K-H3OFe3(SO4)2(OH)6铁矾的XRD研究

根据X射线衍射(XRD)分析发现: A Fe₃(SO₄)₂(OH)₆(A=K⁺、H₃O⁺)系列铁钒的XRD数据十分相近,难以用XRD区别,需通过能谱(EDS)辅助分析,才能区分此类铁矾。另外,此类铁矾的003和107面网间距d随K⁺含量增大而增大,且呈一元三次方程的关系;而033和220面网间距d随K⁺含量增大而减小,呈一元二次方程的关系。对该现象从铁矾晶体结构方面进行解释:K⁺、H₃O⁺离子位于较大空隙中,且沿着Z轴方向排列,当K⁺、H₃O⁺离子之间相互替换时,会导致该铁矾晶体结构在Z轴方向有较明显的变化。     

Dissolution of jarosite [KFe3(SO4)2(OH)6] at pH 2 and 8: Insights from batch experiments and computational modelling

Jarosite [KFe₃(SO₄)₂(OH)₆] is a mineral that is common in acidic, sulphate-rich environments, such as acid sulphate soils derived from pyrite-bearing sediments, weathering zones of sulphide ore deposits and acid mine or acid rock drainage (ARD/AMD) sites. The structure of jarosite is based on linear tetrahedral-octahedral-tetrahedral (T-O-T) sheets, made up from slightly distorted FeO₆ octahedra and SO₄ tetrahedra. Batch dissolution experiments carried out on synthetic jarosite at pH 2, to mimic environments affected by ARD/AMD, and at pH 8, to simulate ARD/AMD environments recently remediated with slaked lime (Ca(OH)₂), suggest first order dissolution kinetics. Both dissolution reactions are incongruent, as revealed by non-ideal dissolution of the parent solids and, in the case of the pH 8 dissolution, because a secondary goethite precipitate forms on the surface of the dissolving jarosite grains. The pH 2 dissolution yields only aqueous K, Fe, and SO₄. Aqueous, residual solid, and computational modelling of the jarosite structure and surfaces using the GULP and MARVIN codes, respectively, show for the first time that there is selective dissolution of the A- and T-sites, which contain K and SO₄, respectively, relative to Fe, which is located deep within the T-O-T jarosite structure. These results have implications for the chemistry of ARD/AMD waters, and for understanding reaction pathways of ARD/AMD mineral dissolution.     

Free energy of formation for green rust sodium sulphate (NaFe6Fe3(OH)18(SO4)2(s))

In a recent study, sulphate-bearing green rust (GR_SO4) was shown to incorporate Na⁺ in its structure (NaFe^II₆Fe^III₃(OH)₁₈(SO₄)_2(s); GR_Na,SO4). The compound was synthesised by aerial oxidation of Fe(OH)_2(s) in the presence of NaOH. This paper reports on its free energy of formation .Freshly synthesised GR_Na,SO4 was titrated with 0.5 M H₂SO₄ in an inert atmosphere at 25 °C, producing dissolved Fe²⁺ and magnetite or goethite. Solution concentrations, PHREEQC and the MINTEQ database were used to calculate reaction constants for the reactions:

     

Solubility of Fe-ettringite (Ca6[Fe(OH)6]2(SO4)3 · 26H2O)

The solubility of Fe-ettringite (Ca₆[Fe(OH)₆]₂(SO₄)₃ · 26H₂O) was measured in a series of precipitation and dissolution experiments at 20 °C and at pH-values between 11.0 and 14.0 using synthesised material. A time-series study showed that equilibrium was reached within 180 days of ageing. After equilibrating, the solid phases were analysed by XRD and TGA while the aqueous solutions were analysed by ICP-OES (calcium, sulphur) and ICP-MS (iron). Fe-ettringite was found to be stable up to pH 13.0. At higher pH-values Fe-monosulphate (Ca₄[Fe(OH)₆]₂(SO₄) · 6H₂O) and Fe-monocarbonate (Ca₄[Fe(OH)₆]₂(CO₃) · 6H₂O) are formed. The solubilities of these hydrates at 25 °C are:      

