ISSN 1069-3513, Izvestiya, Physics of the Solid Earth, 2006, Vol. 42, No. 6, pp. 477–480. © Pleiades Publishing, Inc., 2006. Original Russian Text © V.I. Fel’dman, L.V. Sazonova, V.V. Milyavskii, T.I. Borodina, S.N. Sokolov, A.Z. Zhuk, 2006, published in Fizika Zemli, 2006, No. 6, pp. 32–36.
Shock Metamorphism of Some Rock-Forming Minerals V. I. Fel’dmana, L. V. Sazonovaa, V. V. Milyavskiib, T. I. Borodinab, S. N. Sokolovb, and A. Z. Zhukb a Faculty
of Geology, Moscow State University, Vorob’evy gory, Moscow, 119899 Russia Institute for High Energy Density, Joint Institute for High Temperatures, Russian Academy of Sciences, Izhorskaya ul. 13/19, Moscow, 127412 Russia b
Received August 25, 2005; in final form, November 7, 2005
Abstract—The shock metamorphism of schist consisting of garnet, biotite, quartz, and plagioclase is studied under shock wave loading of a sample in steel recovery ampoules of plane geometry. A maximum shock pressure was reached during several circulations of waves in the sample (stepwise shock compression) and varied within the range 19–52 GPa. The recovered samples were examined by the methods of scanning electron microscopy and microprobe and X-ray phase analysis. The results were compared with natural impactites and with shock-induced alterations in minerals loaded by a spherical convergent wave. It is established that, given a plane geometry of loading (stepwise shock compression), solid-state transformations at the lattice level (migration of chemical elements and formation of shock thermal aggregates) are not observed in all of the studied minerals, in contrast to natural impact processes and spherical geometry experiments. Under the conditions of our experiments, minerals melt at higher pressures than in the case of natural impact processes and spherical geometry experiments. However, for each mineral studied, the mechanical strain patterns at close shock pressures are, on the whole, the same for all of the aforementioned three variants of shock wave loading. PACS numbers: 91.65.Kp DOI: 10.1134/S1069351306060048
INTRODUCTION The collision of cosmic bodies (asteroids, meteorites, or their fragments) with the surface of the Earth or other planets is accompanied by physical processes leading to the formation of a specific type of rocks, impactites. Shock metamorphism and melting played a significant role in the formation of the crusts of terrestrial planets and, therefore, the study of the specific features of impactite formation is of great significance for the general reconstruction of the origination, development, composition, and structure of the Earth’s crust. The effects of impact events should also be taken into consideration in studies of meteoritic substance [Lavrukhina and Mil’nikova, 1984]. The physical simulation of impact processes under laboratory conditions is helpful for gaining insights into processes controlling the shock metamorphism of rocks and minerals. In the present work, we studied the shock metamorphism of schist consisting of garnet (40–45%), biotite (20–25%), quartz (5–10%), and plagioclase (25–30%). The initial samples were taken in the NW area of the Taratashskii metamorphic complex (the Southern Urals) and had a density of 3.13 ± 0.03 g/cm3. The rock samples in the form of disks 15 mm in diameter and ~1 mm in thickness were placed into steel recovery ampoules of plane geometry. Experimental assemblies consisting of a recovery ampoule, a steel guard ring,
and a massive steel basement were loaded by a plane shock of an aluminum plate 90–100 mm in diameter and 4–10 mm in thickness accelerated with the help of explosive projectile systems (EPSs). As the shock wave approached the sample, the pressures in the ampoule material varied within the range 19–52 GPa. The maximum shock pressures in the study samples were attained upon a few circulations of waves in the sample (stepwise shock compression) and were comparable to the pressures in the ampoule material. The duration of the pressure impulse applied to the sample (the interval between the first shock wave front arrival at the sample face and the head wave of rarefaction propagating from the back surface of the projectile) was 0.6–1.9 µs (depending on the EPS type in use). Dependences of the pressure in the study material on the loading time that are typical of the assemblies and EPSs used in our work are presented in [Zhuk et al., 2000]. The preserved material was examined by the methods of scanning electron microscopy (SEM), microprobe analysis, and X-ray phase analysis (XPA). The results obtained from these experiments were compared with natural impactites [Fel’dman, 1990] and with the mineral alterations induced by a spherical convergent shock wave [Kozlov et al., 2003, 2004a, 2004b]. For this purpose, the maximum shock pressures in natural rocks of shock metamorphism were estimated with the help of a quartz geobarometer
477
478
FEL’DMAN et al.
