Cation distribution in low-calcium pyroxenes: Dependence on temperature and calcium content and the thermal history of lunar and terrestrial pigeonites

Four pyroxenes with compositions En₄₈Fs₄₈Wo₄, En_47·5Fs_47·5Wo₅, En₄₅Fs₄₅Wo₁₀ and En₄₀Fs₄₀Wo₂₀, synthesized at 1200°C at atmospheric pressure, were heat-treated at 500, 600, 700, and 800°C for various lengths of time. These pyroxenes are variously ordered with respect to Fe²⁺ and Mg²⁺ without unmixing. The Fe²⁺-Mg²⁺ distribution over the two nonequivalent sites M1 and M2, determined through Mössbauer spectroscopy, is found to be a function of both temperature and concentration of Ca²⁺ at the M2 site. The preference of Fe²⁺ for the M2 site increases with decreasing temperature and increasing Ca²⁺. These data can be used to determine cation equilibration temperatures of lunar and terrestrial pigeonites. The lunar pigeonites usually indicate equilibration temperatures of 700–860°C, except the pigeonite from rock 14053, which may have been subjected to shock heating due to meteoritic impact.     

Iron-magnesium order-disorder in an orthopyroxene crystal from the Johnstown meteorite

Equilibrium reversals of Fe²⁺Mg distribution between the M1 and M2 sites of an orthopyroxene from the Johnstown meteorite were achieved at several temperatures between 700 and 1000°C. One single crystal was used for the whole thermal treatment and for collecting all the X-ray data after quenching. The intracrystalline ion exchange for the bulk chemical composition: (Mg_1.453Fe_0.441Cr_0.024Ca_0.054Mn_0.015Fe_0.005Ti_0.003Al_0.005)(Si_1.960Al_0.040)O₆ is given by: ln K_D = −3027(±39)/T(K) + 0.872(±0.013)> where K_D is the distribution coefficient for the reaction: Fe_M2²⁺ + Mg_M1 = Mg_M2 + Fe_M1²⁺.A transmission electron microscopy (TEM) study of part of the crystal showed the presence of very thin augite lamellae and Guinier-Preston zones indicating a relatively rapid cooling of the host rock at temperatures close to 1000°C. The new temperature scale yields a relatively high quenching temperature of 379 (±8)°C for the pyroxene which appears consistent with a rapid cooling (estimated few degrees per hundred years) of a magmatic cumulate excavated by an impact on its parental body.     

Synthesis of γ-Mg2SiO4

Magnesium orthosilicate with spinel structure (γ-Mg₂SiO₄) was synthesized at about 250 kbar and 1000°C. Unit cell dimension was established to be 8.076 ± 0.001Å. X-ray powder diffraction pattern revealed a significant difference between γ-Mg₂SiO₄ and other γ-M₂SiO₄ spinels (M = Fe, Co, and Ni) in the intensities of (111) and (331) reflections, both of which are virtually absent in the Mg₂SiO₄ spinel. This feature could be thoroughly understood by the calculation of the intensities for several silicate spinels.     

Thermal equation of state of magnesiowüstite (Mg0.6Fe0.4)O

Volume measurements for magnesiowüstite (Mg_0.6Fe_0.4)O, were carried out up to pressures of 10.1 GPa in the temperature range 300–1273 K, using energy-dispersive synchrotron X-ray diffraction. These data allow reliable determination of the temperature dependence of the bulk modulus and good constraint on the thermal expansitivity at ambient pressure which was previously not known for magnesiowüstite. From these data, thermal and elastic parameters were derived from various approaches based on the Birch–Murnaghan equation of state (EOS) and on the relevant thermodynamic relations. The results from three different equations of state are remarkably consistent. With (∂K_T/∂P)_T fixed at 4, we obtained K₀=158(2) GPa, (∂K_T/∂T)_P=−0.029(3) GPa K⁻¹, (∂K_T/∂T)_V=−3.9(±2.3)×10⁻³ GPa K⁻¹, and α_T=3.45(18)×10⁻⁵+1.14(28)×10⁻⁸T. The K₀, (∂K_T/∂T)_P, and (∂K_T/∂T)_V values are in agreement with those of Fei et al. (1992) and are similar to previously determined values for MgO. The zero pressure thermal expansitivity of (Mg_0.6Fe_0.4)O is found to be similar to that for MgO (Suzuki, 1975). These results indicate that, for the compositional range x=0–0.4 in (Mg_1−xFe_x)O, the thermal and elastic properties of magnesiowüstite exhibit a dependence on the iron content that is negligibly small, within uncertainties of the experiments. They are consequently insensitive to the Fe–Mg partitioning between (Mg, Fe)SiO₃ perovskite and magnesiowüstite when applied to compositional models of the lower mantle. With the assumption that (Mg_0.6Fe_0.4)O is a Debye-like solid, a modified equation of heat capacity at constant pressure is proposed and thermodynamic properties of geophysically importance are calculated and tabulated at high temperatures.     

