Transformations of Crystalline Rocks of the Central Uplift and Its Deep-Seated Zones

Abstract

Impact-induced transformations of crystalline rocks of the central uplift and their lateral and vertical gradients are studied in detail due to extensive drilling, particularly in VDW cores down to about 5 km deep. The crystalline rock underwent shock metamorphism, crushing and displacement, injections of impact melt, local post-shock heating and superposed hydrothermal alteration. Based on shock effects in minerals (including appearance of diaplectic minerals and glasses, coesite and impact diamond), the attenuation of shock pressure is established from 40–45 GPa at near-surface zone of the central uplift to 15–20 GPa at a depth of 5 km. Within some linear zones of intense cataclasis, shocked rocks underwent a thermal recrystallization under temperatures up to 900–1100 °C exceeding significantly residual post-shock temperature. This resulted in formation of specific lithologies—coptoblastoliths, which are similar by fabric features to “lunar granulites”, and in emergence of a high-amplitude central magnetic anomaly. At last, hydrothermal alteration caused by the appearance of the fluid circulation in the heated rocks is widespread. Based on petrographic evidences, the thermal evolution of the impact crater has been simulated.

1 Shock Metamorphism

Various mineralogical and petrographic features of the central uplift rock alteration have been studied in the cores of VDW and in other wells, which allowed to estimate main parameters of various processes of transformation. A few general comments need to be made in advance as to the use of various methods of such assessment, especially the parameters of progressive impact metamorphism. To determine the amplitude of shock compression of rocks and minerals, conventional geobarometers were used (e.g. Stöffler 1971; Harrison and Hörz 1981; Basilevsky et al. 1983; Badyukov 1986; Langenhorst and Deutsch 1998; French 1998; French and Koeberl 2010). Various observed deformational features in crystals (dislocations, planar fractures, planar deformational features, mechanical twins, kink bands, mosaicism), as well as different kinds of phase transformations of minerals (formation of high-pressure polymorphs, diaplectic glass, decomposition and melting and evaporation) may be applied in this purpose. The most reliable estimation of the impact compression amplitude may be considered based on of the development of planar deformation features (PDF) in quartz and its phase transitions in non-porous rocks (Langenhorst and Deutsch 1998; Stöffler and Langenhorst 1994; etc.). In addition to the amplitude of the shock wave, it is necessary to bear in mind the influence on these transformations of some other factors, including the initial rock temperature, duration of the compression pulse, reverberation of compression due to reflections, orientation of anisotropic crystals in relation to the shock front and their size, acoustic impedance of the surrounding matrix, the presence of water vapors, etc.

Estimation of the shock compression from the PDF statistics in quartz is reasonable within the limits of quite powerful (many hundreds of meters) concentric zones, on the basis of averaging data on numerous rock samples selected in the massif along the direction of shock wave propagation, as well as taking into account changes in a number of rock-forming minerals. It is necessary to observe the requirements of comparability of local estimates of impact compression amplitude by diaplectic changes with its real gradient on a kilometer scale, neglecting insignificant fluctuations caused by the above-mentioned factors.

To estimate the variability of shock compression by depth and laterally along the radii from the geometric center of the Puchezh-Katunki structure, the maxima of the obtained values are used, based on the critical changes of minerals or newly formed phases, established by petrographic methods, which is a criterion for drawing conditional boundaries between such zones differing in the level of shock-metamorphism. The following shock effects are used as critical: first appearance of PDFs with the orientation $\{ 1\bar{0}13\}$ (ω-system) corresponds to a minimum pressure of 10 GPa, and the orientation $\{ 1\bar{0}12\}$ (π-system) to a pressure in excess to 20 GPa. The occurrence of stishovite indicates a minimum shock pressure of 12–15 GPa, of coesite, of more than 30 GPa, of diaplectic glass, of more than 35 GPa, while shock melting of quartz to yield lechatelierite needs post-shock temperatures that correspond to a shock pressure of more than 50 GPa (e.g., Stöffler and Langenhorst 1994, and references therein).

Other phase transformations, which reliably indicate the level of shock metamorphism, are appearance of maskelynite, i.e. diaplectic plagioclase An_15–35 glass, that indicates shock pressure more than 30 GPa, and formation of plagioclase fusion glass (more than 45 GPa). Conversion of graphite into diamond-lonsdaleite polycrystalline aggregate occurs at shock pressures of 35–45 GPa. Full melting of crystalline rocks takes place at shock loading of about 60 GPa.

An important petrographic indicator correlating with the shock pressure is reduced birefringence of tectosilicates, which is distinguishable at initial stages of shock transformations. It is amplified up to the complete isotropization of minerals with increasing pressure (formation of diaplectic glasses).

A general scheme of radial and concentric zoning of shock transformations based on the mentioned criteria is presented in Table 1 and in Figs. 1 and 2.

Table 1 Principal mineralogical criteria and zoning of shock metamorphism in the Puchezh-Katunki impact structure

Fig. 1

Estimation of shock pressure in the near-surface zone of the central uplift along radii. Vertical coordinate is distance from the crater center (in km); horizontal coordinate, reconstructed shock compression (in GPa). Horizontal lines show pressure limits estimated from the study of PDF in quartz. Numerals are numbers of boreholes, location of which is given in Fig. 2 in Chap. 3

Fig. 2

Estimations of shock pressure values in crystalline target rocks through the VDW section. Vertical coordinate is depth (in km), horizontal coordinate, reconstructed shock compression (in GPa)

Due to superposed post-shock thermal and hydrothermal alterations, shock effects can be significantly obliterated or modified. The barometric estimates can be made on the reconstructed primary traces of compression based on ontogenetic interrelations between new-formed mineral phases. This reconstruction is facilitated by residual post-shock heating and non-monotonous character of superimposed thermal metamorphism. Among its typical features are abrupt temperature fluctuations and caused by them transformations in narrow spatial limits, and besides on the background of rocks of similar composition that have experienced the same shock load.

