orientation relationships between these two phases using naturally shocked zircon grains. Reidite has been characterized in 32 shocked zircon grains (shocked to stages II and III) using a combination of electron backscatter diffraction (EBSD) and focused ion beam cross-sectional imaging techniques. The zircon-bearing clasts were obtained from within suevite breccia from the Nördlingen 1973 borehole, close to the center of the 14.4 Ma Ries impact crater, in Bavaria, Germany. We have determined that multiple sets (up to 4) of reidite lamellae can form in a variety of non-rational habit planes within the parent zircon. However, EBSD mapping demonstrates that all occurrences of lamellar reidite have a consistent interphase misorientation relationship with the host zircon that is characterized by an approximate alignment of a {100}_zircon with a {112}_reidite and alignment of a {112}_zircon with a conjugate {112}_reidite. Given the tetragonal symmetry of zircon and reidite, we predict that there are eight possible variants of this interphase relationship for reidite transformation within a single zircon grain. Furthermore, laser Raman mapping of one reidite-bearing grain shows that moderate metamictization can inhibit reidite formation, thereby highlighting that the transformation is controlled by zircon crystallinity. In addition to lamellar reidite, submicrometer-scale granules of reidite were observed in one zircon. The majority of reidite granules have a topotaxial alignment that is similar to the lamellar reidite, with some additional orientation dispersion. We confirm that lamellar reidite likely forms via a deviatoric transformation mechanism in highly crystalline zircon, whereas granular reidite forms via a reconstructive transformation from low-crystallinity ZrSiO₄ within the reidite stability field. The results of this study further refine the formation mechanisms and conditions of reidite transformation in naturally shocked zircon.     

Phase transition of zircon at high P-T conditions

Abstract In situ observations of the zircon-reidite transition in ZrSiO₄ were carried out using a multianvil high-pressure apparatus and synchrotron radiation. The phase boundary between zircon and reidite was determined to be P (GPa) = 8.5+0.0017×(T-1200) (K) for temperatures between 1100–1900 K. When subducted slabs, including igneous rocks and sediments, descend into the upper mantle, the zircon in the subducted slab transforms into reidite at pressures of about 9 GPa, corresponding to a depth of 270 km. Reidite found in an upper Eocene impact ejecta layer in marine sediments is thought to have been transformed from zircon by a shock event. The peak pressure generated by the shock event in this occurrence is estimated to be higher than 8 GPa.Editorial responsibility: J. Hoefs     

A unified equation for calculating methane vapor pressures in the CH4-H2O system with measured Raman shifts

A unified equation has been derived by using all available data for calculating methane vapor pressures with measured Raman shifts of C-H symmetric stretching band (υ₁) in the vapor phase of sample fluids near room temperature. This equation eliminates discrepancies among the existing data sets and can be applied at any Raman laboratory. Raman shifts of C-H symmetric stretching band of methane in the vapor phase of CH₄-H₂O mixtures prepared in a high-pressure optical cell were also measured at temperatures between room temperature and 200 °C, and pressures up to 37 MPa. The results show that the CH₄υ₁ band position shifts to higher wavenumber as temperature increases. We also demonstrated that this Raman band shift is a simple function of methane vapor density, and, therefore, when combined with equation of state of methane, methane vapor pressures in the sample fluids at elevated temperatures can be calculated from measured Raman peak positions. This method can be applied to determine the pressure of CH₄-bearing systems, such as methane-rich fluid inclusions from sedimentary basins or experimental fluids in hydrothermal diamond-anvil cell or other types of optical cell.     

Raman spectra of hydrous γ -Mg2SiO4 at various pressures and temperatures

 One well-defined OH Raman band at 3651 ± 1 cm⁻¹ and one weak feature near 3700 ± 5 cm⁻¹ are recognized for the hydrous γ-phase of Mg₂SiO₄. Like the hydrous β-phase, the H₂O content in the γ-phase shifts most of the corresponding silicate modes towards lower frequencies. Variations in Raman spectra of the hydrous γ-phase were investigated up to about 200 kbar at room temperature and in the range 81–873 K at atmospheric pressure. Unlike the anhydrous γ-phase, which remains intact up to at least 873 K, the hydrous γ-phase sometimes converts to a defective forsterite structure above 800 K. Although the hydrous γ-phase remains intact up to at least 800 K, Raman signals of the OH bands disappear completely above 423 K. The Raman frequency of the well-defined OH band decreases linearly with increasing temperature between 81 and 423 K. In the region of the silicate vibrations, the Raman frequencies of the two most intense bands increase nonlinearly with increasing pressure, and decrease with increasing temperature. The frequencies for all other weak bands, however, decreased linearly with increasing temperature. The latter most likely reflects the larger scatter of the data for the weak bands. Received: 27 April 2001 / Accepted: 12 September 2001     