Qtz Bt
30 µm Fig. 1. Intensely molten biotite grain (a shock pressure of 52 GPa).
[Fel’dman, 1990] and, in material preserved after spherical shock wave loading, they were determined from the numerical solution of a system of mechanical equations describing a continuous compressible medium [Kozlov et al., 1998]. We should note that the quartz geobarometer is based on the determination of the relative amount of quartz grains with planar elements of different crystallographic orientations in a thin section of the shock-metamorphic rock under study and can be used only up to the pressures of quartz amorphization (~30–36 GPa). There are a number of diagrams of the “property–pressure” type, which are based on the results of laboratory experiments on the shock loading of various rocks and minerals (see the review in [Badyukov, 1986]), but the problem of estimating higher shock pressure loads that cause the formation of a particular type of natural impactites cannot be regarded as presently solved [Fel’dman, 1990]. The accumulation of the experimental information required for the construction of a geobarometer within a wider range of pressures is one of the goals of our work. BIOTITE Within the entire range of pressures studied, mechanical deformations in biotite (Bt) are observed in the form of cracks and kink bands of different crystallographic orientations. The melting of biotite along the cracks and at contacts with other minerals is fixed above a pressure of ~30 GPa. The formation of very fine veinlets in quartz that are composed of biotite glass is additional evidence for biotite melting under this shock loading. Intense melting of biotite starts with a further rise in the loading pressure. The entire mass of a biotite grain is often found to be molten, but some
grains remain unmolten even at a pressure of 52 GPa. In SEM images, the melting is indicated by the presence of numerous bubbles in biotite grains (Fig. 1); these bubbles arise when biotite starts “boiling,” i.e., when the fluid phase leaves the biotite melt. The completely molten grains of biotite can retain their boundaries, but they often lose their initial form and their boundaries become indistinct and diffuse. In the diffraction patterns, the evidence for the presence of molten biotite is a decrease in the integral intensities of diffraction peaks corresponding to this phase. This effect is fixed at pressures above ~30 GPa and is enhanced with an increase in the loading intensity. According to the XPA data, approximately 70% of biotite melts due to the stepwise shock compression of the rock up to 52 GPa. The melt forming from biotite at 52 GPa differs in chemical composition from the initial biotite due to the exchange of chemical elements with the surrounding melts derived from plagioclase and garnet: it is enriched in calcium (1.32 instead of 0.0 wt % of CaO) and in iron (30.33 instead of 19.21 wt % of FeO) but is depleted in potassium (1.11 instead of 9.91 wt % of K2O). A decrease in pressure and temperature initiates the crystallization of new phases from this melt. One of these phases coincides in composition with aluminous ringwoodite (up to 16 wt % of Al2O3), is 1–2 µm in size, and has a weakly elongated shape occasionally close to a hexahedron. The formation of ringwoodite from biotite under loads of ~30 GPa and higher was observed in experiments with spherical convergent shock waves [Kozlov et al., 2003]. The second phase has an irregular shape and sizes of a few micrometers and is close in composition to the grossular-rich garnet (60–68% of grossular, 28–34% of almandine, and 4– 7% of pyrope). Such a combination of minals does not occur in natural impact structures. Estimation based on the equilibrium phase diagrams of the systems SiO2– Al2O3–CaO, SiO2–Al2O3–FeO, and SiO2–Al2O3–MgO [Minerals …, 1974] shows that this combination of minals can exist at pressures less than ~0.5 GPa and temperatures below ~1000°C. Taking into account that the field of stability of ringwoodite is located in the region of higher pressures [Ultrahigh-Pressure …, 1998], we may conclude that these two phases were consecutively crystallized at different times during the unloading of the sample by the rarefaction wave. In experiments with the loading of plagioclase– quartz–biotite schist by a spherical convergent wave [Kozlov et al., 2004b], partial melting of biotite along cracks was observed at ~25 GPa. Intense melting of biotite in the preserved material established by the SEM technique at ~30 GPa was accompanied by the crystallization of new phases from the biotite melt, and nearly complete melting of biotite with the formation of glass was observed at ~40 GPa. In experiments with the loading of biotite–muscovite gneisses by plane shock waves (stepwise shock
IZVESTIYA, PHYSICS OF THE SOLID EARTH
Vol. 42
No. 6
2006
SHOCK METAMORPHISM OF SOME ROCK-FORMING MINERALS
compression), partial melting of biotite was fixed at pressures above ~33 GPa [Lambert, 1979; Lambert and Mackinnon, 1984]. Traces of biotite melting were also fixed in some astroblemes [Lambert and Mackinnon, 1984], although natural impact processes are typically associated with the decomposition of biotite resulting in the formation of new phases and apobiotite residue (in some cases, glass), just as occurs during the heating of biotite under static conditions [Fel’dman, 1990; Lambert and Mackinnon, 1984; Stöffler, 1972]. Experiments with the loading of biotite by a plane shock wave (single-step shock compression) to pressures above ~40 GPa resulted in the decomposition of biotite whose products were glass, magnetite, and spinel [Badyukov, 1986].
