Electron paramagnetic resonance and ENDOR studies of Cr3+ - Al3+ pairs in forsterite

The electron paramagnetic resonance (EPR) spectrum of Cr³⁺ in synthetic crystals of forsterite consists primarily of lines of Cr³⁺ “isolated” at the M1 and M2 positions in a “perfect” crystal environment without local charge compensation. In addition it shows two nonequivalent superhyperfine-split sextets with different intensities which are due to strong interaction of the Cr³⁺ electron spin S (S=3/2) with an adjacent nuclear spin I(I=5/2). Electron nuclear double resonance (ENDOR) experiments revealed that the sextets result from Cr³⁺ (M1) - Al³⁺ and Cr³⁺ (M2) - Al³⁺ pairs, Al being located at adjacent tetrahedral Si sites. The g, D, A, and A′ tensor components of the Cr³⁺ - Al³⁺ pairs have been determined at room temperature. The values of the pairs are distinct although they are not very different from the corresponding data of “isolated” Cr³⁺. From the intensities of the EPR spectra the relative concentration of the Cr³⁺ - Al³⁺ pairs with respect to “isolated” Cr³⁺ has been estimated.     

An EPR study of Mn2+ and Fe3+ in dolomites

Electron paramagnetic resonance (EPR) measurements on dolomites from 9 different localities revealed contents of Mn²⁺ on two axial sites in all of them. The center with largerzero-field splitting (ZFS) was always present in much higher concentrations, except for a sample from Oberdorf it amounted to 95 percent or more of the total. This dolomite was the only one with a considerable content of Fe³⁺ on one axial site, almost certainly substituting for Mg²⁺. With X-ray irradiation the concentration of Fe³⁺ increased by about 30 percent showing that at least some of the divalent iron also substitutes for Mg. The ZFSs for Fe³⁺ and Mn²⁺ with larger ZFS increase with decreasing temperature in the same manner. The previous assignment of this Mn²⁺ to Mg sites is thus confirmed. An almost regular increase of the trigonal distortions at the divalent ions in different carbonates with increasing ionic radius is indicated by their crystal structure data. The very small ZFS for Mn²⁺ on Ca sites in dolomite must thus result from a strong local relaxation in the direction of a more regular octahedral arrangement. It is difficult to explain the different distribution ratios of Mn²⁺ on Ca and Mg sites with differences in growth and/or annealing temperatures alone. Thus different supply of Mg²⁺ and Ca²⁺ in the growth solutions may also contribute.     

Hyperfine interactions and spin transfer between Cr3+ and Al3+ in a synthetic single crystal of Mg2SiO4. A theoretical approach to the interpretation of EPR data

The occurrence of Cr-Al pairs in Mg₂SiO₄ has been detected by EPR spectroscopy. In the case where Cr³⁺ replaces Mg at the M2 position three different neighboring Si sites may be substituted by Al³⁺, which should yield different superhyperfine interactions. A new spectrum is presented which shows the presence of two of these possible pair configurations. An assignment of the spectral features to a specific Cr-Al pair with Cr at M2 from the experimental data alone was not possible, therefore, MSX _α cluster calculations have been performed from which the differences in the superhyperfine interaction for the various pair configurations could be obtained. Best agreement with the data of the Cr³⁺(M2)-Al pair exhibiting the most intense group of lines in the EPR spectrum was obtained for the situation where Al³⁺ is at the Si position with the shortest distance to M2. The second observed Cr³⁺(M2)-Al pair, which is significantly weaker in intensity, could not yet be assigned.     

