Showing papers in "American Mineralogist in 1975"

The structure of pseudorutile and its role in the natural alteration of ilmenite

Abstract: Samples of pseudoruti le from South Australian and Indonesian localit ies have been analyzed chemically and studied by X-ray diffraction techniques. A complete structure analysis has been carried out on a single crystal of the Indonesian material. The structure is based on hexagonal close-packed oxygens with metal atoms randomized with two thirds occupancy over half the available octahedral sites and fully ordered in the remaining sites. The ordered atoms form chains of alternately filled and empty octahedral sites along [001]. Poor correlation between these ordered chains leads to diffuse streaking of the metal-ordering reflections in certain planes perpendicular to [001]. We have obtained a satisfactory refinement of the single crystal intensity data for an azerage metal-ordering model defined by a hexagonal cell with a = 14.375 A and c = 4.615 A. The structural relationships between ilmenite, pseudoruti le, and ruti le are discussed. A twostage mechanism for the natural alteration of ilmenite to pseudorutile to rutile is proposed. An electrochemical corrosion model in the zone ofsaturation is suggested for the initial alteration of ilmenite to pseudorutile. Further alteration to rutile is considered to take place in a zone of oxidation via a dissolution and reprecipitation process. Introduction range of composition, e.g. Fe2Os.nTiOr.mH2O;3 I n ( 5 and I I m ( 2 (Dyadchenko and Khatuntseva, Natural weathering of i lmenite, FeTiOr, involves 1960). Much of the confusion regarding this inoxidation of the ferrous iron to ferric, and progrestermediate alteration compound resulted because the sive removal of iron by leaching, resulting eventually usual method of characterization, powder X-ray in the production of ruti le, TiOr. As early as 1909 diffraction,proved unsatisfactorybecauseof thepoor (Palmer, 1909) it was recognized that the alteration of crystall inity of the alteration phases and because of i lmenite to ruti le proceeded through a distinct inthe coincidence of many of the diffraction l ines from termediate species with approximate composition different phases. These problems were overcome by FerO' ' 3TiOr. The intermediate alteration phase was Teufer and Temple ( 1966) by the application of single termed arizonite by Palmer (1909). Whether arizonite crystal X-ray diffraction techniques to single crystal is an independent mineral with a definite structure grains of altered i lmenite from commercial i lmenite and composition has been the subject of considerablb concentrates. They identif ied a new compound with debate (Miller, 1945; Lynd, Sigurdson, North, and hexagonal symmetry and with composition approxAnderson, 1954; Dyadchenko and Khatuntseva, imating FerTirO, as the main constituent of the 1960; Bykov, 1964); a number of researchers supaltered product. Small amounts of auxil iary i lmenite ported the alternative thesis that arizonite was a and ruti le gave separate, easily identif iable single polymineralic mixture of ruti le, anatase, hematite, crystal X-ray patterns. Structure factor calculations and ilmenite (Ernst, 1943; Overholt, Vaux, and were carried out for the new compound, which these Rodda, 1950). However, regardless of whether it was authors named pseudoruti le. The general structure called afizonite, proarizonite (Bykov, 1964), or was shown to consist of a hexagonal close-packed oxamorphous iron-titanate (Bailey, Cameron, Spedden, ygen lattice with metal atoms statistically distributed and Weege, 1956), most researchers were agreed that over the available octahedral sites. However, such a the alteration of i lmenite proceeded through a disdisordered model for pseudoruti le, with a random distinct intermediate isotropic material with a narrow tribution of metal atoms, does not readily explain the

Mica polytypism: Dissimilarities in the crystal structures of coexisting 1M and 2M1 biotite