1
2
Grt
QUARTZ Planar elements are observed in quartz (Qtz) at shock pressures of 19–36 GPa. The number of crystallographic directions of their development increases with the shock load. At a pressure of 52 GPa, planar elements in quartz are not observed but numerous rounded cavities arise in quartz grains. This phenomenon is indirect evidence for the melting of the material associated with the escape of fluid phases that could exist as inclusions in the quartz grains. The presence of numerous glasses of mixed composition with a high concentration of SiO2 in the sample also indicates the melting of quartz at this pressure. Shock-induced X-ray diffraction changes in quartz are a drop in the integral intensities and broadening of its diffraction lines that are fixed at pressures of 19 GPa and more. According to XPA data, about 96% of the initial quartz becomes amorphous after the application of a shock load of 52 GPa. Amorphous silica gives a halo of a significant integral intensity in the diffraction spectra. The samples preserved after the shock loading up to 36 and 52 GPa contained small amounts of low temperature SiO2-cristobalite. The high pressure phases of silica (coesite and stishovite) were not discovered in the preserved material. It is likely that coesite does not form in our experiments because of the insufficient duration of loading [Stöffler and Langenhorst, 1994] and the relatively low temperature realized in the cycle of stepwise shock compression. It is known that, under single shock compression, the phase transformation of quartz into coesite (of the diffusive type) is observed only in highly porous samples. Otherwise, the direct transition of quartz into stishovite via the martensite mechanism is observed [Podurets and Trunin, 1987]. The probable reason for the absence of stishovite in the preserved material is its amorphization due to the annealing after the pressure release [Stöffler and Langenhorst, 1994]. We should note that, under natural conditions, high pressure modifications of quartz are observed only in impact glasses that experienced fast quenching as a result of ejection outside a crater. On the whole, the transformations experienced by quartz under the conditions of our IZVESTIYA, PHYSICS OF THE SOLID EARTH
Vol. 42
479
10 µm Fig. 2. Fragment of a crushed and partially molten grain of garnet. The separate garnet microblocks are cemented by bubbly glass of a mixed composition (1). Glass of the garnet composition is observed (2) in the cracks and at the edges of the microblocks. The shock pressure is 52 GPa.
experiments are in agreement with the changes observed in experiments with monomineral samples [Anan’in et al., 1974]. GARNET The SEM examination of schist samples preserved after their shock loading up to pressures lower than 36 GPa revealed only mechanical deformations in garnet (Grt) represented by fracturing, planar elements, and fragmentation with rotation of the grain fragments. The XPA data revealed no changes in the crystal lattice. On shock loading to 36 GPa, a phase close in composition to garnet arose in both peripheral and central regions of garnet grains, in the zones of intense fragmentation. This phase is observed in the form of thin branching veinlets filling cracks, and its color is darker than the surrounding garnet. It is homogeneous and has no mineral characteristics (primarily, boundaries and grain shapes). Based on these specific features, the phase can be considered as glass arising during the melting of garnet. In comparison with the initial garnet, this vitreous phase is depleted in silica and calcium and enriched in potassium and magnesium. On shock loading to 52 GPa, the SEM examination of the preserved material reveals intense melting of garnet, often involving a large part of a grain, but it is evident that, in this case as well, melting was initiated at cracks and grain boundaries and propagated from them (Fig. 2). According to the XPA data, about 35% of garnet became amorphous under a load of 52 GPa. In spherical convergent wave loading experiments with schist of the same type No. 6
2006
480
FEL’DMAN et al.
as the sample studied in our work, partial melting of garnet (along cracks) was observed at a pressure of about 30 GPa and intense melting (more than 90% according to SEM data) was observed at pressures above ~40 GPa [Kozlov et al., 2004a]. CONCLUSIONS In conclusion, we should note that, for the minerals investigated in this work, solid-state transformations at the lattice level (migration of chemical elements and the formation of shock thermal aggregates) were not fixed in experiments with a plane geometry of loading, in contrast to natural impact processes and spherical loading experiments. The melting of minerals under the conditions of our experiments is observed at pressures higher than in the case of natural impact processes and spherical loading experiments. This is due to the fact that, in the cycle of stepwise shock compression, a lower shock temperature is realized (as compared with the other shock loading variants considered above). However, the mechanical deformation patterns obtained at similar shock pressures, on the whole, coincide for all three variants of shock loading mentioned above.