Thermodynamics of multicomponent pyroxenes: III. Calibration of Fe2+(Mg)-1, TiAl2(MgSi2)-1, TiFe 2 3+ (MgSi2)-1, AlFe3+(MgSi)-1, NaAl(CaMg)-1, Al2(MgSi)-1 and Ca(Mg)-1 exchange reactions between pyroxenes and silicate melts

A thermodynamic model for the Gibbs free energy of igneous pyroxenes with the general formula [Na, Ca, Fe²⁺, Mg]^M2[Fe²⁺, Mg, Ti, Al, Fe³⁺]^M1[Al, Fe³⁺, Si]^TetSiO₆ is calibrated from experimentally determined compositions of coexisting pyroxene and silicate melt. The model is based upon the general formulation, and relies upon the calibration of the “quadrilateral” subsystem, previously published by the present authors. The calibration database of pyroxene-liquid equilibria spans a broad spectrum of temperature, pressure and oxygen fugacity conditions, ranging from 1000°–1600°C, 0.001–30 kbar and iron-wüstite to air. Chemical potentials of endmember pyroxene components as well as exchange potentials between pyroxenes and coexisting liquids are defined utilizing the present authors' thermodynamic melt model. Model parameters are extracted from these relations by regression analysis. The resulting model and derivative endmember properties are internally consistent with an existing standard state thermodynamic database. The success of the model and its applicability to igneous petrogenesis are demonstrated by comparing calculated and experimentally determined liquidus compositions, temperatures and symmetry states for pyroxenes crystallizing from a variety of silicate melts, ranging in composition from tholeiites and angrites through rhyolites to potash ankaratrites.     

The distribution of Fe2+ and Fe3+ in iron-bearing tourmalines: A Mössbauer study

Summary Room-temperature Mössbauer spectra of five iron-bearing tourmalines were measured and analyzed. The Fe²⁺/Fe³⁺ ratio and the iron occupancy of the Y and Z positions could be assigned to all samples, with the help of two previously well characterized samples, from Mexico and Madagascar. Ferric or ferrous ions or both partially occupy the Z as well as the Y octahedra. This fact of observation is interpreted as the chemical response, during crystal growth, to the requirement of size matching for the edge-sharing Y and Z oxygen octahedra. It accounts for the inexistence of solid solution between the Mg and (Li, Al) tourmalines.

---

Die Verteilung von Fe²⁺ und Fe³⁺ in eisenhaltigen Turmalinen: Eine Mössbauer-Untersuchung

Zusammenfassung Mössbauer-Spektren von fünf eisenhaltigen Turmalinen wurden gemessen und analysiert. Das Verhältnis Fe²⁺/Fe³⁺ und die Eisenverteilung konnten mit Hilfe von zwei gut identifizierten Turmalin-Kristallen von Mexiko und Madagascar für die Y-und Z-Lagen aller Exemplare bestimmt werden. Zweiwertiges sowie dreiwertiges Eisen findet sich sowohl in der Z-als auch in der Y-Lage. Da sich die Y-und Z-Oktaeder in einer gemeinsamen Kante treffen, wird diese Beobachtung als chemische Antwort des Kristalles auf die erforderte Größenanpassung der Y-und Z-Oktaeder während seines Wachstums erklärt. Die Abwesenheit der festen Lösung zwischen Dravit und Elbait kann somit erklärt werden.