Stable boron isotope fractionation between dissolved B(OH)3 and B(OH)4

The stable boron isotope ratio (¹¹B/¹⁰B) in marine carbonates is used as a paleo-pH recorder and is one of the most promising paleo-carbonate chemistry proxies. Understanding the thermodynamic basis of the proxy is of fundamental importance, including knowledge on the equilibrium fractionation factor between dissolved boric acid, B(OH)₃, and borate ion, B(OH)₄⁻ (, hereafter α_(B3-B4)). However, this factor has hitherto not been determined experimentally and a theoretically calculated value (Kakihana and Kotaka, 1977, hereafter KK77) has therefore been widely used. I examine the calculations underlying this value. Using the same spectroscopic data and methods as KK77, I calculate the same α_(B3−B4) = 1.0193 at 300 K. Unfortunately, it turns out that in general the result is sensitive to the experimentally determined vibrational frequencies and the theoretical methods used to calculate the molecular forces. Using analytical techniques and ab initio molecular orbital theory, the outcome for α_(B3-B4) varies between ∼1.020 and ∼1.050 at 300 K. However, several arguments suggest that α_(B3-B4) ? 1.030. Measured isotopic shifts in various ¹⁰B-, ²D-, and ¹⁸O-labeled isotopomers do not provide a constraint on stable boron isotope fractionation. I conclude that in order to anchor the fundamentals of the boron pH proxy, experimental work is required. The critics of the boron pH proxy should note, however, that uncertainties in α_(B3-B4) do not bias pH reconstructions provided that organism-specific calibrations are used.     

Acid dissociation mechanisms of Si(OH)4 and Al(H2O)6 in aqueous solution

Silicic acid and the hexa-aqua of Al³⁺ are fundamental model aqueous species of chemical importance in nature. In order to investigate their hydroxyl dissociation mechanisms, Car-Parrinello molecular dynamics (CPMD) simulations were carried out, which allow treating the solutes and solvents on the same footing. The method of constraint was employed to trigger the reactions by taking coordination number as the reaction coordinate and the thermodynamic integration was used to obtain the free-energy profiles. The approximate transition states were located and the reactant and product states were also characterized. The free-energy changes of dissociation are found about 15.0 kcal/mol and 7.7 kcal/mol for silicic acid and Al-aqua, respectively. From the simulation results, the first pKas were calculated by using two approaches, which are based on the pristine thermodynamic relation and the RDF (radial distribution function)-free energy relation, respectively. Because of more uncertainties involved in the RDF way, it is suggested that the pristine way should be favored, which shows an error margin of 1 pKa unit. This study provides an encouraging basis for applying the present methodology to predict acidity constants of those groups that are difficult to measure experimentally.     

Thermochemistry of yavapaiite KFe(SO4)2: Formation and decomposition

Yavapaiite, KFe(SO₄)₂, is a rare mineral in nature, but its structure is considered as a reference for many synthetic compounds in the alum supergroup. Several authors mention the formation of yavapaiite by heating potassium jarosite above ca. 400°C. To understand the thermal decomposition of jarosite, thermodynamic data for phases in the K-Fe-S-O-(H) system, including yavapaiite, are needed. A synthetic sample of yavapaiite was characterized in this work by X-ray diffraction (XRD), scanning electron microscopy (SEM), Fourier transform infrared spectroscopy (FTIR), and thermal analysis. Based on X-ray diffraction pattern refinement, the unit cell dimensions for this sample were found to be a = 8.152 ± 0.001 Å, b = 5.151 ± 0.001 Å, c = 7.875 ± 0.001 Å, and β = 94.80°. Thermal decomposition indicates that the final breakdown of the yavapaiite structure takes place at 700°C (first major endothermic peak), but the decomposition starts earlier, around 500°C. The enthalpy of formation from the elements of yavapaiite, KFe(SO₄)₂, ΔH°_f = −2042.8 ± 6.2 kJ/mol, was determined by high-temperature oxide melt solution calorimetry. Using literature data for hematite, corundum, and Fe/Al sulfates, the standard entropy and Gibbs free energy of formation of yavapaiite at 25°C (298 K) were calculated as S°(yavapaiite) = 224.7 ± 2.0 J.mol⁻¹.K⁻¹ and ΔG°_f = −1818.8 ± 6.4 kJ/mol. The equilibrium decomposition curve for the reaction jarosite = yavapaiite + Fe₂O₃ + H₂O has been calculated, at pH₂O = 1 atm, the phase boundary lies at 219 ± 2°C.     