5.
6.
7.
8. 9.
10.
ACKNOWLEDGMENTS This work was carried out within the framework of the program “Physics and Mechanics of Highly Compressed Substance and Problems of the Internal Structure of the Earth and Planets” of the Division of Energetics, Mechanical Engineering, Mechanics, and Control Processes, Russian Academy of Sciences, and was supported by the Russian Foundation for Basic Research, project no. 0.5-05-64496.
11. 12.
13.
REFERENCES 1. A. V. Anan’in, O. N. Breusov, A. N. Dremin, et al., “Shock Loading of Silicon Dioxide,” Fiz. Goreniya Vzryva 10 (3), 426–436 (1974). 2. D. D. Badyukov, “Shock Loading Effects on the Main Types of Rock-Forming Minerals,” Meteoritika, No. 45, 122–130 (1986). 3. V. I. Fel’dman, Petrology of Impactites (MGU, Moscow, 1990) [in Russian]. 4. E. A. Kozlov, L. V. Sazonova, G. V. Kovalenko, and Yu. N. Zhugin, Enstatite Phase Transformations under
14.
15.
16.
Spherical Shock Wave Loads, RFYaTs-VNIITF Preprint, Snezhinsk, 1998, no. 136, p. 42. E. A. Kozlov, L. V. Sazonova, V. I. Fel’dman, et al., “Formation of Ringwoodite during shock Loading of TwoMica Schist from Experimental Data,” Dokl. Akad. Nauk 390 (3), 379–381 (2003). E. A. Kozlov, L. V. Sazonova, and I. V. Belyatinskaya, “Chemical Characteristics of the Garnet Transformation in Shock Waves from Experimental Data,” Elektronnyi Nauch.-Inf. Zh. “Vestnik Otdeleniya nauk o Zemle RAN,” No. 1 (22)' (2004a); http://www.scgis.ru/Russian/cp1251/h_dgggms/1-2004/informbul-1 2004/planet4.pdf. E. A. Kozlov, V. I. Fel’dman, and L. V. Sazonova, “Crystallochemical Structure of Rock-Forming Minerals and Peculiarities, Sequence and Completeness Profile Physico-Chemical Transformations in Weak and Strong Shock Waves,” in Shock Compression of Condensed Matter—2003. AIP Conf. Proc. Ed. by M. D. Furnish, Y. M. Gupta, and J. W. Forbes (New York, 2004b), Vol. 706, pp. 1458–1461. P. Lambert, “Shock-Induced Melting of Biotite and Muscovite,” Meteoritics 14, 466–468 (1979). P. Lambert and I. D. R. Mackinnon, “Micas in Experimental Shocked Gneiss,” J. Geophys. Res. 89, B685– B699 (1984). A. K. Lavrukhina and Z. K. Mil’nikova, “Shock Effects in Meteorites: Observations and Experiment,” Meteoritika, No. 43, 67–82 (1984). Minerals: A Handbook, Ed. by F. V. Chukhrova F.V. et al. (Nauka, Moscow, 1974), Vol. 1 [in Russian]. M. A. Podurets and R. F. Trunin, “Shock Compressibility Features of Silicon Dioxide as Evidence for Kinetics of Phase Transformations,” Fiz. Goreniya Vzryva, No. 1, 98–101 (1987). D. Stöffler, “Deformation and Transformation of RockForming Minerals by Natural and Experimental Shock Process. I. Behavior of Minerals under Shock Compression,” Fortschr. Mineral. 49, 50–113 (1972). D. Stöffler and F. Langenhorst, “Shock Metamorphism of Quartz in Nature and Experiment: I. Basic Observation and Theory,” Meteoritics 29, 155–181 (1994). Ultrahigh-Pressure Mineralogy: Physics and Chemistry of the Earth’s Deep Interior. Mineralogical Society of America Reviews in Mineralogy, Vol. 37, Ed. by R. J. Hemley (Washington, 1998). A. Z. Zhuk, T. I. Borodina, V. V. Milyavskii, and V. E. Fortov, “Shock Wave Synthesis of Carbyne from Graphite,” Dokl. Akad. Nauk 370 (3), 328–331 (2000).
IZVESTIYA, PHYSICS OF THE SOLID EARTH
Vol. 42
No. 6
2006