---

---

With 4 Figures     

尾矿的砷污染试验研究及治理措施

金矿尾矿是砷污染的重要来源,不同冶炼工艺形成的尾矿中砷的赋存状态有很大差异,不同的砷形态会直接影响其在环境中的迁移转化行为,确认砷的存在形态是修复砷污染场地研究中的重要内容。本文针对浮选法和生物氧化法两种不同浸金方法的尾矿,采用pH计和氧化还原电位自动测定仪测量尾矿的物理性质（pH、氧化还原电位）,采用X射线衍射和X射线荧光光谱分析尾矿的矿物物相和主要成分,电子探针分析砷的存在形态。表征结果表明,浮选法尾矿和生物氧化法尾矿的物理性质都会因堆置环境的变化而变化,其Eh、pH和零电荷点都大致相同,pH≈8.5,样品零电荷点（pH_PZC）大约为8.5。由于两种尾矿都属于碱性环境,因此在修复方法的选择上也受到限制,如只适合在酸性条件下进行的电动修复法就不适用于这两种尾矿,修复试剂的选择也以碱性物质和铁的氧化物（氢氧化物）为主。浮选法尾矿的主要矿物类型为石英、白云石和黏土矿物,化学主要成分是Si和Al,尾矿呈灰白色,其中砷的含量约为754 μg/g左右,主要以毒砂（FeAsS）形式存在;而生物氧化法尾矿的主要矿物类型为石膏,化学主要成分是Fe、Ca和S,尾矿呈红棕色,其中砷的含量约为26350 μg/g,主要以毒砂（FeAsS）和五氧化二砷（As₂O₅）形式存在。两种浸金工艺每一道工序的不同,都会造成两种尾矿在矿物相、主要成分和其中砷存在形态上的差异。因此在进行尾矿原位修复工作时,应考虑不同浸金方法对尾矿堆置环境和砷存在形态的影响,从而选择出一种更加合适、廉价、高效的修复方法和试剂。

     

Electronic absorption and luminescence spectroscopic studies of kyanite single crystals: differentiation between excitation of FeTi charge transfer and Cr3+ dd transitions

A selected set of five different kyanite samples was analysed by electron microprobe and found to contain chromium between <0.001 and 0.055 per formula unit (pfu). Polarized electronic absorption spectroscopy on oriented single crystals, R₁, R₂-sharp line luminescence and spectra of excitation of λ₃- and λ₄-components of R₁-line of Cr³⁺-emission had the following results: (1) The Fe²⁺–Ti⁴⁺ charge transfer in c-parallel chains of edge connected M(1) and M(2) octahedra shows up in the electronic absorption spectra as an almost exclusively c(||Z′)-polarized, very strong and broad band at 16000 cm⁻¹ if <, in this case the only band in the spectrum, and at an invariably lower energy of 15400 cm⁻¹ in crystals with  ≥ . The energy difference is explained by an expansion of the O_f–O_k, and O_b–O_m edges, by which the M(1) and M(2) octahedra are interconnected (Burnham 1963), when Cr³⁺ substitutes for Al compared to the chromium-free case. (2) The Cr³⁺ is proven in two greatly differing crystal fields a and b, giving rise to two sets of bands, derived from the well known dd transitions of Cr^{3+ 4}A_2g→⁴T_2g(F)(I), →⁴T_1g(F)(II), and →⁴T_1g(P)(III). Band energies in the two sets a and b, as obtained by absorption, A, and excitation, E, agree well: I: 17300(a, A), 17200(a, E), 16000(b, A), 16200(b, E); II: 24800(a, A), 24400(a, E); 22300(b, A), 22200(b, E); III: 28800(b,A) cm⁻¹. Evaluation of crystal field parameters from the bands in the electronic spectra yield Dq(a)=1730 cm⁻¹, Dq(b)=1600 cm⁻¹, B(a)=790 cm⁻¹, B(b)=620 cm⁻¹ (errors ca. ±10 cm⁻¹), again in agreement with values extracted from the λ³, λ⁴ excitation spectra. The CF-values of set a are close to those typical of Cr³⁺ substituting for Al in octahedra of other silicate minerals without constitutional OH⁻ as for sapphirine, mantle garnets or beryl, and are, therefore, interpreted as caused by Cr³⁺ substituting for Al in some or all of the M(1) to M(4) octaheda of the kyanite structure, which are crystallographically different but close in their mean Al–O distances, ranging from 1.896 to 1.919 A (Burnham 1963), and slight degrees of distortion. Hence, band set a originates from substitutive Cr³⁺ in the kyanite structural matrix. The CF-data of Cr³⁺ type b, expecially B, resemble those of Cr³⁺ in oxides, especially of corundum type solid solutions or eskolaite. This may be interpreted by the assumption that a fraction of the total chromium contents might be allocated in a precursor of a corundum type exsolution. Received: 3 January 1997 / Revised, accepted: 2 May 1997     

