Identificatie van natuurlijke alkaliveldspaten met behulp van röntgen-poederdiagrammen

Identification of natural alkali felspars with X-ray powder photographs. X-ray powder analysis is becoming an important tool for the petrographer when identification problems can not be solved with the usual optical and chemical methods. It is the aim of this paper to provide data to identify alkali felspars in groundmasses of extrusive rocks, perthites and other fine grained structures. Moreover the variation of the intensities and the position of spacings of the powder patterns of natural alkali felspars is compared with the variation in optical properties and chemical composition. To this purpose alkali felspars of different localities, chemical composition, crystallization temperature and rate of cooling are investigated with optical methods, X-ray powder analysis and as far as possible, chemical analysis. The optical examination of the alkali felspars was made with the four axes universal stage. The position of the poles of crystallographic elements and twinning axes was determined with respect to the axes of the indicatrix N\u03b1, N\u03b2, N\u03b3. The co-ordinates are recorded according to Nikitin (1936). The quadrant in which each pole is situated is indicated by the sign + or —. In plate III the measurements on the potash-soda felspars are plotted in a projection normal to N\u03b21. The interpretation normal orthoclase-Naorthoclase was made with the aid of the co-ordinates given by Nikitin (partly reproduced in table I) who did not give a chemical definition of these terms. The available chemical data in this investigation proved that thus defined normal orthoclase contained < 25 % Ab and Na-orthoclase > 25 % Ab in solid solution. Determination of refractive indices was used to distinguish anorthoclase from both “low temperature” albite and potashfelspar. The alkali felspars investigated were grouped according to their natural paragenesis. Crystallization temperature, rate of cooling and stability within these groups are discussed. 1. Alkali felspar phenocrists from extrusive rocks. Large sanidine phenoerists (d. 5,5 m.m.) from Lagno de Pollena, Vesuvius, show a zoned structure // (010), (fig. 3). In sanidine of Siebengebirge wedged in between large homogeneous crystals (d. 8—10 m.m.) appear small zoned sanidine crystals (d. 1—3 m.m.) which show polysynthetic twinning lamellae in many directions (fig. 2). Probably this is a product of later crystallization under stress. Anorthoclase of Puy de Dôme (fig. 7), Pantelleria and Mnt. Anakie, Australia (fig. 4) show an extremely fine albite twinning which seems to be typical for anorthoclase. Refractive indices (n\u03b3=1,529) and X-ray powder pattern (fig. 18) are characteristic and different from those of “low temperature” albite. In trachites of Colli Euganei, Italy, phenocrists were observed (fig. 5) with a core of “high temperature” oligiclase (26 % An, 2V=—84°) passing in a rim of anorthoclase (2V=—60°). This proves the existence of a continuous series of solid solutions between h.t. oligoclase and anorthoclase. 2. Alkali felspars from plutonic rocks and dykes. Examples of cryptoperthites, orthoclase- and microcline microperthites and untwinned microcline are described. 3. Alkali felspars from pegmatites. Different structures of microcline perthites are described. In fig. 15 is shown how vein albite // (001), with an irregular surface regulates the position of adjacent microcline twinning lamellae. In this case the microcline twinning lamellae seem to be younger than the vein albite. On the other hand simultaneous crystallization as suggested by Spencer (1938, p. 107) seems not impossible. The most frequent occuring type of vein albite in microcline is reproduced in fig. 23, cutting the microcline lamellae under an angle of 60° with the (010) cleavage in (001). The vein albite is consequently younger than the microcline. Therefore Andersens (1928) suggestion that this vein albite is produced by infiltration of albite solutions in oriented shrinkage cracks may explain the constant orientation of the vein albite. Spencer’s hypothesis of the cotectic origin of vein albite can only hold for isolated examples as mentioned in the description of fig. 15. The majority of vein albite in microcline is of secondary origin. Examples of patch perthite produced by replacement are shown in fig. 14 and fig. 24. As examples of “high temperature” pegmatites a cryptoperthite from Larvik, Norway, and orthoclase from Itrongay, Madagascar, are described. A number of crystals of the well known monoclinic “orthoclase” of Baveno produced X-ray powder patterns characteristics for microcline with additional albite reflections. Optical examination showed that these crystals are strongly altered to kalinite and invaded by secondary albite (see Baveno twin of fig. 8). Other crystals showed recrystallization of fine grained microcline and albite (fig. 9). With high magnification an initial microcline twinning is observed (fig. 10). It seems probable that most crystal of Baveno “orthoclase” on display in mineralogical musea, on optical examination will be found to show a pseudomorphosis of orthoclase by microcline. 