Elasticity of single crystal pyrope and implications for garnet solid solution series

The elastic moduli of a synthetic single crystal of pyrope (Mg₃Al₂Si₃O₁₂) have been determined using a technique based on Brillouin scattering. These results are used in an evaluation of the effect of composition on the elastic properties of silicate garnet solid solution series (Mg, Fe, Mn, Ca)₃ (Al, Fe, Cr)₂ Si₃O₁₂. In the pyralspites (Mg FeMn aluminum garnets), for which a large amount of data is available, this analysis indicates that the bulk modulus K is independent of the Fe²⁺/Mg²⁺ ratio, which is similar to the behavior observed in olivines and pyroxenes. However, the shear modulus μ of the garnets increases by 10% from the Mg to the Fe end member, in contrast to the decrease of μ with Fe content which is observed in olivines and pyroxenes. This contrasting behavior is most probably related to the oxygen coordination of the cation site occupied by Mg²⁺ and Fe²⁺ in these different minerals.     

Magnetic properties of synthetic titanomagnetites

Magnetic properties and crystal structure parameters of synthetic solid solutions Fe₃O₄Fe₃TiO₄, Fe₂O₄MgFe₂O₄ and Fe₃O₄Mg₂TiO₄ have been studied. Basic regularities in the behaviour of saturation magnetisation (I_s), Curie temperature (T_C) and cubic lattice parameter a during the substitution of Ti and Mg ions for Fe ions have been found. As the concentration of Ti ions increases, I_s reduces from 70 Gs·cm³ g^?1 to 0, T_C changes from 580 to 130°C and a from 8.391 to 8.520 Å. Growth of the Mg concentration leads to changes in I_s to 19.8 Gs·cm³, g^?1, T_C, to 440°C and a, to 8.360 Å. The full Fe ions substitution gives $\text{“a”=8.440}\text{A}\text{?}$.Chemical compositions of the samples, in which the valency variation of Fe ions at oxidation leads to an increase in susceptibility and T_C, have been determined.     

Crystal structure and compositional variation of Angra dos Reis fassaite

The crystal structure of fassaite from the Angra dos Reis meteorite has been determined by least-squares refinement of three-dimensional X-ray data to anR value of 3.3%. The pyroxene is monoclinic, space groupC2/c, with unit-cell dimensionsa = 9.738(1),b = 8.874(2),c = 5.2827(5)Å, β = 105.89(1)°, andV = 439.1(1)ų. Average bond lengths are (Si,Al)-O = 1.651, M1-O = 2.061, and M2-O = 2.489Å. The distribution of iron and magnesium between M1 and M2 suggests a temperature of equilibration greater than 1000°C.Electron microprobe analysis of several fassaite grains reveals small but statistically significant variations of (Mg + Si) versus (Al - Ti). The range of fassaite composition may be represented byEn₃Hd₂₂TiCpx₆(Di_53±2CaTs_16?2) whereEn=Mg₂Si₂O₆,Hd=CaFeSi₂O₆,TiCpx=CaTiAl₂O₆,Di=CaMgSi₂O₆,CaTs=CaAl₂SiO₆. Most fassaite analyses calculated on the basis of four cations yielded greater than six anions, suggesting that part of the titanium or chromium might be reduced to Ti³⁺ or Cr²⁺.     