The kinetics of oxygen exchange between the GeO4Al12(OH)24(OH2)12(aq) molecule and aqueous solutions

We report rates of oxygen exchange with bulk solution for an aqueous complex, ^IVGeO₄Al₁₂(OH)₂₄(OH₂)₁₂⁸⁺(aq) (GeAl₁₂), that is similar in structure to both the ^IVAlO₄Al₁₂(OH)₂₄(OH₂)₁₂⁷⁺(aq) (Al₁₃) and ^IVGaO₄Al₁₂(OH)₂₄(OH₂)₁₂⁷⁺(aq) (GaAl₁₂) molecules studied previously. All of these molecules have ε-Keggin-like structures, but in the GeAl₁₂ molecule, occupancy of the central tetrahedral metal site by Ge(IV) results in a molecular charge of +8, rather than +7, as in the Al₁₃ and GaAl₁₂. Rates of exchange between oxygen sites in this molecule and bulk solution were measured over a temperature range of 274.5 to 289.5 K and 2.95 < pH < 4.58 using ¹⁷O-NMR.Apparent rate parameters for exchange of the bound water molecules (η-OH₂) are k_ex²⁹⁸ = 200 (±100) s⁻¹, ΔH^‡ = 46 (±8) kJ · mol⁻¹, and ΔS^‡ = −46 (±24) J · mol⁻¹ K⁻¹ and are similar to those we measured previously for the GaAl₁₂ and Al₁₃ complexes. In contrast to the Al₁₃ and GaAl₁₂ molecules, we observe a small but significant pH dependence on rates of solvolysis that is not yet fully constrained and that indicates a contribution from the partly deprotonated GeAl₁₂ species.The two topologically distinct μ₂-OH sites in the GeAl₁₂ molecule exchange at greatly differing rates. The more labile set of μ₂-OH sites in the GeAl₁₂ molecule exchange at a rate that is faster than can be measured by the ¹⁷O-NMR isotopic-equilibration technique. The second set of μ₂-OH sites have rate parameters of k_ex²⁹⁸ = 6.6 (±0.2) · 10⁻⁴ s⁻¹, ΔH^‡ = 82 (±2) kJ · mol⁻¹, and ΔS^‡ = −29 (±7) J · mol⁻¹ · K⁻¹, corresponding to exchanges ≈40 and ≈1550 times, respectively, more rapid than the less labile μ₂-OH sites in the Al₁₃ and GaAl₁₂ molecules. We find evidence of nearly first-order pH dependence on the rate of exchange of this μ₂-OH site with bulk solution for the GeAl₁₂ molecule, which contrasts with Al₁₃ and GaAl₁₂ molecules.     

Solid-liquid equilibria of Mg(OH)2(cr) and Mg2(OH)3Cl·4H2O(cr) in the system Mg-Na-H-OH-Cl-H2O at 25°C

The solubility of crystalline Mg(OH)₂(cr) was determined by measuring the equilibrium H⁺ concentration in water, 0.01-2.7 m MgCl₂, 0.1-5.6 m NaCl, and in mixtures of 0.5 and 5.0 m NaCl containing 0.01-0.05 m MgCl₂. In MgCl₂ solutions above 2 molal, magnesium hydroxide converted into hydrated magnesium oxychloride. The solid-liquid equilibrium of Mg₂(OH)₃Cl·4H₂O(cr) was studied in 2.1-5.2 m MgCl₂. Using known ion interaction Pitzer coefficients for the system Mg-Na-H-OH-Cl-H₂O (25°C), the following equilibrium constants at I = 0 are calculated:

     

Phosphoinnelite, Ba4Na3Ti3Si4O14(PO4,SO4)2(O,F)3, a new mineral species from peralkaline pegmatite of the Kovdor pluton, Kola Peninsula