Hydrous olivine (Mg1? y? Fe2+y)2? x ?v x ? SiO4H2 x – a new DHMS phase of variable composition observed as nanometer-sized precipitations in mantle olivine

 An olivine grain from a peridotite nodule 9206 (Udachnaya kimberlite, Siberia) was investigated by TEM methods including AEM, HRTEM, SAED and EELS techniques. A previous study of the 9206 olivine sample revealed OH absorption bands in the IR spectrum and abundant nanometer-sized OH-bearing inclusions, of hexagonal-like or lamellar shape. Inclusions, which are several hundred nm in size, consist of 10 ? phase, talc and serpentine (chrysotile and lizardite). The lamellar (LI) and hexagon-like small inclusions of several ten nm in size (SI) are the topic of the present paper. AEM investigations of the inclusions reveal Mg, Fe and Si as cations only. The Mg/Si and Fe/Si atomic ratios are lower in the inclusions than in the host olivine. The Si concentration in the olivine host and both lamellar inclusions and small inclusions is the same. A pre-peak at 528eV was observed in EEL spectra of LI and SI, which is attributed to OH⁻ or Fe³⁺. From these data it is concluded that there is a OH⁻- or Fe³⁺-bearing cation-deficient olivine-like phase present. HRTEM lattice fringe images of LI and SI exhibit modulated band-like contrasts, which are superimposed onto the olivine lattice. Diffraction patterns (Fourier-transforms) of the HREM images as well as SAED patterns show that the band-like contrasts in HRTEM images of the inclusions are caused by periodic modulations of the olivine lattice. Three kinds of superperiodicity in the olivine structure such as 2a, 3a and 3c, were observed in SAED patterns. The corresponding olivine supercells labelled here as Hy-2a, Hy-3a and Hy-3c were derived. The M1-vacancies located in the (100) and (001) octahedral layers of the olivine lattice are suggested to form ordered arrays of planar defects (PD), which cause the band-like contrasts in HRTEM patterns as well as the superperiodicity in the SAED patterns. The vacancy concentrations as well as the chemical composition of Hy-2a, Hy-3a and Hy-3c olivine supercells were calculated using crystal chemical approaches, assuming either {(OH)^< _O−V^" _Me−(OH)^< _O}^↔, or {F e ^< _Fe −H _Me ^′}^↔ or {2F e ^< _Fe −V _Me ^"}^↔ point defect associates. The calculated theoretical compositions Mg_1.615Fe⁺² _0.135v_0.25SiO₄H_0.5 (Hy-2a) and Mg_1.54Fe²⁺ _0.12v_0.33SiO₄H_0.66 (Hy-3a and Hy-3c) are in a good agreement with the AEM data on inclusions. Hy-2a, Hy-3a and Hy-3c are considered to be a hydrous olivine with the extended chemical formula (Mg_1-yFe²⁺ _y)_2−xv_xSiO₄H_2x. The crystal structure of hydrous olivine is proposed to be a modular olivine structure with Mg-vacant modules. The crystal chemical formula of hydrous olivines in terms of a modular structure can be written as [MgSiO₄H₂] · 3[Mg_1.82Fe_0.18SiO₄] for Hy-2a, [MgSiO₄H₂] · 2[Mg_1.82Fe_0.18SiO₄] for Hy-3a and Hy-3c. Hydrous olivine is suggested to be exsolved from the olivine 9206, which has been initially saturated by OH-bearing point defects. The olivine 9206 hydration as well as the following exsolution of hydrous olivine inclusions is suggested to occur at high pressure-high temperature conditions of the upper mantle. Received: 15 January 2001 / Accepted: 2 July 2001