A new high-pressure phase of spinel

A new phase which is much denser than the component oxides of spinel (MgAl₂O₄) was synthesised at loading pressures greater than 250 kbar and at about 1000°C in a diamond-anvil press coupled with laser heating. The new phase (ε-MgAl₂O₄) was indexed on the basis of an orthorhombic cell with a = 8.507 ± 0.004, b = 2.740 ± 0.003, c = 9.407 ± 0.005Å, and Z = 4 at room temperature and 1 bar pressure. Thus the molar volume for ε-MgAl₂O₄ at the above conditions was calculated to be 33.01 ± 0.07 cm³, which is 10.3% less than that of the mixture corundum plus periclase. The dense phase of spinel found in shock-wave experiments can be reasonably interpreted as ε-MgAl₂O₄, and this may be a potentially important mineral component of the earth's lower mantle. The new structure may also provide a possible candidate for the dense phases of Fe₃O₄ and Mg₂SiO₄ which were found by shock experiments.     

Phase petrology with special reference to pyroxenes of the acid igneous complex of Barda hills,Western Saurashtra (Gujarat)

The Barda igncous complex of western Saurashtra (Gujarat) consists of a group of volcanic and subvolcanic igneous rocks, viz., pitchstone, rhyolite, felsite, granophyre and dioritic rocks. They are generally grouped under the Deccan Trap series and have been emplaced through the Deccan Trap basalt flows. Using different varieties of clinopyroxenes as characteristic phases several units of felsites and granophyres have been distinguished in the southeastern part of the Barda hills. The augites of the adjacent Deccan Trap basalt flows range between Ca₃₃Mg₆₁Fe₆ and Ca₄₁Mg₄₄Fe₁₅. (All compositions of minerals are derived from optical data). The clinopyroxene of a monzodiorite is Ca₄₁Mg_42.5Fe_16.5. In the felsites and granophyres the clinopyroxenes range from salite (Ca₄₆Mg₃₁Fe₂₃) through ferrosalite (Ca₄₆Mg₂₀Fe₃₄) and ferroaugite (Ca₄₃Mg₁₇Fe₄₆) to hedenbergite (Ca₄₆Mg₉Fe₄₅). Beyond that a peralkaline trend is indicated by common occurrence of aegirine-augite granophyre in which the clinopyroxene is alkalic with 20 percent of NaFe³ Si₂O₆ (acmite) molecule. Several bodies of acmite-riebeckitebearing granophyres have been found; they also indicate a high degree of oxidation. A fayalitic olivine-bearing granophyre is of interest as it shows the strongest iron enrichment with 13 weight percent FeO + Fe₂O₃; its mineralogy indicates a very low oxygen fugacity of 10¹³ atmospheres during crystallization.     

Mineralogy and petrology of sample 67075 and the origin of lunar anorthosites

Plagioclase in cataclastic anorthosite 67075 occurs as angular matrix grains and as recrystallized clasts of micro-anorthosite. Olivines are Fe-rich and fall into two compositional groupings. Large grains of pyroxene show exceptionally well-developed exsolution lamellae analogous to those observed in pyroxenes from layered complexes. The low-Ca component in both pigeonites and augites shows varying degrees of inversion to orthopyroxene. The lattices of host and lamellae may deviate slightly (up to 2°) from the ideal orientation. Very slow cooling from magmatic temperatures is required to produce the coarse exsolution textures and inversion features. Augite macrocrystals are distinctly subcalcic indicating crystallization at temperatures around1100 ± 50°C while host-lamellae pairs and small grains in lithic clasts and matrix indicate reequilibration on a micron scale to temperatures less than 800°C. Pyroxene compositions tend to cluster into two groups both of which are among the most Fe-rich reported for highland pyroxenes. Ti and Al contents of pyroxenes are very low and Ti, Cr, and Mn follow well-established magmatic differentiation trends. The high Cr content may reflect low?O₂ conditions and/or early crystallization of olivine and plagioclase.The⁸⁷Sr/⁸⁶Sr ratios in lunar anorthosites are the lowest reported for any lunar rock. It is likely that anorthosites formed as cumulates during the major differentiation episode which occurred prior to～4.3AE. Recrystallization features are common and³⁹Ar/⁴⁰Ar ages cluster around 4.0 AE. This may be the result of the intense bombardment prior to 4.0 AE which caused repeated cycles of in-situ fracturing and granulation followed by recrystallization. The low siderophile element content and the inferred slow cooling indicate a plutonic source region (10km) not frequently plumbed by impact events. The Fe-rich silicates indicate crystallization from a melt at an advanced stage of fractionation. However, the low REE abundances are not consistent with late-stage crystallization. Plagioclase apparently crystallized relatively early and was concentrated by flotation and/or convection currents while the mafic minerals crystallized from a fractionated trapped liquid. The chemical, isotopic, and mineralogical data place stringent constraints on the nature of genetically related rocks and the relationship of anorthosites to other members of the ANT suite does not appear to be one ofsimple fractionation. The data presented in this paper are consistent with the Taylor-Jake?model of lunar evolution.     