Investigation of time-dependent reactions of H+ ions with variable and constant charge soils: a comparative study

A sampling-separation method and a dynamic monitoring method were used to investigate the time-dependent reactions of H⁺ ions with two contrasting types of soil, variable charge soils (VCS) and constant charge soils (CCS), by directly evaluating H⁺ ion consumption and other relevant consequences. The results for both CCS and VCS show that H⁺ ion consumption, increase in positive surface charge and increase in soluble Al are all characterized by a rapid step followed by a slow one. The higher the content of free Fe oxides in the soil, the larger the increase in positive surface charge and in H⁺ ion consumption in the initial rapid step. This is due mainly to protonation on external surfaces. The gradual increase in positive surface charge in the slow step for the 3 VCSs is a result of H⁺ ion diffusion to the reactive sites of Fe–OH on internal surfaces. The very low content of free Fe oxides on internal surfaces of the 2 CCSs render a negligible increase in positive surface charge in the slow step. For the 3 VCSs, the gradual consumption of H⁺ ions in the slow process is the result of protonation, Al dissolution and/or transformation into exchangeable acidity. For the 2 CCSs, however, the gradual consumption is mainly the result of Al dissolution and/or transformation into exchangeable acidity. The time-dependent Al dissolution from both VCS and CCS is influenced by several factors such as mineral components, solubility and dissolution rates of the soils, and H⁺ ion concentration in soil suspensions.     

Use of ultrasonically modulated electron resonance to study S-state ions in mineral crystals: Mn2+, Fe3+, in tremolite

The use of ultrasonically modulated electron resonance (UMER) to study S-state ions in substitutional sites of mineral single crystals is discussed. Mn²⁺ and Fe³⁺ in natural single crystals of tremolite are used as examples. Combined electron paramagnetic resonance (EPR) and UMER measurements establish almost certainly that Mn²⁺ enters predominantly into the distorted M4 sites occupied by Ca²⁺ in the ideal tremolite structure and only to a minor extent into the M1, M2 and M3 sites normally occupied by Mg²⁺. Fe³⁺ in tremolite gives rise to the well known high spin resonance with g _eff?4.3 but there is considerable uncertainty as to the site of the impurity ion.     

Correlation of spectroscopic properties and substitutional sites of Cr3+ in aluminosilicates: I. Kyanite

Using the superposition model in conjunction with our crystal field analysis package recently developed for 3d ions doped at arbitrary low symmetry sites in crystals, the energy levels and statevectors have been predicted within the whole 3d ³ configuration of Cr³⁺ at the four possible triclinic sites in kyanite (Al₂O₃∶SiO₂). The values of the ground state zero-field splitting for each of the four Al sites are evaluated. The splittings of the lower excited state ² E as well as the admixture of ⁴ T ₂ state into ² E have also been determined. The predicted results are compared with the available experimental data on the four possible, but so far not uniquely identified, substitutional Cr³⁺ sites in kyanite thus enabling correlation of the spectroscopic properties and substitutional sites.     

Biosorption of Cr(III) and Cr(VI) species from aqueous solution by Cabomba caroliniana: kinetic and equilibrium study

This study reports the potential ability of non-living biomass of Cabomba caroliniana for biosorption of Cr(III) and Cr(VI) from aqueous solutions. Effects of contact time, biosorbent dosage, pH of the medium, initial concentration of metal ion and protonation of the biosorbent on heavy metal–biosorbent interactions were studied through batch sorption experiments. Cr(III) was sorbed more rapidly than Cr(VI) and the pH of the medium significantly affected the extent of biosorption of the two metal species differently. Surface titrations showed that the surface of the biosorbent is positively charged at low pH while it is negatively charged at pH higher than 4.0. Protonation of the biosorbent increased its capacity for removal of Cr(III), while decreasing that of Cr(VI). FT-IR spectra of the biosorbent confirmed the involvement of –OH groups on the biosorbent surface in the chromium removal process. Kinetic and equilibrium data showed that the sorption process of each chromium species followed pseudo second-order kinetic model and both Langmuir and Freundlich isothermal models. A possible mechanism for the biosorption of chromium species by non-living C. caroliniana is suggested.     

The Hydrolysis of Al(III) in NaCl solutions: A Model for M(II), M(III), and M(IV) Ions

Understanding the identity and stability of the hydrolysis products of metals is required in order to predict their behavior in natural aquatic systems. Despite this need, the hydrolysis constants of many metals are only known over a limited range of temperature and ionic strengths. In this paper, we show that the hydrolysis constants of 31 metals [i.e. Mn(II), Cr(III), U(IV), Pu(IV)] are nearly linearly related to the values for Al(III) over a wide range of temperatures and ionic strengths. These linear correlations allow one to make reasonable estimates for the hydrolysis constants of +2, +3, and +4 metals from 0 to 300°C in dilute solutions and 0 to 100°C to 5 m in NaCl solutions. These correlations in pure water are related to the differences between the free energies of the free ion and complexes being almost equal $$ \Updelta {\text{G}}^\circ \left( {{\text{Al}}^{3 + } } \right) - \Updelta {\text{G}}^\circ \left( {{\text{Al}}\left( {\text{OH}} \right)_{j}^{{\left( {3 - j} \right)}} } \right) \cong \Updelta {\text{G}}^\circ \left( {{\text{M}}^{n + } } \right) - \Updelta {\text{G}}^\circ \left( {{\text{M}}\left( {\text{OH}} \right)_{j}^{{\left( {n - j} \right)}} } \right) $$ The correlation at higher temperatures is a result of a similar relationship between the enthalpies of the free ions and complexes $$ \Updelta {\text{H}}^\circ \left( {{\text{Al}}^{3 + } } \right) - \Updelta {\text{H}}^\circ \left( {{\text{Al}}\left( {\text{OH}} \right)_{j}^{3 - j} } \right) \cong \Updelta {\text{H}}^\circ \left( {{\text{M}}^{n + } } \right) - \Updelta {\text{H}}^\circ \left( {{\text{M}}\left( {\text{OH}} \right)_{j}^{n - j} } \right) $$ The correlations at higher ionic strengths are the result of the ratio of the activity coefficients for Al(III) being almost equal to that of the metal. $$ \gamma \left( {{\text{M}}^{n + } } \right)/\gamma \left( {{\text{M}}\left( {\text{OH}} \right)_{j}^{n - j} } \right) \cong \gamma \left( {{\text{Al}}^{3 + } } \right)/\gamma \left( {{\text{Al}}\left( {\text{OH}} \right)_{j}^{3 - j} } \right) $$ The results of this study should be useful in examining the speciation of metals as a function of pH in natural waters (e.g. hydrothermal fresh waters and NaCl brines).     

Correlation of spectroscopic properties and substitutional sites of Cr3+ in aluminosilicates: II. Andalusite and sillimanite

This paper is an extension of the earlier one dealing with kyanite in which the best fitting value of the oxygen ligand distance for Cr³⁺ is adopted to study the spectroscopic properties of Cr³⁺ ions doped at the two possible Al sites in the other two polymorphs of the aluminosilicate group (Al₂O₃ · SiO₂), namely, andalusite and sillimanite. The superposition model and the crystal field analysis package recently developed for 3d ions doped at arbitrary low symmetry sites in crystals are used to predict energy levels and statevectors within the whole 3d ³ configuration. Then the values of the ground state zerofield splitting for Cr³⁺ ions at each Al sites in the two crystals are obtained. The splittings of the lower excited states ² E and ⁴ T ₂ as well as the admixture of ⁴ T ₂ into ² E have also been predicted. Comparison of our results with the available experimental data enable us to correlate the optical and EPR Spectroscopic properties with the substitutional Cr³⁺ sites. The conclusion is that in andalusite and sillimanite only the Al sites with nearly-octahedral six-fold coordination seem to be occupied by Cr³⁺ ions.     

Synthesis and stability of deerite,Fe 12 2+ Fe 6 3+ [Si12O40](OH)10, and Fe3+⇋Al3+ substitutions at 15–28 kb

Deerite, Fe ₁₂ ²⁺ Fe ₆ ³⁺ [Si₁₂O₄₀](OH)₁₀, thus far known from ten localities in glaucophane schist terranes, was synthesized at water pressures of 20–25 kb and temperatures of 550–600 °C under the of the Ni/NiO buffer. The X-ray powder diagram, lattice constants and infrared spectrum of the synthetic phase are closely similar to those of the natural mineral. A solid solution series extends from this ferri-deerite end member to some 20 mole % of a hypothetical alumino-deerite, Fe ₁₂ ²⁺ Al ₆ ³⁺ [Si₁₂O₄₀](OH)₁₀. The upper temperature breakdown of ferri-deerite to the assemblage ferrosilite +magnetite+quartz+water occurs at about 490 °C at 15 kb, and 610 °C at 25 kb fluid pressure for the of the Ni/NiO buffer. Extrapolation of these data to lower water pressures indicates that deerite can be a stable mineral only in very low-temperature, high-pressure environments.     

Experimental investigation of the metamorphism of siliceous dolomites III. Equilibrium data for the reaction: 1 Tremolite + 11 dolomite ⇌ 8 forsterite + 13 calcite + 9 CO2 +1H2O for the total pressures of 3,000 and 5,000 bars

The temperature-X _{CO 2}-equilibrium data for the reaction 1 tremolite + 11 dolomite 8 forsterite + 13 calcite + 9 CO₂ +1H₂O have been determined at total pressures (P _{CO 2} + P _H2_O) of 3,000 and 5,000 bars. The results are shown in Figure 2 along with the data for the total pressure of 1,000 bars (Metz, 1967).The MgCO₃ contents of the magnesian-calcites formed during the experiments agree very well with the calcite-dolomite-solvus which can be recalculated from Equation (1) and the activity coefficients for MgCO₃ in magnesiancalcite as given by Gordon and Greenwood (1970).If the T-X _{CO 2}-equilibrium data are calculated from the equilibrium constant as given by Skippen (1974), assuming ideal mixing of CO₂ and H₂O, good agreement is achieved for the total pressure of 1,000 bars (see Figs. 4 and 5). At a total pressure of 3,000 bars, however, the calculated equilibrium temperatures are about 40 ° C below the experimentally determined values (see Fig. 6). This difference increases up to 70 ° C for a total pressure of 5,000 bars (see Fig. 7).From the experimentally determined equilibrium conditions of the assemblage: tremolite + dolomite + forsterite + magnesian calcite (see Fig. 8) the pressure of metamorphism can be estimated if the temperature is determined by the MgCO₃-content of the magnesian-calcite from the calcite-dolomite solvus. However, when using the data of Figure 8, attention has to be drawn to the limiting condition of X _{CO 2}0.2.Simplified reaction equation not considering solid solution in the carbonates     

High-pressure transitions and thermochemistry of MGeO3 (M=Mg, Zn and Sr) and Sr-silicates: systematics in enthalpies of formation of A2+B4+O3 perovskites