Shock-metamorphosed changes are established in rocks and minerals for a distance of about 5 km from the Puchezh-Katunki crater center and are traced along the VDW core down to its bottom, i.e. for more than 4.2 km from the top of the central uplift (Fig. 3).

Fig. 3

Distribution of deformed target rocks undergone shock and post-shock thermal metamorphism to a different degree, within the central uplift of the Puchezh-Katunki impact structure. Shock pressure values estimated from mineralogical effects recorded in cores is given for certain numbered boreholes Areas of thermal transformation of crystalline basement rocks are hatched (hatching density reflects the intensity thermal of transformations). In addition, this intensity in the upper part of the authigenic breccia column is shown by differently filled circles for numbered boreholes, in which thermal metamorphic effects were estimated: a—intense, b—moderate, c—weak, d—not appeared. Because of magnetization of rocks is mostly due to post-shock thermal metamorphism, ΔT isogams (in nT) are given. Dot-dashed line shows the crest of the central uplift, dotted line, contour of the central pit. In addition, a section cut across central uplift at A–B line is shown

In the VDW, which locates approximately 1500 m northeast of the geometric center of the crater, fluidal quartz fusion glass (lechatelierite) is found in shocked biotite-amphibole gneisses just beneath of the base of allogenic breccia. Plagioclase is converted into fusion glass, but in some spots maskelynite is preserved. Amphibole is characterized by four systems of planar fractures. One of such systems closely spaced (about 0.01 mm apart) is oriented parallel to the coarse cleavage cracks aligned to (010). Biotite displays several intersecting systems of closely spaced kink bands.

Within a radius of 1300 m from the center of the crater (boreholes 13, 19, 61, 64 etc.), shock fusion plagioclase glasses are observed in gneisses, but the main part of this mineral is transformed into maskelynite. These glasses display fluidal structure or micro-perlitic cracks. In places, the transformation follows only one of the polysynthetic twin systems, which shows the influence of the orientation of shock wave front on transformation of individual crystals. Quartz and microcline are converted into diaplectic glasses; the changes of biotite and amphibole are similar to mentioned above. Samples from boreholes 19, 54, 55 also contain plagioclase fusion glasses and completely amorphized quartz. Due to post-shock annealing, felsic minerals glasses are recrystallized, while mafic minerals are opacitized.

The shock melting of quartz is rarely observed in the uppermost of the authigenic breccia in the VDW section. This makes it possible to assume that in the central zone of the shock compression its amplitude reached 50 GPa. The lower boundary of this zone (about 45 GPa) is determined by the coexistence of diaplectic quartz and microcline glasses, as well as plagioclase fusion glasses, which are traced down to a depth of about 600 m. The state of mafic minerals, which underwent decomposition under high post-shock temperatures confirms this assessment. Relatively rapid quenching of shock-heated rocks directly beneath allogenic breccia leads to the preservation of incompletely decomposed biotite and amphibole in some cases.

At the depth from 600 to about 1800 m in the VDW as well as at a distance between 1300 and 3400 m from the crater center in other wells, the shock pressure can be restored only through relics of diaplectic glasses after quartz and feldspars. Quartz diaplectic glasses are usually preserved in the central parts of grains, but at peripheral parts they are recrystallized in quartz with monocrystal structure and palimpsest traces of planar deformation elements. Similar features are observed in boreholes S-3, 18, 40 etc. In places, cryptocrystalline finger-like and kidney-like aggregates of fine-grained quartz are recorded in recrystallized diaplectic glass; these are products of coesite inversion (borehole 13). Both in the depth interval of 1520–1800 m in VDW and in boreholes S-3, 7, 13, 18, and 40, some sections of plagioclase diaplectic glass in feldspar grains with partially recovered by annealing crystalline structure with cloudy extinction are recorded. In addition, kink bands in biotite (Fig. 4) as well as two or three systems of planar microstructures (including shock twins) and some reduction of birefringence in hornblende occur.

Fig. 4

A biotite grain with kinkbands (VDW, depth 2001 m) in a shocked biotite gneiss. All quartz grains are characterized by multiple PDFs. Photomicrograph, plane-polarized light

In a sample of biotite-amphibole gneiss from the depth of 607 m in the VDW section, some impact diamonds were found, they were revealed together with numerous graphite scales. This fact indicates a lower limit of shock compression of about 35 GPa. The lack of fusion feldspar glasses constrains the upper limit of shock compression by 45 GPa.

Between 1800 and 3300 m in the VDW section and at a radius of 3400–5000 m laterally (boreholes 40, S-3, 42 etc.), the following shock transformations occur: within annealed quartz with recovered monocrystal texture, sections with lower refractive index and systems of decorated planar elements are preserved (Fig. 5). Both recrystallized and relict domains within quartz grains do not differ in extinction; this indicates the preservation of optical orientation in the recrystallized material. This feature makes it possible to determine the orientation of planar deformation features relative to the optical axis if their traces are present. The average number of relict systems per one grain varies from four (in the depth interval of 1670–2106 m) to three (2257–2420 m). In both cases PDFs orientated along ω- and π-rhombohedrons are predominant (Fig. 6). Other PDF planes are also observed, but they are subordinate. It is possible that some of these PDF planes could have been destroyed by the annealing and recrystallization of shocked rocks.