Oxysulfates of copper, sodium, and potassium in the lava flows of the 2012–2013 Tolbachik Fissure Eruption

Samples from the surface of lava flows discharged by the 2012–2013 Tolbachik Fissure Eruption were found to contain oxysulfates of copper, sodium, and potassium: K₂Cu₃O(SO₄)₂ (fedotovite), NaKCu₂O(SO₄)₂, and Na₃K₅Cu₈O₄(SO₄)₈. The last two phases have no naturally occurring or synthetic analogues that we are aware of. They form flattened crystals of prismatic to long-prismatic habits. The crystals of Na₃K₅Cu₈O₄(SO₄)₈ have a chemical composition corresponding to the empirical formula Na_2.22K_5.47Cu_8.02S_8.05O₃₆. An X-ray analysis of this compound showed that it has a monoclinic symmetry, P2/c, a = 13.909(4), b = 4.977(1), c = 23.525(6) Å, β = 90.021(5)°, V = 1628.3(7) ų. The crystal structure was determined by direct techniques and refined to yield R ₁ for 3955 reflexes//web// with F ² > 4σF. The compound NaKCu₂O(SO₄)₂ also belongs to the monoclinic system, P2/c, a = 14.111(4), b = 4.946(1), c = 23.673(6) Å, β = 92.052(6)°, V = 1651.1(8) ų. The structure was determined by direct techniques to yield a tentative structural model that has been refined up to R ₁ = 0.135 for 4088 reflexes with F ² > 4σF. The crystal structure of Na₃K₅Cu₈O₄(SO₄)₈ is based on chains of [O₂Cu₄]⁴⁺ consisting of rib-coupled oxy-centered tetrahedrons of (OCu₄)⁶⁺. The chains are surrounded by sulfate radicals, resulting in columns of {[O₂Cu₄](SO₄)₄}^4? aligned along the b axis. The interchain space contains completely ordered positions of Na⁺ and K⁺ cations. The principle underlying the connection of NaKCu₂O(SO₄)₂ columns in the crystal structure of {[O₂Cu₄](SO₄)₄}^4? is different, in view of the relation Na:K = 1 as contrasted with 3:5 for the compound Na₃K₅Cu₈O₄(SO₄)₈. The presence of oxy-centered tetrahedrons in the structure of these new compounds furnishes an indirect hint at the importance of polynuclear copper-oxygen radicals with centering oxygen atoms as forms of transport of copper by volcanic gases.     

Shock behavior of zircon: phase transition to scheelite structure and decomposition

Both single-crystal and powdered specimens of zircon (ZrSiO₄) were shocked to peak pressures between 30 and 94 GPa using the gun method, and specimens recovered were studied by means of X-ray diffraction analysis, transmission electron microscopy and infrared spectroscopy. Transformation to the scheelite structure started above 30 GPa, and was completed above 53 GPa in the case of single crystal specimens. Tetragonal unit cell parameters of the scheelite type ZrSiO₄ at room condition are measured to bea = 4.7341(1)Å, c = 10.51(1)Å, c/a = 2.219(2) andV = 235.5(2)ų, which is smaller than that of the zircon type by 9.9%. The recovered scheelite-type ZrSiO₄ reverts to the zircon type after rapid heating to 1200°C at room pressure. This transformation from the zircon type to the scheelite type is unique in that it is fast, displacive-like, but does not reverse. Tetragonal ZrO₂ was detected as decomposition product in the single-crystal specimen shocked to 94 GPa, and further confirmed in a powdered specimen shocked to 53 GPa where enhancement of temperature is expected because of high porosity. Decomposition behavior of zircon observed in natural shock events is discussed on the basis of present experimental results.     