Phase transitions in MgGeO₃ and ZnGeO₃ were examined up to 26 GPa and 2,073 K to determine ilmenite–perovskite transition boundaries. In both systems, the perovskite phases were converted to lithium niobate structure on release of pressure. The ilmenite–perovskite boundaries have negative slopes and are expressed as P(GPa)=38.4–0.0082T(K) and P(GPa)=27.4−0.0032T(K), respectively, for MgGeO₃ and ZnGeO₃. Enthalpies of SrGeO₃ polymorphs were measured by high-temperature calorimetry. The enthalpies of SrGeO₃ pseudowollasonite–walstromite and walstromite–perovskite transitions at 298 K were determined to be 6.0±8.6 and 48.9±5.8 kJ/mol, respectively. The calculated transition boundaries of SrGeO₃, using the measured enthalpy data, were consistent with the boundaries determined by previous high-pressure experiments. Enthalpy of formation (ΔH _f°) of SrGeO₃ perovskite from the constituent oxides at 298 K was determined to be −73.6±5.6 kJ/mol by calorimetric measurements. Thermodynamic analysis of the ilmenite–perovskite transition boundaries in MgGeO₃ and ZnGeO₃ and the boundary of formation of SrSiO₃ perovskite provided transition enthalpies that were used to estimate enthalpies of formation of the perovskites. The ΔH _f° of MgGeO₃, ZnGeO₃ and SrSiO₃ perovskites from constituent oxides were 10.2±4.5, 33.8±7.2 and −3.0±2.2 kJ/mol, respectively. The present data on enthalpies of formation of the above high-pressure perovskites were combined with published data for A²⁺B⁴⁺O₃ perovskites stable at both atmospheric and high pressures to explore the relationship between ΔH _f° and ionic radii of eightfold coordinated A²⁺ (R _A) and sixfold coordinated B⁴⁺ (R _B) cations. The results show that enthalpy of formation of A²⁺B⁴⁺O₃ perovskite increases with decreasing R _A and R _B. The relationship between the enthalpy of formation and tolerance factor ( R _o: O²⁻ radius) is not straightforward; however, a linear relationship was found between the enthalpy of formation and the sum of squares of deviations of A²⁺ and B⁴⁺ radii from ideal sizes in the perovskite structure. A diagram showing enthalpy of formation of perovskite as a function of A²⁺ and B⁴⁺ radii indicates a systematic change with equienthalpy curves. These relationships of ΔH _f° with R _A and R _B can be used to estimate enthalpies of formation of perovskites, which have not yet been synthesized.     

The reaction MgCr2O4 + SiO2 = Cr2O3 + MgSiO3 and the free energy of formation of magnesiochromite (MgCr2O4)

Low-temperature heat capacity measurements for MgCr₂O₄ have only been performed down to 52 K, and the commonly quoted third-law entropy at 298 K (106 J K⁻¹ mol⁻¹) was obtained by empirical extrapolation of these measurements to 0 K without considering the magnetic or electronic ordering contributions to the entropy. Subsequent magnetic measurements at low temperature reveal that the Néel temperature, at which magnetic ordering of the Cr³⁺ ions in MgCr₂O₄ occurs, is at ∼15 K. Hence a substantial contribution to the entropy of MgCr₂O₄ has been missed. We have determined the position of the near-univariant reaction MgCr₂O₄+SiO₂=MgSiO₃+Cr₂O₃. The reaction, which has a small positive slope in P-T space, has been bracketed at 100 K intervals between 1273 and 1773 K by reversal experiments. The reaction is extremely sluggish, and lengthy run times with a flux (H₂O, BaO-B₂O₃ or K₂O-B₂O₃) are needed to produce tight reversal brackets. The results, combined with assessed thermodynamic data for Cr₂O₃, MgSiO₃ and SiO₂, give the entropy and enthalpy of formation of MgCr₂O₄ spinel. As expected, our experimental results are not in good agreement with the presently available thermodynamic data. We obtain Δ _f H ^∘ ₂₉₈=−1759.2±1.5 kJ mol⁻¹ and S ^∘ ₂₉₈=122.1±1.0 J K⁻¹ mol⁻¹ for MgCr₂O₄. This entropy is some 16 J K⁻¹ mol⁻¹ more than the calorimetrically determined value, and implies a value for the magnetic entropy of MgCr₂O₄ consistent with an effective spin quantum number (S') for Cr³⁺ of 1/2 rather than the theoretical 3/2, indicating, as in other spinels, spin quenching. Received: 9 May 1997 / Accepted: 28 July 1997