Fig. 5

Three systems of planar deformation elements in quartz (VDW, depth 1711 m) in a leucocratic biotite-plagioclase gneiss. Photomicrograph, cross-polarized light

Fig. 6

Number of PDF systems (I) and PDF orientation histograms (II) of PDF in quartz grains for different zones through the VDW section. n and n₁ are number of measured samples. At vertical axis, parts of total number of measured grains in % (I) and measured orientations of PDFs in % (II). On the x-axis, the amount of different PDF systems per quartz grain (I), and the angle between the quartz c-axis and the pole to the planar feature is plotted. Y-axis indicates a percentage of total number of measured grains (I) and frequency for each given angle

In plagioclase, some of the twin lamellae are converted to maskelynite while others have a reduced birefringence and characteristic intersecting PDE systems that are orientated relative to the twin plane (Fig. 7). In microcline, systems of thin parallel planar fractures oblique to the microcline lattice are recorded. In mafic minerals, deformation feature systems differing by orientation are observed as well. The occurrence in the considered zone of diaplectic quartz with a highly reduced birefringence, maskelynite, and numerous PDE systems in all minerals allows to estimate the shock pressure between 35 and 25 GPa (Fig. 6).

Fig. 7

Photomicrograph of the plagioclase, in which one of twins systems transformed into maskelynite with traces of planar deformation (VDW, depth 3781 m). cross-polarized light, ×150

Downward in the VDW section to a depth of about 4500 m, diaplectic transformations in the minerals also occur, but their intensity is decreasing. Relict areas with reduced birefringence occur rarely in quartz grains. The average number of PDFs per a quartz grain is three like in the above depth interval, but their total number in the depth range of 3353–3491 m decreases. Predominant PDF orientations are the same—ω- and π-rhombohedrons, but a small peak corresponding to the direction $\{ 11\bar{2}2\}$ appears on the histogram (Fig. 6). Isotropic feldspars are very rare. Lamellae of multiple twins in a feldspar grain show a various degree of lowering of birefringence and refractive indices. At depths of 3900 m and more, only some twin lamellae in plagioclase (in places, in microcline) have a reduced refractive indice, but systems of planar fractures are preserved (Fig. 7). If to total, the presence of plagioclase with the mentioned optical characteristics, the predominance of ω- and -π orientations of PDFs in quartz, the presence of PDFs in feldspars and biotite (Fig. 8) suggest that the shock pressure at the depth from 3300 down to 4500 m was at least 20 GPa (Table 1).

Fig. 8

A biotite scale with two systems of planar deformational bands. VDW, depth 3113 m; the field size is 0.6 mm wide. Photomicrograph, cross-polarized light

In crystalline rocks occurring below 4500 m, the total amount of quartz grains with PDF decreases sharply; the average number of PDF systems per grain reduces to be 2. The percentage of PDFs along with ω-rhombohedron increases sharply, while the share of the π-system decreases significantly. Planar fractures, which are parallel to the basal plane {0001} become more abundant. Plagioclase and microcline are characterized by selective reduction of birefringence and, in places, by short fractures oblique to the twin planes. These transformations expressed by a noticeable optical effect are due to the preferred compression of feldspar crystal lattices along the directions [100] and [010] in comparison with [001] (Dworak 1969; Ostertag 1983). For plagioclase, this effect is provided by a shock pressure from 18 to 30 GPa, and for microcline, from 15 to 27 GPa. Since the marked effects in quartz are realized under the shock pressure less than 20 GPa, the shock compression at this depth interval may be estimated at 15–20 GPa.

Thus, the attenuation of shock deformation of crystalline rocks in the central uplift both downwards and outwards from the center is clearly established. An extrapolation of the shock compression attenuation downward in the section shows that the complete disappearance of shock features is expected at a depth from 8 to 10 km down from the upper boundary of the central uplift. This estimated attenuation of shock effects with depth is well correlated with the decreasing degree of crushing and cataclasis of crystalline rocks, as well as with a gradual increase in their density.

2 Thermal Metamorphism

The thermal metamorphism of crystalline rocks of the central uplift occurred after passing of shock wave. It resulted in recrystallization of minerals, in their decomposition, and in some cases in melting and remobilization. Simultaneously, significant changes in the magnetic properties of the rocks occurred there. As early as during initial core study in the process of deepening of the VDW it was found that thermal transformations of the rocks at many places occurred at temperatures significantly exceeding the post-shock temperatures estimated from the respective levels of shock pressure (Masaitis and Mashchak 1990). These data initiated a detailed study of cores from other boreholes penetrating the central uplift as well as experiments on annealing of shock-metamorphosed rocks from the Puchezh-Katunki impact structure.

The intervals where rocks display intense thermal transformations were found both in the pilot well and in the VDW as well as in many other boreholes penetrating into deformed crystalline rocks of the central uplift (Fig. 3). The areal of intensely thermally-metamorphosed rocks is confined with the area of the central pit, weak and, rarely, moderate thermal transformations are recorded in places in rocks from the crest and outer slope of the central uplift. Inferred from distribution of depth intervals of the most intense thermal metamorphism in VDW and the pilot well, these intervals may be interpreted as intersections of a single zone, which is steeply sinking and splintered with depth. The zone is to be probably several tens of meters thick.

The scheme proposed by Stöffler (1972, 1984) is commonly used for the estimation of post-shock temperatures in non-porous rocks; the temperatures depend mainly on the shock pressure. As it was shown above, the shock pressure in the central uplift of the Puchezh-Katunki exceeded 45 GPa; it is quite probably that it was slightly higher. Hence, the residual temperature there could reach 900 °C at a depth about 550–600 m in the VDW. In the peripheral part of the shock transformation zone where the pressure did not exceed 35 GPa, the post-shock temperature is estimated to about 300 °C (at a depth about 1800 m). Nevertheless, in this zone as well as in the surrounding zone, where the pressure was ~25–35 GPa and the residual temperatures were less than 200–300 °C, various mineralogical features indicate thermal transformations at much higher temperatures than post-shock temperatures given above.