Phosphoinnelite, an analogue of innelite with P > S, has been found in a peralkaline pegmatite vein crosscutting calcite carbonatite at the phlogopite deposit, Kovdor pluton, Kola Peninsula. Cancrinite (partly replaced with thomsonite-Ca), orthoclase, aegirine-augite, pectolite, magnesioarfvedsonite, golyshevite, and fluorapatite are associated minerals. Phosphoinnelite occurs as lath-shaped crystals up to 0.2 × 1 × 6 mm in size, which are combined typically in bunch-, sheaf-, and rosettelike segregations. The color is yellow-brown, with vitreous luster on crystal faces and greasy luster on broken surfaces. The mineral is transparent. The streak is pale yellowish. Phosphoinnelite is brittle, with perfect cleavage parallel to the {010} and good cleavage parallel to the {100}; the fracture is stepped. The Mohs hardness is 4.5 to 5. Density is 3.82 g/cm³ (meas.) and 3.92 g/cm³ (calc.). Phosphoinnelite is biaxial (+), α = 1.730, β = 1.745, and γ = 1.764, 2V (meas.) is close to 90°. Optical orientation is Z^c ∼ 5°. Chemical composition determined by electron microprobe is as follows (wt %): 6.06 Na₂O, 0.04 K₂O, 0.15 CaO, 0.99 SrO, 41.60 BaO, 0.64 MgO, 1.07 MnO, 1.55 Fe₂O₃, 0.27 Al₂O₃, 17.83 SiO₂, 16.88 TiO₂, 0.74 Nb₂O₅, 5.93 P₂O₅, 5.29 SO₃, 0.14 F, −O=F₂ = −0.06, total is 99.12. The empirical formula calculated on the basis of (Si,Al)₄O₁₄ is (Ba_3.59Sr_0.13K_0.01)_Σ3.73(Na_2.59Mg_0.21Ca_0.04)_Σ3.04(Ti_2.80Fe _0.26 ³⁺ Nb_0.07)_Σ3.13[(Si_3.93Al_0.07)_Σ4O₁₄(P_1.11S_0.87)_Σ1.98O_7.96](O_2.975F_0.10)_Σ3.075. The simplified formula is Ba₄Na₃Ti₃Si₄O₁₄(PO₄,SO₄)₂(O,F)₃. The mineral is triclinic, space group P or P1. The unit cell dimensions are a = 5.38, b = 7.10, c = 14.76 ?; α = 99.00°, β = 94.94°, γ = 90.14°; and V = 555 ?³, Z = 1. The strongest lines of the X-ray powder pattern [d, ? in (I)(hkl)] are: 14.5(100)(001), 3.455(40)(103), 3.382(35)(0 2), 2.921(35)(005), 2.810(40)(1 4), 2.683(90)(200, 01), 2.133(80)( 2), 2.059(40)(204, 1 3, 221), 1.772(30)(0 1, 1 7, 2 2, 2 3). The infrared spectrum is demonstrated. An admixture of P substituting S has been detected in the innelite samples from the Inagli pluton (South Yakutia, Russia). An innelite-phosphoinnelite series with a variable S/P ratio has been discovered. The type material of phosphoinnelite has been deposited at the Fersman Mineralogical Museum, Russian Academy of Sciences, Moscow. Original Russian Text ? I.V. Pekov, N.V. Chukanov, I.M. Kulikova, D.I. Belakovsky, 2006, published in Zapiski Rossiiskogo Mineralogicheskogo Obshchestva, 2006, No. 3, pp. 52–60. Considered and recommended by the Commission on New Minerals and Mineral Names, Russian Mineralogical Society, May 9, 2005. Approved by the Commission on New Minerals and Mineral Names, International Mineralogical Association, July 4, 2005 (proposal 2005-022).     

Microscope-photometric methods for non-destructive Fe2+ – Fe3+ determination in chloritoids (Fe2+, Mn2+, Mg)2 (Al,Fe3+)4 Si2O10(OH)4

Six natural chloritoid crystals with Fe²⁺ and Fe³⁺ contents ranging from 4.15 to 12.81 and from 0.411 to 0.849g-atoms/l, respectively, as determined by means of microprobe and Mössbauer techniques, served as reference material to develop non-destructive microscope-spectrophotometric methods for quantitative Fe²⁺ – Fe³⁺ determinations in chloritoids from unpolarized spectra of (001) platelets. Fe²⁺ concentrations in g-atom/l can be obtained from [ [Fe³⁺]=C₁xD₁/t where D₁ = log₁₀(I_0/I at 28,000 cm^-1 and t=crystal thickness in cm; C₁ is a conttant that may be influenced somewhat by experimental conditions and is found to be 0.002289 with the experimental set-up used in this study. Fe²⁺ concentrations in g-atom/l can be obtained from [Fe²⁺]=C₁xD₁/D₁-C₃ with D₂=log₁₀(I₀/I) at 16,300 cm^?1 and constants C₄ = 45.36 and C₅ = 3.540. Due to the uncertainties in absorbance measurements, D₁ and D₂ and the thickness measurements, the accuracies are ±0.05 and ±0.15 g-atom/l for [Fe³⁺] and [Fe²⁺], respectively. The determinations may be carried out on chloritoid grains in normal thin sections with an areal resolution of ～10 μm.     