Water and nutrient fluxes in a cool temperate rainforest at the Cordillera de la Costa in southern Chile

Water and nutrient fluxes were studied during a 12-month period in an alerce (Fitzroya cupressoides) forest, located in a remote site at the Cordillera de la Costa (40°05′S) in southern Chile. Measurements of precipitation, throughfall, stemflow, effective precipitation, soil infiltration and stream flow were carried out in an experimental, small watershed. Simultaneously, monthly water samples were collected to determine the concentrations and transport of organic-N, NO₃-N, total-P, K⁺, Ca²⁺, Na⁺ and Mg²⁺ in all levels of forest. Concentration of organic-N, NO₃-N, total-P and K⁺ showed a clear pattern of enrichment in the throughfall, stemflow, effective precipitation and soil infiltration. For Ca²⁺ and Mg²⁺, enrichment was observed in the effective precipitation, soil infiltration and stream flow. Annual transport of K⁺, Na⁺, Ca²⁺ and Mg²⁺ showed that the amounts exported from the forest via stream flow (K⁺=0·95, Na⁺=32·44, Ca²⁺=8·76 and Mg²⁺=7·16 kg ha⁻¹ yr⁻¹) are less than the inputs via precipitation (K⁺=6·39, Na⁺=40·99, Ca²⁺=15·13 and Mg²⁺=7·61 kg ha⁻¹ yr⁻¹). The amounts of organic-N and NO₃-N exported via stream flow (organic-N=1·04 and No₃-N=3·06 kg ha⁻¹ yr⁻¹) were relatively small; however, they represented greater amounts than the inputs via precipitation (organic-N=0·74 and NO₃-N=0·97 kg ha⁻¹ yr⁻¹), because of the great contribution of this element in the superficial soil horizon, where the processes of decomposition of organic material, mineralization and immobilization of the nutrients occurs. © 1998 John Wiley & Sons, Ltd.     

Impact of Ca2+ and Mg2+ on the Removal of F− by Magnesium Potassium Phosphate

This study investigates the influence of Ca²⁺ and Mg²⁺ on the removal of F^? by magnesium potassium phosphate (MPP) from water. The kinetic experiments reveal that the F^? concentration decreased from 3.5 to 3.31 mg L^?1 in a single (F^?) system and to 1.45 mg L^?1 in a ternary system (F^?, Ca²⁺, and Mg²⁺) after 1 min, respectively. Thus, the F^? removal efficiencies are found to increase by about 53% with the co‐active effect of Ca²⁺ and Mg²⁺ in the solution. Moreover, Ca²⁺ and Mg²⁺ are almost completely removed in the F^?, Ca²⁺, and Mg²⁺ system. According to the pseudo‐first‐order modeling, the rate constants k for F^?, Ca²⁺, and Mg²⁺ are 0.00348, 0.0106, and 0.0159 min^?1 respectively; thus, Mg²⁺ > Ca²⁺ > F^?. In the ternary system, the removal efficiencies are 53.29–66.03% for F^?, 99.99–100% for Ca²⁺, and 87.21–95.19% for Mg²⁺ with initial pH 5–10. The removal efficiencies of F^? increases with increases in initial concentrations of F^?, Ca²⁺, and Mg²⁺. The removal of F^? is governed by two routes: 1) adsorption by electrostatic interactions and outer sphere surface complexation; 2) co‐precipitation with Ca₃(PO₄)₂, CaHPO₄, Mg₃(PO₄)₂, and Mg(OH)₂.     

High-pressure phase transformations in the system of MgSiO3-FeSiO3

Two synthetic pyroxenes (FeSiO₃, MgSiO₃) and five natural pyroxenes with compositions of about Fs₈₀En₂₀, Fs₆₀En₄₀, Fs₅₀En₅₀, Fs₄₀En₆₀, and Fs₂₀En₈₀ have been subjected to pressures up to250 ± 50kbars at a temperature of about1500 ± 200°C in a diamond anvil cell heated by an infrared laser beam. After quenching and unloading X-ray data analysis indicates that (1) those with Mg less than 50% undergo the following reactions: 2(Mg,Fe)SiO₃ (pyroxene) → (Mg,Fe)₂SiO₄ (spinel) + SiO₂ (stishovite) → 2(Mg,Fe)O (magnesiowu¨stite) + SiO₂ (stishovite) with increase of pressure, and (2) those with Mg higher than 60%, undergo the following reactions: 2(Mg,Fe)SiO₃ (pyroxene) → (Mg,Fe)₂SiO₄ (spinel) + SiO₂ (stishovite) → 2(Mg,Fe)SiO₃ (hexagonal phase) → 2(Mg,Fe)O (magnesiowu¨stite) + SiO₂ (stishovite) with increase of pressure.     