4. The adularia-albite paragenesis. In most of the examined adularia crystals from St. Gotthard, Bristenstock and Maderanerthal locally triclinic lamellae were observed which show extinction angles of 2°—6° with the (010) cleavage in (001). These triclinic zones are nearly always situated round inclusions (fig. 21) and may be found in the core as well along the faces of the crystals. They are to be compared with the triclinic zones found in sanidine (fig. 2). Axial angles and extinction angles are different from microcline. Chemical analysis in weight percents of some of the alkali felspars investigated are listed in table 2 and fig. 16. The Or-Ab-An components are expressed in molecular percents. SiO2 values are generally too low and Al2O3 and Fe2O3 values to high. For the samples no. 4, 48, 49, 33 and 23 this may be explained by the occurrence of alteration products. X-ray powder photographs were obtained with an iron target, Mn filter and a 9 c.m. diameter Unicam powder camera. The diameter of the diafragma slit was 0,3 m.m.. Tube current and voltage were 18 m.A. and 40 k.V. respectively. The accuracy of the measurement of spacings was 0,02 m.m. corresponding with 1,9’ \u03c6. Measurements were corrected by the admixture of 10 % Nall. Intensities were estimated visually. Examining the powder patterns of the alkali felspars, five groups could be distinguished, classified independently of chemical composition and optical properties. Group A (plate I A and II A and B). A similar pattern was observed for sanidine, orthoclase of plutonic rocks, dykes and pegmatites and hydrothermal adularia. Samples investigated are listed in table 8. In table 3 intensities, \u03c6Fe- and d-values are recorded for St. Gotthard adularia and Drachenfels sanidine. Characteristic are the two strongest reflections (202) and (002) (040). Group B (plate I B). All microclines and untwinned microclines give a similar pattern which différa from the group A pattern by showing a single strong (002) (040) reflection followed by three groups of each three reflections with the same intensity (p, q and r in fig. 17). Intensities, \u03c6Fe- and d-values are recorded in table 4. Samples investigated are listed in table 9. Group C (plate I C). The powder pattern data of “low temperature” albite are recorded in table 5. Samples investigated are listed in table 10. Additional albite reflections of orthoclase- and microcline perthites are indicated respectively with AC and BC in table 8 and 9. Group D (plate II D). In table 6 are recorded the intensities, \u03c6Fe and d-values of a typical anorthoclase. The investigated samples are listed in table 11. Group E (plate II C). In table 7 the powder pattern properties are recorded of a ciwptoperthite with a high An-content. The facts recorded in table 8—12 show complete agreement between the classification of alkali felspars with powder patterns and the classification on optical properties. It is not possible tot distinguish between sanidine and orthoclase with the aid of powder photographs. So the optical properties seem to be more sensitive to small changes in structure. The powder patterns of all felspars have the strong reflection (002) (040) in common. The powder patterns of the alkali felspars with the exeption of “low temperature” albite differ from those of the plagioclases by the possession of an isolated strong reflection (043) (062), (d=1, 79—1, 78, s in fig. 17 and fig. 18). Characteristic for sanidine, orthoclase and adularia (group A) with a composition up to 45 % Ab is the strong reflection pair (202) and (002) (040). The microclines (group B) are characterized by a single strong (002) (040) reflection followed by three groups of each three reflections of the same moderate intensity (p, q and r in fig. 17). The anorthoclase powder pattern which differs distinctly from the “low temperature” albite pattern is distinguished from the other alkali felspars by the presence of an isolated reflection of moderate intensity with d=3,15 (t in fig. 18). The distance between the two strongest reflections (202) and (002) (040) of the powder patterns of sanidine, orthoclase from plutonic rocks, dykes, pegmatites and adularia proved to vary nearly linear with the Ab-content contained in solid solution. The distances were measured with the microscope with low magnification (X 19). In fig. 19 the variation of the distance between (202) and (002) (040) expressed in minutes (\u03c6) is plotted against the Ab-content in molecular percents, calculated out of the chemical analyses available of homogeneous crystals of group A. The strong reflection of anorthoclase (106) seems to be doubled under the microscope. The corresponding distance does not fit in the diagram of fig. 19. A similar variation diagram for group A is plotted in fig. 20 in which the distances between the reflectons a and b (indicated in table 3) are used. The more time consuming absolute measurements of the position of certain spacings may also be used for the determination of the composition (see table 3—7 and fig. 17 and 18). The Ab-component of orthoclase- and