Rates of oxygen exchange between the Al2O8Al28(OH)56(H2O)26(aq) (Al30) molecule and aqueous solution

Rates of steady exchange of oxygens between bulk solution and the largest known aluminum polyoxocation: Al₂O₈Al₂₈(OH)₅₆(H₂O)₂₆¹⁸⁺(aq) (Al₃₀) are reported at pH≈4.7 and 32-40°C. The Al₃₀ molecule is a useful model for geochemists because it is ≈2 nm in length, comparable to the smallest colloidal solids, and it has structural complexity greater than the surfaces of most aluminum (hydr)oxide minerals. The Al₃₀ molecule has 15 distinct hydroxyl sites and eight symmetrically distinct bound waters. Among the hydroxyl bridges are two sets of μ₃-OH, which are not present in any of the other aluminum polyoxocations that have yet been studied by NMR methods. Rates of isotopic equilibration of the μ₂-OH and μ₃-OH hydroxyls and bound water molecules fall within the same range as we have determined for other aluminum solutes, although it is impossible to determine rate laws for exchange at the large number of individual oxygen sites. After injection of ¹⁷O-enriched water, growth of the ¹⁷O-NMR peak near 37 ppm, which is assigned to μ₂-OH and μ₃-OH hydroxyl bridges, indicates that these bridges equilibrate within two weeks at temperatures near 35°C. The peak at +22 ppm in the ¹⁷O-NMR spectra, assigned to bound water molecules (η-OH₂), varies in width with temperature in a similar fashion as for other aluminum solutes, suggesting that most of the η-OH₂ sites exchange with bulk solution at rates that fall within the range observed for other aluminum complexes. Signal from one anomalous group of four η-OH₂ sites is not observed, indicating that these sites exchange at least a factor of ten more rapidly than the other η-OH₂ sites on the Al₃₀.     

Actinide crystal-liquid partitioning for clinopyroxene and Ca3(PO4)2

Using fission and alpha track radiography techniques, we have measured partition coefficients (D) for the actinide elements Th, U and Pu between diopsidic clinopyroxene, whitlockite [β-Ca₃ (PO₄)₂] and silicate liquid at 20kbar. Equilibrium partitioning at the crystal-liquid interface is assumed, and corrections for actinide zoning have been applied to the measured D values. Reproducibility for both actinide and minor element D values is carefully examined as a criterion for crystal-liquid interface equilibrium. The data are mostly compatible with interface equilibrium except for experiments at high cooling rates ( ? 30 deg/hr). Partition coefficients for Th/U/Pu of about 0.002/0.002/0.06 are measured for clinopyroxene and 1.2/0.5/3.4 for whitlockite. At an oxygen fugacity of 10^?8.5, Pu is much more readily incorporated into the crystalline phases than is U or Th because of the importance of trivalent Pu. The D_Pu(cpx) is similar to D(cpx) of the light rare earths supporting the concept of Pu/(rare earth) dating.     

Solubility of pyromorphite Pb5(PO4)3Cl-mimetite Pb5(AsO4)3Cl solid solution series

Pyromorphite Pb₅(PO₄)₃Cl and mimetite Pb₅(AsO₄)₃Cl are isostructural minerals with apatite. Due to their high environmental stability, they have gained considerable attention as metals sequestration agents in water treatment and contaminated soil remediation. Pyromorphite and mimetite can form a continuous solid solution series in near-Earth surface environments. Precipitation of the end members and intermediate members of the series is likely to occur in the areas where the cost-effective in situ immobilization reclamation method, based on phosphate amendments, is applied. In contrast to the widely studied thermodynamic parameters of pyromorphite and mimetite, knowledge of the thermodynamics of their solid solutions is sparse. To supplement the data, a number of compounds from the pyromorphite-mimetite series were synthesized at room temperature using a method to simulate the conditions in the near-Earth surface environments. Afterwards, batch dissolution and dissolution-recrystallization experiments of seven synthesized precipitates were conducted at 25 °C, pH = 2 and in a 0.05 M KNO₃ background electrolyte. The experiments were carried out for a period of 6 (dissolution) and 14 (dissolution-recrystallization) months. A plateau in the [Pb] evolution patterns was used to determine equilibrium. All seven dissolutions were congruent, and the ionic activity products (IAP) of the minerals from the pyromorphite-mimetite solid solution series were calculated based on the dissolution reaction: . The IAPs for pyromorphite and mimetite exhibit a significant difference in values over three orders of magnitude between approximately 10⁻⁷⁹ for pyromorphite and approximately 10⁻⁷⁶ for mimetite. The series appeared to be ideal, and Lippmann and Roozboom diagrams were used for better understanding of its thermodynamics. The results indicated a strong tendency of pyromorphite to partition into the solid phase in the series, which explains some of the naturally observed phenomena. The improvement of the lattice stability of the mimetite due to isostructural phosphate substitutions in anionic sites was observed. The thermodynamic data reported in this study supplement existing databases used in geochemical modeling.     