Abstract: Crystal-structure of coexisting lM and 2M, btotite polytypes from a rhyodacite lava flow, Ruiz Peak, New Mexico, shows that the two forms possess essentially identical unit-cell parameters (in the lM setting), bond length averages, and site occupancies and probably, therefore, very similar chemical compositions. The unit layers of the two polytypes, however, are different structually in that the x and y atomic positions of the 03 and 04 oxygens differ in the two forms by an average of 14 standard deviations. The average difference in the relative values of all other atomic positions is one standard deviation. The displacement of 03 and 04 in the 2Mrbiotitecauses a type of deformation not before reported in micas and manifests itself by a relative shifting of the upper and lower triads of octahedral oxygens as a unit along the aD directions (2Mt setting). This causes a reduction in symmetry of the2M, unit-layer from the ideal C2/m as found in the lM polytype, to Cl. It is postulated that the structural differences between the unit layers of thelM and,2M, biotite polytypes are due to the different atomic and geometric constraints imposed upon the unit layer by the adjacent layers, the latter being in different relative orientations in different polytypes. An origin for biotite polytypism is proposed on the basis of these unit-layer constraints coupled with the mica platelet growth and spiral growth observations of A. Baronnet.

The crystal structures of pyrope and grossularite at elevated temperatures

Abstract: Single crystal X-ray intensity data have been collected for a grossularite at 25\" ,365\" and 675\"C and a pyrope at25o,350\",550o, and 750oC. Anisotropic least-squares ref inements reveal that the oxygen positional parameters remain approximately constant in grossularite but change in pyrope. Linear thermal expansion coefficients per 1'C l(h lrr.)/(T 25\\/1\"*) for the mean interatomic distances Ca-O and Al-O in grossularite are 0.8 X lO-u and 1.3 X 10-6, respectively, while those from Mg-O and Al-O in pyrope are 1.3 X 10-5 and 0.8 X 10-5, respectively. The Si-O interatomic distances do not increase significantly as a function of temperature. In pyrope, with increasing temperature, the rigid SiO. tetrahedra rotate to allow the shared octahedral edge to lengthen at a greater rate than the unshared edge, thereby decreasing the octahedral bond angle strain. In grossularite the tetrahedra do not rotate, and the octahedral bond angle strain does not change. Structural changes in aluminum silicate garnets as a function of heating are compared to changes resulting from chemical substitution at the triangular dodecahedral site.

Thermochemical approximations for clay minerals

Abstract: The Gibbs energies of formation for clay minerals have been approximated by a model which assumes that these minerals are formed by the combination of silicon hydroxide with metal hydroxides. The change in free energy during the polymerization ofhydroxides to form silicates plus water is shown to be generally small and has been represented by a correction term. The dispersion between the predicted and the experimental data of better than * 2.0 kcal mol-r is comparable to the total uncertainty in the experimentally determined constants. The approximated free energies of formation for vermiculites and ferromagnesium chlorites are presented. Introduction, The study of the chemical constraints which predicate the environments of formation (or alteration) of clay minerals has been hindered by the sparsi ty of relevant thermochemical constants. Although several workers have determined the Gibbs free energy of formation (AGi) for kaolinite (e.g., see Polzer and Hem, 1965; Kittrick, 1969; Reesman and Keller, 1968; Huang and Keller, 1973), few (if any) AGy0 for the other clay minerals are available. The recently reported AGro for montmorillonites and illites (Reesman and Keller, 1968; Helgeson, 1969; Huang and Keller, 1973) pertain to specific or idealized minerals and may be of limited geochemical applicability. The AG10 for ferromagnesium chlorites and vermiculites remain to be determined. This report presents a method for approximating the AG10 for clay minerals with an accuracy which is adequate for many geochemical calculations. It is intended as a first step towards developing better methods for approximating the AGro for these minerals; the gambit will undoubtedly be modified and extended as accurate experimental data become available. Obviously there is a need for methods to estimate the AGro values for clay minerals. The experimental determinations are difficult, and many have considerable room for improvement. Clay minerals show considerable variations in composition, and it is unlikely that the AGro for every one of them will ever be determined experimentally. The Predictive Method The predictive method employed in this report considers the clay minerals to be formed by the combination (\"polymerization\") of simple metal hydroxides with silicon hydroxide, Si(OH),. The change in free energy during the polymerizatton of the hydroxides to form clay minerals is usually small and has been approximated by a one parameter correction term. The model in its present form is essentially heuristic and the validity for its use rests chiefly on the good agreement between the predicted and experimental data. The mechanics of the procedure may be illustrated by considering the formation reaction of a typical montmorillonite: n,NaOH * nrKOH * naCa(OH), * naMg(OH), * nuFe(OH)' + n.Al(OH! + n,Si(OH). NaorKorCaorAlnrFeouMgorSirTO,o(OH), * (Znpr l2)HzO ( l ) where z1 is the reaction coefficient of the i-th hydroxide and z1 is the charge on the i-th cation (including Si). Usually z1 corrospoflds to the amount of the i-th cation (or silicon) in the formula unit for the clay mineral. The free energy of formation for the clay mineral whose formation is described by the above reaction is calculated by the expression: AGyo (Montm.) : Znt AGlo (r1) (2np1 l2)LGto(H,O)Q Q) where AG10 (r) is the free energy of formation of the i-th hydroxide component and Q is an empirical correction factor defined as Q = o(n121 12) (3) i.e., Q is expressed as a function of the number of moles of water released during the reaction.

A sample holding technique for study of crystal growth in silicate melts

Abstract: A thin platinum wire loop is an effective way to hold silicate melts during experimentation in a gas-mixing furnace. This method results in a minimum of physical and chemical interaction between the sample and container but maximum interaction between sample and gas mixture. However, volatilization of sodium occurs while the silicate is molten. By minimizing the chance of heterogeneous nucleation, the method is ideal for experimental investigation of the origin of rock textures.

Cation ordering in Ni-Mg olivine

Abstract: A synthetic crystal of Ni-Mg olivine-Ni,.*MgorSiO,;a = 4.7366(4); b : l0.l7l6(13); c: 5.9374(4) A-has been investigated by X-ray methods in order to determine the intracrystalline distribution of Ni and Mg. Anisotropic least squares refinement of the structure, including cation occupancies, to R = 0.029 indicates significant ordering of Ni on Ml (0.767 Ni + 0.233 Mg; o = +0.002 Ni) and of Mg on M2 (0.737 Mg + 0.263 Ni). The intracrystalline distribution coefficient, Kp : (Mg[M21/N|IM2])/(M\IMll/ Ni[Ml]), equals 9.22. This amount of Ni-Mg ordering is surprising in view of the small size difference [Ar = 0.03 A] between the two cations and the slight ordering observed in some Fe-Mg olivines [Ko = 1.00 - 1.30] and suggests that crystal field effects strongly favor Ni on the Ml site of olivine, in agreement with the predictions of Burns (1970a) and others. Comparison of the available site occupancy data for transition metal olivines and pyroxenes indicates that site preferences of divalent Fe, Co, Ni, and Mg depend on a balance between such crystal chemical factors as cation size, charge, electronegativity, crystal field effects, and the nature of the two octahedral sites in each structure. In particular, the differences in residual charge, f, at these sites [Zyr,o', : Zar

The crystal chemistry of complex niobium and tantalum oxides. IV. The metamict state: Discussion

Abstract: Graham and Thornber (1974b) have proposed a mechanism for metamictization in complex niobium-tantalum oxides in which complex compositions, rather than radiation damage, are the primary requirement. This note reviews the literature in support of radiation damage as a necessary condition for metamictization and suggests that the structural stability of the metamict "glass" relative to that of the crystalline pre-metamict material is an important consideration in predicting whether a substance will occur in the metamict state.

Spectroscopic and magnetic studies of ferric iron hydroxy sulfates: Intensification of color in ferric iron clusters bridged by a single hydroxide ion

Abstract: Optical and infrared spectroscopic data and magnetic susceptibility data are reported for the ferric hydroxy-sulfate minerals magnesiocopiapite and botryogen and the related mineral coquimbite. The physical properties of intensity of color, pleochroism, and magnetic susceptibility associated with the hydroxobridged Fe^(3+) dimer of magnesiocopiapite and with the hydroxobridged Fe^(3+) infinite chain of botryogen are compared to those of isolated Fe^(3+). The hydroxobridged minerals show strong pleochroism with intensity enhancement of the spin-forbidden ligand field absorption bands occurring when the electric vector of the incident light is oriented in the direction of the Fe-OH-Fe axis. The intensity enhancement and pleochroism are associated with antiferromagnetic exchange coupling of the Fe^(3+) ions.

Refinement of the margarite structure in subgroup symmetry

Abstract: The crystal structure of margarite-2M1 from Chester County, Pennsylvania, has been refined as an ordered derivative structure in subgroup Cc of the ideal space group C2/ c. Because of the high pseudosymmetry involved, successful refinement by least-squares required initial movement of atomic parameters away from those of the disordered phase towards those of the possible ordered models predicted by a distance least-squares program. Ordering of tetrahedral Si and AI is nearly complete in a pattern that violates centrosymmetry between the two tetrahedral sheets within a 2: I layer. Compositionally similar tetrahedra in adjacent sheets are related instead by a pseudo two-fold axis that extends laterally through the oc­ tahedral AI atoms and is normal to the direction of intralayer shift. The two tetrahedral sheets differ si.in Si,Al contents. A similar ordering pattern is theoretically possible in muscovite-2M, also, but was not found to be adopted by muscovite from the Diamond mine, South Dakota.

Phase relations in the systems PbS-Ag2S-Sb2S3 and PbS-Ag2S-Bi2S3

Abstract: Phase relations in the systems pbS-AgrS-SbrS, and pbS-AgrS-BirSs were studied in the temperature range 300o-500'c using evacuated glass capsule technique. The pbS_AgsbS, join shows a complete solid solution at 500oC, a miscibility gap between 65 and 90 moti percent PbS at 400oC, and at 300oC the presence of two small solid solutions between 50 and 52 mole percent PbS, and 25 and 40 mole percent PbS. Another solid solution includes the compositions of andorite (pbAgSbrS.), ramdohrite (pbrAgrSb.Srr), and fizelyite (pbuAgrSbsS16). An extensive solid solution exists along the lillianite-pavonite join, and iras a misciUitity gap between 25 and 46 mole percent PbS at 500'C. Pavonite also exhibits solid sotution toward galenobismutite. Along the PbS-AgBiS, join, a complete solid solution exists between 400o and 500'C. Correlations of experimental data obtained in this study with some natural mineral assemblages were made.

Deformation and Recrystallization of a Plagioclase Grain

Abstract: A large deformed grain of plagioclase (Anno) contains numerous deformation (kink) bands and fewer broader kink-like zones that may have involved fracturing. The plagioclase has been recrystallized along narrow zones of relatively high strain. The orientations of the new grains are close to that ofthe original grain, implying crystallographic control in the recrystalIization process. Small, consistent chemical differences exist between old and new grains, suggesting ihat small chemical changes provided some of the driving force for nucleation. The lrain stro*, evidence that subgrain nucleation, bulge nucleation, and sintering of fragments have contributed to the production of the recrystallized aggregates'

High resolution electron microscopy of enstatite. II: Geological application

Abstract: A number of polytypes, existing over short intervals of structure, have been observed in enstatites. These include the27,36, and 54 A periodicities, in addition to the well known 9 A monoclinic enstatite (CLEN) and l8 A orthorhombic enstatite (OREN) polymorphs. The polytypes, as well as all other features described in this paper, can be explained on the basis of a twinning model. This includes "antiphase boundaries," which we interpret in terms of the normal (100) twin planes stepping across adjacent (100) planes. Parting planes can be seen in their earliest stages of development as simple extensions of the (100) twin planes. They presumably occur in regions where lattice strains have been most highly localized. These strains can be observed as dark contrast in the electron photomicrographs. Parting is produced by shear plus rotation. It is possible to distinguish between CLEN that formed by solid state transformation (a) directly from OREN in the absence of shearing, (b) from OREN by shearing, and (c) by temperature quench from the protoenstatite stability field. This is done by measuring the width of CLEN fields within OREN and noting the degree of CLEN twinning. Most meteoritic CLEN formed by shearing.