Fe-Mg interdiffusion in magnesiowüstite up to 35 GPa

Fe-Mg interdiffusivities in (Fe,Mg)O magnesiowüstite have been measured in experiments conducted at pressures of 7-35 GPa and temperatures of 1573-1973 K using a Kawai-type high-pressure apparatus. The diffusion profiles were measured across the interface between MgO and (Fe_0.5,Mg_0.5)O samples by electron microprobe analysis, and the Fe-Mg interdiffusivities were determined as D_Fe-Mg=D₀exp{−E*(1+PV*_Mg/E*_Mg)/RT}, where D₀=4.1(^+16.1_−3.3)×10⁻⁷ m²/s, E*=(1−C_Mg)E*_Fe+C_MgE*_Mg (activation energy for the concentration of Mg, where E*_Fe=113(±74) kJ/mol and E*_Mg=226(±32) kJ/mol), the activation volume V*_Mg=1.8(±1.2)×10⁻⁶ m³/mol. By extrapolating these results to the P-T conditions of the core-mantle boundary, we conclude that the interdiffusivity of Fe-Mg in magnesiowüstite at the bottom of the lower mantle is at least one order of magnitude larger than that at the top of the lower mantle.     

Oxygen self-diffusion in forsterite: Implications for the high-temperature creep mechanism

The diffusivity of¹⁸O in forsterite Mg₂SiO₄ has been measured in the temperature range 1150–1600°C. The activation energy of oxygen self-diffusion in this silicate is found to equal0.32 ± 0.04MJ/mol(77 ± 10kcal/mol), and there is no dependence of the diffusivity upon the oxygen partial pressure surrounding the samples. The diffusion profiles were analysed either with an ion probe or by means of the¹⁸O(p, α)¹⁵N nuclear reaction. The latter method made use of a resonance in the nuclear cross-section in the case of diffusion profiles shorter than 100 nm (1000Å); for diffusion profiles up to 4 μm the same reaction was used, but in a non-resonant mode. New data on creep in forsterite and natural olivine are also given, including the influence of the oxygen partial pressurep_O2 which is zero for forsterite and proportional to(p_O2)¹⁶ for natural olivine. From this set of data we infer the possible relationship between diffusion and creep for these materials. This relationship may be more complicated than that predicted by simple climb mechanism.     

Ni2SiO4-enthalphy of the olivine-spinel transition by solution calorimetry at 713°C

The high pressure spinel polymorph of Ni₂SiO₄ persists metastably at 713°C and atmospheric pressure. The enthalpy of the olivine-spinel transition was obtained by measuring the heats of solution of both polymorphs in a molten oxide solvent, 2PbO · B₂O₃, at that temperature. For Ni₂SiO₄(ol)→Ni₂SiO₄, ΔH₉₈₆⁰ = +1.4 ± 0.7kcal/mol. The heat content increments, H₉₈₆ ? H₂₉₇, were found to be: olivine, 25.73 ± 0.42kcal/mol, and spinel, 25.39 ± 0.20kcal/mol. The measured enthalpy of the transformation is consistent with the low slope of the phase boundary, ?P/?T = ～ 12b/deg, observed by Akimoto and others. The entropy of the olivine-spinel transition in Ni₂SiO₄ is accordingly about a factor of three smaller in magnitude (ΔS = ～ ?1cal/deg mol) than that for Co₂SiO₄,Fe₂SiO₄,Mg₂SiO₄or Mg₂GeO₄ (ΔS = ?3to?3.5cal/deg mol).     

Disproportionation of spinels to mixed oxides: significance of cation configuration and implications for the mantle

Experimental study of the phase boundary for the disproportionation of the inverse spinel Mg₂SnO₄ into its isochemical mixed oxides indicates a slope dP/dT = 40 ± 10bars/°K. This positive slope is consistent with the large entropies of inverse (relative to normal) spinels predicted from high-temperature entropy-molar volume systematics. Thermodynamic data do not support the existence of a distinctly negative slope for the proposed disproportionation of Mg₂SiO₄ normal spinel. Evidence from X-ray and phase equilibria studies suggests the possibility that Si⁴⁺, Mg²⁺, and Fe²⁺ share the octahedral sites in silicate spinels under mantle conditions. The consequences of this partial inverse character are a positive slope for the postulated spinel-mixed oxide phase boundary near 650 km depth, removal of a widely accepted constraint on mantle-wide convection, and anomalous elasticity-density behaviour within the transition zone.