microcline perthites was easily observed in the diffraction patterns. Comparison with artificial mixtures of “low temperature” albite with orthoclase and microcline are shown in plate I, D, E, F, G. Excepting a cryptoperthite of Larvik, Norway, with an exceptional high An-content (group E) the albite component of the cryptoperthites (f.i. moonstone from Ceylon) could be easily detected. As in most cases only the strongest reflections of the albite component were present, is was not possible to make ure that “high temperature” albite was present 1). As contrasted with the cryptoperthites the investigated anorthoclases of Puy de Dôme, Pantelleria, Colli Euganei and Mnt. Anakie, Australia, proved to be optical and roentgenographical homogeneous. Although no natural or artificial “high temperature” albite was available for investigation it seems probable that the powder pattern of anorthoclase (plate II D, table 6, fig. 18) must be similar to that of “high temperature” albite. Felspars of rhomb porphyries, Oslo district, showed a powder pattern characteristic for oligoclase in agreement with the optical investigation of Oftedahl (1948). Investigation of X-ray powder photographs of the plagioclases gave similar results as obtained by Claisse (1950). Powder patterns of anorthite from efflata of Monte Somma, Vesuvius (92 % An), anorthite of Pesmeda, Tyrol (94 % An) and anorthite of Kamitsuki, Miyake-Jima, Japan (98 % An), although very similar, showed differences in spacings and intensities which can not be explained by changes in composition. Differences in crystallization temperature and rate of cooling may be responsible for these structural differences. X-ray powder photographs of groundmasses of trachites, rhyolites, andesites, bostonites, pantellerites and helleflints showed the presence of alkali felspars, plagioclases and quartz (cristobalite, tridymite), see table 14. Comparison powder photographs of mixtures of quartz and felspar of known concentration permitted the estimation of the quartz content of the groundmasses. In plate II E a powder photograph of charnockite is reproduced. With optical methods is was impossible to determine whether the mesoperthite present consisted of orthoclase- or microcline perthite. Comparison with diffraction patterns of quartz (II F), a mixture of 80 % l.t. albite and 20 % quartz (II G) and a mixture of 80 % microcline and 20 % quartz proved the presence of quartz and microcline perthite in the charnockite. In the last part of the paper the relation orthoclase-microcline is discussed and the existing opinions reviewed. The hypothesis Mallard-Michel-Lévy states that orthoclase consists out of submicroscopical twinned microcline units. The starting-point of this hypothesis is the supposed general occurence of intimately intergrown orthoclase and microcline. Now observations made by Mäkinen (1917), Baier (1930), Gysin (1928, 1938) and the present author tend to the conclusion that untwinned and partly twinned microcline are common; intergrowths of orthoclase and microcline however are limited to contactmetamorphic phenomena as described by Wimmenauer (1950). Triclinic lamellae in sanidine and adularia are not identical with microcline. The influence of stress, advocated by Brauns (1891) as the cause of microcline formation is negligible, as is demonstrated by the common occurrence of free grown microcline crystals. The general occurrence of microcline in slightly metamorphosed rocks is due to the fact that these rocks attained equilibrium in the temperature region of 750°—500° C. (Spencer 1937, p. 481). Our optical investigation shows that there is a certain variation of the optical properties of microcline. A continuous change towards the optics of orthoclase was not observed. Considering these facts, together with the arguments put forward by Spencer (1938, p. 88), the submicroscopical twinning hypothesis seems improbable. According tot the hypothesis of Barth (1934), modified by Buerger (1948) microcline is formed by ordering of the Si and Al atoms with declining temperatures. The difference in spacings and intensities found in the powder pattern of microcline indicates that the microcline structure shows a small distortion compared with the orthoclase structure. Finally the optical anomalies of adularia are discussed. The difference between the symmetrie relations of microcline and triclinic adularia is demonstrated in fig. 21 and 22. The crystal structure of adularia seems to be similar to the orthoclase structure. Locally triclinic may originate round inclusions and disturbed areas during the crystallization. The structure of these triclinic lamellae is essentially different from the microcline structure originated by the complete ordering of the Si and Al atoms. Contrary to the opinion of Köhler (1948) it is evident that alkali felspars with an orthoelase structure crystallize at relatively low temperatures (450°—200° C.) which is also proved hy the occurence of authigenic felspar. Considering the polymorphism of the alkali felspars, exceptional conditions during the crystallozation must explain the formation of these “low temperature” forms.