Residence times for protons bound to three oxygen sites in the AlO4Al12(OH)24(H2O)12 polyoxocation

Although, the kinetic reactivity of a mineral surface is determined, in part, by the rates of exchange of surface-bound oxygens and protons with bulk solution, there are no elementary rate data for minerals. However, such kinetic measurements can be made on dissolved polynuclear clusters, and here we report lifetimes for protons bound to three oxygen sites on the AlO₄Al₁₂(OH)₂₄(H₂O)₁₂⁷⁺ (Al₁₃) molecule, which is a model for aluminum-hydroxide solids in water. Proton lifetimes were measured using ¹H NMR at pH ∼ 5 in both aqueous and mixed solvents. The ¹H NMR peak for protons on bound waters (η-H₂O) lies near 8 ppm in a 2.5:1 mixture of H₂O/acetone-d₆ and broadens over the temperature range −20 to −5 °C. Extrapolated to 298 K, the lifetime of a proton on a η-H₂O is τ²⁹⁸ ∼ 0.0002 s, which is surprisingly close to the lifetime of an oxygen in the η-H₂O (∼0.0009 s), but in the same general range as lifetimes for protons on fully protonated monomer ions of trivalent metals (e.g., Al(H₂O)₆³⁺). The lifetime is reduced somewhat by acid addition, indicating that there is a contribution from the partly deprotonated Al₁₃ molecule in addition to the fully protonated Al₁₃ at self-buffered pH conditions. Proton lifetimes on the two distinct sets of hydroxyls bridging two Al(III) (μ₂-OH) differ substantially and are much shorter than the lifetime of an oxygen at these sites. The average lifetimes for hydroxyl protons were measured in a 2:1 mixture of H₂O/dmso-d₆ over the temperature range 3.7-95.2 °C. The lifetime of a hydrogen on one of the μ₂-OH was also measured in D₂O. The τ²⁹⁸ values are ∼0.013 and ∼0.2 s in the H₂O/dmso-d₆ solution and the τ²⁹⁸ value for the μ₂-OH detectable in D₂O is τ²⁹⁸ ∼ 0.013 s. The ¹H NMR peak for the more reactive μ₂-OH broadens slightly with acid addition, indicating a contribution from an exchange pathway that involves a proton or hydronium ion. These data indicate that surface protons on minerals will equilibrate with near-surface waters on the diffusional time scale.     

La sénégalite,Al2 (PO4) (OH)3 · H2O,un nouveau minéral

Senegalite is orthorhombic, mm2, a:b:c:=1.296:1:1.007; $\text{a}{}_0\text{=9.673,}\text{b}{}_0\text{=7.596,}\text{c}{}_0\text{=7.668}\text{A}\text{?}$, Z=4, G_calc=2551; space group Pna2. The strongest lines in the powder pattern are: 5.41(7); 4.089(9); 3.834(10); 3.610(8); 2.990(9); 2.348(8); 2.070(7) 1.929(7); 1.505(7) Å. The chemical analysis: Al₂O₃ ? 46.23; Fe₂O₃ ? 0.28; P₂O₅ ? 31.85 H₂O ? 21.00; sum 99.34, gives a formula Al₂(PO₄)(OH)₃ · H₂O. Colourless optically biaxial positive, n_S: α=1.562, β=1.566, γ=1.587, plane of optical axies (001), Z=a, Y=c; 2V=53°, weak dispersion r > v. Measured density 2.552. The DTA curve shows endothermic reactions at 250, 370 and 440°C corresponding to the dehydration of mineral. Infrared spectrum indicates the presence of OH and H₂O groups. Found in oxidation zone of Kouroudiako iron deposit, Senegal, associated with turquoise, augelite, wavellite and crandallite.