The general scheme of attenuation of post-shock temperature with depth based on estimation of corresponding shock pressure together with the assessment of annealing temperatures of pyrometamorphic transformations, is given in Fig. 9. In addition, the rocks were at some elevated temperature at the moment of the impact event due to the geothermal gradient. With this correction, the curve of post-shock temperature should be displaced to the right in Fig. 9 by 120–150 °C, particularly in the lower part.

Fig. 9

Estimated temperature of pyrometamorphic transformations through the VDW section. Vertical axis, depth in km; horizontal axis, temperature values. Solid curves show: 1—attenuation of post-shock temperature evaluated base on shock pressure estimates (see Sect. 1); 2—attenuation of the same post-shock temperature with correction to existed geothermal gradient. Horizontal lines show temperature values estimated from petrographic study of core samples (see text)

The temperature regime of annealing is estimated from comparison of the observable mineralogical changes in rocks of the central uplift with experimental data at pressures of P < 0.1 GPa that neglected an influence of a steam fluid (Tchukhrov and Bonstedt-Kupletskaya 1965, 1981, 1983; Deer et al. 1978, 1986, 2001). The presence of water as well as the disorder of crystalline structure due to impulse compression would expect a lower temperature of transformations. Therefore, it is reasonable to assume the lower values of the experimentally established temperature intervals.

An appearance of superimposed thermal metamorphism is definitely established in the cases where the observed mineralogical effects cannot be attributed to the residual heat because temperature estimate exceeds considerably an expected post-shock temperature.

In order to clarify the physical parameters of post-shock thermal effects taking into account their imposition on the diaplectic minerals and diaplectic glasses, a series of annealing experiments was performed with biotite-amphibole gneisses naturally shocked up to 45 GPa. The gneisses were sampled from wells drilled within the central uplift. An unshocked gneiss from the Vladimirskoe-2 well situated 70 km northeast of the central uplift (see Fig. 1 in Chap. 2), was used as a reference. Each sample was sawn into 5 pieces, and four of them were heated in a muffle furnace to 500, 750, 950 and 1050 °C and annealed at these temperatures for 30 min with free access of air.

Petrographic study of the naturally shocked to a various degree and subsequently experimentally annealed rocks shows that biotite and amphibole due to iron oxidation first acquire irregular spotty extinction, turn red, become opacitized at the grain edges, and under maximum heating (950 °C), they turn opaque. In gneisses naturally shocked up to 20 GPa, any thermal alteration in quartz and plagioclase were not observed even after annealing at 1050 °C. In rocks compressed up to 25 GPa and higher, planar elements are partially obliterated first in feldspars at 950 °C and then in quartz at 1050 °C. In samples shocked at pressure between 35 and 40 GPa, planar elements in quartz vanish after annealing at 950 °C. The quartz attains spotted extinction and higher birefringence (up to 0.016). The coexisting feldspars are melted in some places after experimental annealing at 1050°. In gneisses shocked at pressures 40–45 GPa, maskelynite is melted at the annealing temperature of 950 °C. In places, it recrystallizes with a fine granoblastic texture. In respective samples annealed at 1050 °C, porous polymineralic glasses with relics of altered plagioclase and quartz appear together with small spots of recrystallized feldspar.

Melanocratic rocks naturally shocked at lower pressure, are more stable during experimental annealing than shocked leucocratic gneisses.

A general scheme of pyrometamorphic changes in diaplectic minerals and glasses and an estimation of respective temperatures are given in Table 2. This scheme was compiled on the basis of petrographic and mineralogical data for thermally transformed rocks of the central uplift. Thermal effects of quartz begin at a temperature of about 573 °C (the inversion temperature of α to β quartz), but they are not reflected in optical properties of newly formed quartz aggregates. Experiments showed that the PDFs disappear when annealed to 600 °C in an atmosphere of steam containing ions of sodium and carbon dioxide (Feldman 1990), but they are stable in dry experimental conditions at 950 °C (Bunch et al. 1968); this was confirmed by our experiments.

Table 2 Pyrometamorphic transformations of diaplectic minerals and diaplectic glasses

Quartz formed by recrystallization of dialectic quartz glass in the rocks from VDW and other wells at temperatures about 600 °C shows a higher birefringence (yellow color interference), and a brown color in transmitted light due to optical dispersion on the boundary between glass and newly formed crystallites. At higher temperature (600–800 °C), spots of micro-granular or globular texture appear in quartz inverted from diaplectic glass (Fig. 10). Metacristobalite, which is a metastable cryptocrystalline modification of β-cristobalite in combination with an amorphous phase, crystallizes from diaplectic quartz and diaplectic glass at a more significant heating. The inversion points for quartz crystal are ca. 1200 °C, and for silica gel, ca. 900 °C (Tchukhrov and Bonstedt-Kupletskaya 1965, 1981). Respective transition temperatures for both diaplectic quartz and, especially, for diaplectic glass are closer to the lower value, but exceed it.

Fig. 10

Thermally-transformed shocked biotite-plagioclase gneiss. In the centre of the image, a partly recrystallized quartz with domains of globular texture and relics of planar deformational elements. The bulk consists of maskelynite (dark-colored) with linearly-orientated isotropized biotite flakes. Photomicrograph, borehole 61, depth 283 m; cross-polarized light

More intensive (t > 950°) annealing of quartz and various products of its transformation is established on a basis of zonal distribution of silica modifications in primary grain boundaries and of tridymite occurrence that is caused not only by high temperature value, but also by rate of its growth (Sinelnikov 1956; Ziegler et al. 1988). A rapid rise of the temperature into the field of tridymite stability (870–1470 °C at 1 atm) results at the beginning in crystallization of metacristobalite rather than tridymite. Then, a stable form of high-temperature cristobalite crystallizes forming needles, prisms, idioblasts, or microspherulites. Finally, at temperature in excess to 1000–1200 °C tridymite with characteristic spear-shaped twins appears in the rocks (Fig. 11). In the center of grains of recrystallized zonal quartz, the transition is completed at the stage of cristobalite formation as a result of the short-term action of the peak temperature. A slower temperature decreasing frequently yields an inversion into quartz with inherited shapes of the replaced phases: lanceolate at the margin zone and an aggregate of isometric rounded individuals in the center of the grain.

Fig. 11

Lanceolate tridymite twins and granoblastic quartz after diaplectic quartz glass within thermally-recrystallized shocked leucocratic gneiss. Black—relics of diaplectic quartz glass. The groundmass consists of granoblastic quartz-feldspar aggregate. Photomicrograph, VDW, depth 1469.5 m; cross-polarized light

The recrystallization of diaplectic plagioclase and diaplectic plagioclase glass (maskelynite) apparently starts at a temperature of about 600 °C, but an appreciable effect of recrystallization of maskelynite is observed only when upon heated to 800 °C (Bunch et al. 1968; Ostertag 1983). At temperatures of 800–950 °C, diaplectic plagioclase and diaplectic plagioclase glass turn into the so-called “white ceramics”, which appears brown and dark brown in transmitted light. In places, there are signs of melting along one of the twin systems or at grain edges. X-ray diffraction analysis of completely opaque glassy grains shows that they have a crystalline structure with the ordering index of Ι ≤ 50 and Ι ≤ 70 (samples from the VDW, depths 1575 and 1747 m).

Recrystallization of maskelynite and “white ceramics” results in formation of micro-laths aggregates up to a 0.1 mm across with finely crystalline, in places axiolitic and spherulitic textures (Fig. 12). Maskelynite recrystallization products usually have a more basic composition than the original plagioclase.

Fig. 12

A finely crystalline aggregate of plagioclase microlites after maskelynite. At right bottom and top, tridymite-quartz aggregate after diaplectic quartz glass are observed. The groundmass is partly replaced by saponite. Photomicrograph, VDW, depth 2524 m; cross-polarized light

The beginning of opacitization of biotite and the change of optical characteristics of hornblende can be attributed to the temperature range of 600–800 °C. Biotite is characterized by narrow opacite rims, while amphibole, by deepening of colour at grain edges. The effect of change in the colour of amphibole is observed during annealing at temperatures of 500–800 °C (Tchukhrov and Bonstedt-Kupletskaya 1983). At higher temperatures, brown and red tints begin to predominate in amphibole and biotite; the higher the level of preceding shock transformation, the more intense the colour appears. The birefringence in hornblende increases to 0.040–0.050. The edges of grains represent optically an opaque cryptocrystalline aggregate, in which monoclinic pyroxene and hematite are recorded by XRD.

A cryptocrystalline opaque aggregate of clinopyroxene and maghemite originate after common hornblende in annealing experiments at temperatures of 1000–1100 °C (Tchukhrov and Bonstedt-Kupletskaya 1983), but for diaplectic amphibole these temperatures may be lower. In the rocks from the VDW, this opaque aggregate consists of clinopyroxene, plagioclase, and iron ore (magnetite, hematite). The clinopyroxene often forms rims of relatively larger microlites around decomposed amphibole (Fig. 13). Their compositions on the polythermal diagram for P = 1 atm (Lindsley 1983) are located close to the isotherm 1000 °C.

Fig. 13

Opaque aggregate of neogenetic minerals after hornblende (black) with rims of fine (<0.01 mm) clinopyroxene (salite) microlites (at right). The opaque aggregate consists of clinopyroxene, plagioclase, magnetite, and hematite. The rock is thermally-recrystallized shocked amphibole gneiss. The texture of the rock is demonstrated at the left side of the image. It consists mainly of feldspar-dominated granoblastic aggregate with quartz-tridymite nests after diaplectic quartz glass. Photomicrograph, VDW, depth 1400 m; cross-polarized (left) and plane-polarized (right) light

Experimental data show that the decomposition of biotite (with an iron content close to biotite from gneisses of VDW) occurs at pressures below 1 kbar at temperatures above 700–800 °C. Biotite is replaced by an aggregate of titanomagnetite, potassium feldspar, more magnesium mica, and occasionally plagioclase. Thus, both input and output of some chemical components takes place during this transformation.

The formation of dusty titanomagnetite and magnetite during the decomposition of femic minerals in annealed rocks resulted in sharp increasing of their magnetic susceptibility, which are recorded by a positive magnetic anomaly (with an amplitude up to 1800 nT) of almost rounded shape and ca. 4 km in diameter in plan. The nature of the magnetic field may indicate that the magnetic disturbing body (in fact, the zone of intensely thermally-metamorphosed rocks) is steeply dipping to SSW; its exposure under the crater fill corresponds in outline to contours of the central pit (Fig. 3). At the crest of the central uplift, natural remanent magnetization values fall sharply, local low-intensity anomalies being recorded at the outer slope of the uplift, though.

The lower stability limit for the almandine garnet at atmospheric pressure is 785 °C (Tchukhrov and Bonstedt-Kupletskaya 1983). In the experiment on thermal stability for a shocked almandine (Gnevushev et al. 1982), a weak exothermic effect caused by oxidation of the ferrous iron was observed at 800–1000 °C. This effect precedes the decomposition of the garnet into a cryptocrystalline aggregate of newly formed minerals including pyroxene and plagioclase.

In individual cases, the high temperatures yield small amounts of melt, which forms at grain boundaries glass films enriched in alkalis. Clinopyroxene similar in composition to that formed at the expense of amphibole, occasionally crystallizes from this melt.

The given data show that assemblages of anhydrous relatively high-temperature minerals are formed as a result of thermal decomposition of hydrous silicates in studied shock-metamorphosed rocks. At the same time, the vapor occurring in the system facilitates easier recrystallization of diaplectic minerals, ensures the transfer of chemical components and yields blasthesis. At final stages, small amounts of newly formed mica and amphibole appear. As it documented by microprobe analyses, an intense re-distribution of alkalis, calcium, magnesium, and, to a lesser extent, silica takes place during this process.

The high degree of post-shock thermal transformations almost completely erases many of the rock features, first of all the texture, and then the mineral composition. The bulk chemical composition remains almost unchanged, and some textural features may be preserved in places. Such thermally transformed shock-metamorphosed rocks, which usually occur in the areas of intensive cataclasis and crushing, have been named coptoblastoliths (from Greek κοπτο—to destroy by blows and βλαστεσισ—shoot) (Masaitis and Mashchak 1996). The appearance of this lithology depends on both the composition of initial rocks and the degree of their preceding shock and thermal (and in places superposed hydrothermal) transformations. Coptoblastoliths are developing mostly at zones where shocked gneisses or amphibolites underwent intense cataclasis and crushing. Due to thermal influence, either massive or porous, fine-grained, dark-colored rocks arisen in these zones; they are eutaxitic frequently (Fig. 14). Coptoblastoliths form irregularly-shaped parts with vague contacts within thermally-transformed shocked rocks. They contribute no more than 5 vol.% to the total length of zones of intense thermal transformation that are recorded mainly in depth intervals of 1000–1100, 1250–1300, 1425–1498, 1830–1910, and 2400–2600 m of the VDW (Masaitis and Mashchak 1996).

Fig. 14

Intensely thermally-metamorphosed and cataclased biotite-amphibole gneiss changing to completely recrystallized rock (coptoblastolith, dark parts). In the latter, porphyroclasts of recrystallized quartz and feldspar are abundant within a dark fine-grained groundmass. The rock has an eutaxitic structure. VDW, depth of 1000 m

The most developed are coptoblastoliths formed at the expense of essentially quartz-feldspathic rocks (Fig. 15); coptoblastoliths after amphibolites and schists as well as after rocks of intermediate composition (mesocratic plagioclase-dominated gneisses), also occur (Fig. 16). All these usually have a spotted appearance due to the presence of relic minerals of original rocks and tridymite porphyroblasts (Fig. 17), which compose about a half of the coptoblastoliths volume. In places where the cataclasis was less intense, relics of gneissic structure can be preserved. The predominant texture of coptoblastoliths is micro-granoblastic combined with micro-ophitic (Fig. 18), ophitic-taxitic, and for rocks formed at the expense of quartz-feldspathic varieties also heteroblastic or felsitic ones (Fig. 15).

Fig. 15

Coptoblastolith after a leucocratic lithology (granite gneiss). The rock is spotted due to numerous evenly distributed quartz-tridymite glomeroblastic accumulations and rare shadow relict minerals within an inequigranular quartz-feldspar groundmass with heteroblastic texture. The groundmass is partly replaced by saponite, which also fills in small pores. Photomicrograph, VDW, depth 1869 m; cross-polarized light

Fig. 16

Coptoblastolith after a mesocratic biotite-amphibole gneiss. Texture is micro-ophitic, formed by plagioclase laths and isometric clinopyroxene microlites. Irregular light spots are granoblastic quartz and feldspar aggregates. Photomicrograph, VDW, depth 1105 m; plane-polarized light

Fig. 17

Coptoblastolith after a migmatized biotite-amphibole gneiss. The rock consists of fine-grained aggregate of feldspars and clinopyroxene microlites hosting tridymite porphyroblasts. Photomicrograph, VDW, depth 1467 m; cross-polarized light

Fig. 18

Ternary Wo-Fs-En diagram for clinopyroxenes from rocks of the central uplift of the Puchezh-Katunki impact structure. Dots for pyroxenes from coptoblastoliths, tagamites, post-impact hydrothermal associations as well as dots of mean composition for pyroxenes from target crystalline rocks (see Masaitis and Pevzner 1999), tagamites and coptoblastoliths are plotted. EMP analyses of pyroxenes are given in Tables 5 (in Chap. 3), 3, and 8

The most widespread minerals of coptoblastoliths formed during annealing are clinopyroxene, plagioclase, alkali feldspars, tridymite, titanomagnetite, magnetite, ilmenite, and phlogopite. Clinopyroxene appears as rounded or, less often, short prismatic grains. Its composition it corresponds to salite (Table 3). In this aspect, it differs significantly from pyroxenes from gneisses and amphibolites, as well as from pyroxenes in tagamites (Fig. 18). Plagioclase forms elongated (0.02 × 0.15 mm) sometimes twinned laths of basic oligoclase, or, less often, of acidic andesine composition. Plagioclase in coptoblastoliths after amphibolites is a basic andesine (Table 4). In general, it is more basic than that in the precursor rocks, but it is more acidic than plagioclase in tagamites, which has a higher potassium content (Fig. 19). Phlogopite occurs as short prismatic plates or, less frequently, as flakes. Its composition is even more variable than that of pyroxene and plagioclase, with the Fe/Mg ratio ranging from 1:3 to 1:20 (Table 5).

Table 3 Selected electron microprobe analyses of clinopyroxenes from coptoblastoliths

Table 4 Selected electron microprobe analyses of feldspars from coptoblastoliths

Fig. 19

Ternary An-Ab-Ort diagram for feldspars from rocks of the central uplift of the Puchezh-Katunki impact structure. In addition to feldspars from coptoblastoliths, mean values for plagioclases from crystalline target rocks (see Masaitis and Pevzner 1999), coptoblastoliths, and tagamites are shown. EMP analyses of feldspars are given in Tables 5 (in Chap. 3) and 4

Table 5 Selected electron microprobe analyses of phlogopite and ore minerals from coptoblastoliths

The mentioned ore minerals demonstrate a spotted distribution and form a dust-like impregnation with particles of 0.001 mm across or isometric segregations of 0.01 mm across, but in places, up to 0.1 mm. Compositions of magnetite and titanomagnetite are given in Table 5.

Alkali feldspars fill the interstices between plagioclase laths and they are also preserved as recrystallized relicts of primary minerals. These relicts are characterized by broom-like and spherulite textures. Feldspars, which fill interstices have micro-granoblastic, radial, lath-like and other textures. The composition of K-feldspars is variable, orthoclase and anorthoclase are present (Table 4). Tridymite occurs as spear-shaped twins within the spotted areas of granoblastic quartz; its presence is particularly typical for leucocratic coptoblastoliths (Fig. 17).

The chemical composition of coptoblastoliths is generally similar to that of the precursor crystalline rocks. The data points for coptoblastoliths plot in the AFM ternary diagram are located in the same field as the points of shocked crystalline rocks that host the zone of coptoblastoliths (see Fig. 36 in Chap. 3). As noted above, coptoblastoliths show signs of initial selective melting. This melt is relatively more acid and alkaline, and enriched in volatiles. Thin veinlets of such granophyre (up to 2–3 cm thick) with sharp contacts are found in coptoblastoliths at depths of 1380 and 1460 m in the VDW.

These granophyres have spotted and fluidal texture, they contain irregular and lenticular voids filled with heulandite and chalcedony. The groundmass is fine-crystalline, prismatic-grained, in some places spherulitic, and composed of pyroxene (Fs₁₉En₄₀Wo₄₁), alkali feldspar and tridymite. Compared to host shock- and thermally-metamorphosed gneisses, granophyres are distinguished by higher contents of silica and alkalis, and lower contents of femic components (Table 6). The granophyres may be considered as rheomorphic injections.

Table 6 Whole rock analyses (wt%) of a granophyre and host coptoblastolith after biotite-amphibole gneiss (VDW, depth 1380 m)

The pressure–temperature (PT) conditions of the coptoblastoliths formation are similar to those of the pyroxene-hornfels facies of contact metamorphism for metapelite, quartz-feldspathic and basic rocks. Maximum temperatures of metamorphism of this facies at low pressures are up to 1100 °C. Rocks similar to those found in the Puchezh-Katunki have been found in central uplifts of the Boltysh and Terny impact structures in the Ukraine (Masaitis et al. 1980; Masaitis and Mashchak 1994), they also may be considered as coptoblastoliths (Masaitis and Mashchak 1994). Peculiar annealed crystalline rocks (so-called leucogranofelses) in the central uplift of the Vredefort impact structure (Schreier 1983; Hart et al. 1995; Gibson et al. 1998, and others) also resemble the above described coptoblastoliths by many features. The annealing temperatures of these rocks are estimated at 900–1000 °C, which are also significantly higher than the residual post-shock temperatures there. In some cases, the annealing resulted in partial melting and remobilization of a relatively leucocratic melt there.

The coptoblastoliths of the deep sub-crater zones of the Puchezh-Katunki and some other impact structures may be correlated with the so-called lunar granulites in a number of structural and textural features and reconstructed conditions of transformation of initial rocks (Masaitis and Mashchak 1995). The lunar granulites are recrystallized polymict or monomict impact breccias with a granoblastic texture formed after ferroan anorthosites, norites and troctolites. Their origin is disputed (e.g., Lindström and Lindström 1986; McGee 1987, 1989; James et al. 1989, and others). The most likely that the formation of coptoblastoliths as well as lunar granulites is stimulated by heating caused by intense shifting stresses accompanied by differential movements under conditions of unloading after pressure decay. Differences between the two types of recrystallized impact breccias—terrestrial and lunar—result mainly from differences in the composition of their precursor rocks, in their fluid content, and in the repeated character of impacts on the Moon. The inferred original geological position of lunar granulite samples may have been the same as that of their terrestrial analogues.

One more specific appearance of thermal transformation of shocked crystalline rocks in the Puchezh-Katunki should be noted. This is a near-surface oxidation, which took place prior to the formation of hydrothermal veins. The oxidation is appearing in the annular crest of the central uplift, i.e. in the most elevated part of the latter. From drilling data, the depth of the zone of oxidation reaches 200 m. It is manifested by ochre colour of shocked and brecciated crystalline rocks due to formation of hematite and limonite after mafic minerals and magnetite. It is quite possible that hematite and limonite developed in the most elevated part of the authigenic breccia as a result of interaction of the rocks heated up to 200–400 °C and atmospheric oxygen immediately following the origin of the central uplift, inasmuch as this part of the uplift was elevated to the surface.

3 Hydrothermal Alteration

Evidence of impact-induced hydrothermal activity includes both occurrence of secondary mineral assemblages (including ore-forming minerals) and chemical alteration of materials at the impact site. Due to drilling, secondary mineralization can be traced in the Vorotilovo Deep Well to 5.3 km depth, as well as laterally in the area of the central uplift. The degree of hydrothermal alteration of rocks depends on both their permeability and degree of preceding thermal recrystallization; Hydrothermal minerals commonly contribute no more than 1–2% to the rock volume but in places, they peak 30 vol.% in coptoblastoliths. Hydrothermal alterations become less pronounced downwards and outwards reflecting the rather regular decrease in rock porosity. Thus, noticeable appearance of hydrothermal alteration minerals does not extend beyond the central uplift of the Puchezh-Katunki astrobleme.

The Puchezh-Katunki impact structure is stand out for its diversity of hydrothermal minerals. The list of the latter includes smectites, chlorite, illite, calcite, sulfides, zeolites, anhydrite, gypsum, apophyllite, quartz, opal, prehnite, epidote, andradite, ferrosalite, actinolite, and albite. The spatial distribution of the hydrothermal mineralization is characterized by both vertical and lateral zonation. The vertical zoning is most completely represented in the middle part of the central uplift, where it was studied in the VDW section. There, two alteration zones, named after the main minerals that occur in these zones, can be distinguished: the upper smectite-zeolite zone and the lower chlorite-anhydrite zone (Fig. 20).

Fig. 20

Distribution of main hydrothermal minerals in the Vorotilovo Deep Well section. The intensity of thermal metamorphism is shown based on data from Sect. 2 in arbitrary units by rectangle areas for 50-m depth intervals. The intensity of hydrothermal alteration is in arbitrary units, reflecting the content of secondary minerals calculated for 15-m depth intervals from thin section study: 1, <0.5%; 2, 0.5–2%; 3, 2–4%; 4, >4%

The smectite-zeolite zone comprises layers of coptomict deposits, suevites, and polymict allogenic breccia, plus the shocked and brecciated target rocks to 2550–2600 m depth. The compositions of main minerals of the zone—both smectites and, especially, zeolites vary regularly with depth (Table 7).

Table 7 Representative EMP analyses of main minerals from the smectite-zeolite zone

According to DTA, XRD, and microprobe data, montmorillonite, Mg–Fe-saponite, and nontronite are distinguished among the smectites (Naumov 2002). Montmorillonite (b₀ = 8.98 Å) replaces impact glass fragments in suevites and redeposited suevites, and fills locally interstices in the authigenic breccia and in tagamites in peripheral parts of the central uplift. Mg–Fe-saponite (b₀ = 9.18–9.22 Å) replaces ubiquitously shocked and thermally transformed rock-forming minerals in the authigenic breccia and is developed after the matrix and in tagamite interstices. With depth, the Al^IV content increases in the saponites. Nontronite in association with calcite occurs as thin veinlets cutting saponitized rocks. For smectites, this variation appears in the progressive substitution of dioctahedral high-alu-mina varieties by trioctahedral Mg–Fe varieties (saponites) downward in the section, with the content of tetrahedral alumina increasing with depth in the saponites. Thus, a general compositional variation for smectites appears in substitution of dioctahedral Al-montmorillonite by trioctahedral Mg–Fe varieties (saponites) downward in the section, the content of tetrahedral aluminia increasing with depth in these saponites. Finally, at the depths of 2500–2600 m saponites are gradually replaced by chlorites (Fig. 20). So, the boundary between the smectite-zeolite, and the chlorite-anhydrite zone is drawn at this level. XRD data failed to reveal the presence of chlorite-smectite (mixed-layered phases in this interval.

The zeolites form numerous veins up to 10 cm thick in the authigenic and polymict allogenic breccias and suevites and locally, form the cement of coptomict gritstones; they associate with three generations of calcite, and with pyrite, apophyllite, and gypsum. Among zeolites, heulandite, chabazite, analcime, stilbite, laumontite, clinoptilolite, erionite, phillipsite, harmotome, and scolecite are identified by EMP, XRD, and DTA analyses (Naumov 2001).

Both a vertical zonation and some lateral inhomogeneity of the distribution of zeolites is noted. On the whole, the Ca/Na and Al/Si ratios in zeolites increase downwards in the section (Table 7, Fig. 21), so that some zeolite subzones can be distinguished, from top to bottom (Fig. 20):

Fig. 21

Diagram of main components for chemical compositions of zeolites from different parts of a generalized vertical section through the central uplift of the Puchezh-Katunki impact crater. Results of 58 X-ray fluorescence spectroscopy and electron microprobe analyses are used. Representative chemical data of zeolites are given elsewhere (Naumov 1993, 2002). The data show a general increase of Ca and Al contents in zeolite compositions downward in the vertical section as well as the rating of zeolites from peripheral segments of the central uplift to Na zeolite–calcite subzone

• (a) The Na zeolites-calcite subzone embraces redeposited suevites, and, locally, underlying suevite layer. Zeolites are represented by Na-heulandite, Na-chabazite, and, in some locations, erionite and clinoptilolite. Together with calcite, they form the cement of the redeposited suevites and suevites within the central pit and the disseminated mineralization in these rocks within the annular depression. Among clay minerals, which replace impact glass, montmorillonite dominates over saponite.

• (b) The Na–Ca zeolites—calcite subzone contains suevites, polymict allogenic breccia, and shocked and brecciated basement rocks to 550–850 m depth. In this subzone, zeolite-bearing assemblages vary laterally, that indicates a significant chemical variability of the mineral-forming solutions. Common are heulandite-calcite, analcime-calcite, chabazite-analcime-(heulandite)-calcite, stilbite-analcime, apophyllite-analcime-heulandite-gypsum, apophyllite-erionite-heulandite associations. Just within this subzone, the radial variation of the zeolite distribution is revealed (Naumov 2002). It consists in both an alternation of areas with high-silica zeolites (heulandite, erionite) and low-silica zeolites (chabazite, analcime), and in increase of the Si and Na contents in chabazite, a zeolite occupying only a thin layer (no more 100 m thick) at the top of the authigenic breccia, outward from the crater center (Naumov 1993). Clay minerals are mainly represented by saponite within this subzone.

• (c) The Ca zeolites subzone is penetrated by only VDW core in the depth interval 850–1800 m. Laumontite and Ca-heulandite, with minor apophyllite, chabazite, scolecite, and quartz occur there.

• (d) The Ca zeolites—anhydrite subzone (the depth interval 1800–2550 m) is distinguished by the appearance of anhydrite, together with laumontite and pyrite, as a common secondary mineral.

The chlorite-anhydrite zone comprises brecciated and shocked basement rocks at depths more than 2550 m. Common are chlorite, anhydrite, and pyrite, and locally (mainly in the depth interval 2900–3500 m), relatively high-temperature Ca–Fe-silicates (andradite, ferrosalite, epidote, prehnite, actinolite), and, in some cases, quartz also occur. The representative chemical analyses of main minerals from this zone are given in Table 8. