The crystal structures of low-temperature and high-temperature albites

Abstract

Starting with the structure of (low) albite published by Taylor, Darbyshire & Strtmz in 1934, the crystal structures of a low albite and of a high albite, both nearly pure NaA1SiaO s, have been refined by means of a series of Fo and (Fo--Fc) Fourier projections parallel to all three axes.The mean bond lengths within the four non-equivalent tetrahedra are in low albite, 1-743, 1.590, 1.636 and 1.616 /i~, and in high albite 1.653, 1-639, 1.643 and 1.647 /~, with a standard deviation 0.02 A. It is concluded that in low albite the first site Sil(0) contains nearly all the A1 and that in high albite the A1 and Si atoms are randomly distributed throughout the four tetrahedral sites. The dimorphism of soda felspar is thus due primarily to differences in the degree of A1-Si ordering.Four features of the Na atom were noted: (1) In both albites the temperature factor of this atom is much greater than that implied in the f curve of Bragg & West. (2) The temperature factor is greater in high albite than in low. (3) In low albite this atom behaves as though it had an anisotropic thermal vibration with a maximum amplitude nearly along y equivalent to an atomic separation of ~ 0-1 ~. (4) In high albite the effect is similar but much more intense, equivalent to a separation of N 0.6 /~. A possible interpretation is that the Na atom occupies at random through the structures one or other of two positions, within the same large cavity, separated by N 0.1 /~ or ---0.6 A nearly along y. On this view, when the A1-Si atoms are disordered as in high albite the cavity available for Na is effectively larger than that atom can fill, whereas when the A1-Si atoms are largely ordered, as in low albite, the cavity is small enough to nearly enclose the Na atom.On the assumption that maximum stability at room temperature corresponds to local balance of electrostatic charges throughout the structure, a detailed discussion of the two albites and of sanidine and intermediate microcline leads to a number of unexpected and important conclusions: (1) In low-temperature felspars the most stable structure is not necessarily, as is generally assumed, one in which the A1-Si atoms are completely ordered. (2) The most stable potassium felspar is not 'maximum' microcline but a monoclinic (C2/m) orthoclase with A1 partially ordered into one half of the tetrahedra. (3) Intermediate microcline has an unstable charge distribution and all microclines lie outside the normal stability range of the potassium felspars, which runs from disordered sanidine to partially ordered orthoclase. (4) Most microclines probably form as the exsolution product of an alkali felspar which cooled from fairly high temperatures and which, at those high temperatures, contained a sufficient proportion of Na to confer partial ordering and the associated triclinic symmetry on the tetrahedral framework. General introductionSodium felspar (NaA1SiaOs) may exist in one of two well established modifications. The ordinary lowtemperature form, called variously low-temperature * Progress reports...

Refinement of the atomic parameters of α-quartz

“…The edge length of the Si04 tetrahedron varies much more widely in the silicates, evidently being sensitive to such factors as the mode of linking of the tetrahedra and the presence and nature of interstitial cations, so that it can only be said that the intramoleeular O-O distances in a-quartz of 2.604 to 2.640 A lie somewhat below the midpoint of the range observed in the silicates. The Si-O-Si bond angle of 144.0 ° lies almost precisely at the midpoint of the range of values observed in silicates containing linked tetrahedra; for example, in the felspars some of the reported ranges of the Si-O-Si angle are as follows: high albite 125-160 ° and low albite 128-155 ° (Ferguson, Taylor & Traill, 1958), microcline 130-155 ° (Bailey & Taylor, 1955), orthoclase 131-153 ° (Jones & Taylor, 1961. By X-ray radial distribution analysis of silica glass Zarzycki (1957) has deduced that the average Si-O-Si angle is 143 °, but the precision of the method, _+ 17 °, is not good enough to justify quantitative comparisons with crystalline systems.…”

Ab initio calculations of elastic constants of plagioclase feldspars

“…The effect of Al distribution within T-sites has been widely studied (e.g., Ferguson et al 1958;Ribbe et al 1969;Prewitt et al 1976;Winter et al 1979;Carpenter et al 1985Carpenter et al , 1990Kunz and Armbruster 1990;Sochalski-Kolbus et al 2010), and has been found to play an important role in the topology of the tetrahedral framework. Because Al atoms are larger than Si atoms, the Al-O bonds are longer than Si-O bonds.…”

The crystal structure of an anorthoclase: an intermediate alkali feldspar

“…We may mention, in this connexion, the determination and refinement of two sanidine structures obtained by heating orthoclase (Cole, S~rum & Kennard, 1949;Ribbe, 1963), of natural sanidine (Onorato, Penta & Sgarlata, 1963), of an orthoclase (Jones & Taylor, 1961;Colville & Ribbe, 1968), of an adularia (Colville & Ribbe, 1968), of an intermediate microcline (Bailey & Taylor, 1955), and of two maximum microclines (Brown & Bailey, 1964;Finney & Bailey, 1964). In addition, the high and low albite structures are well known (Ferguson, Traill & Taylor, 1958;Ribbe, Megaw & Taylor, 1969). It would appear, therefore, that apart from some as yet unresolved details the end-members of the alkali feldspar series are well determined.…”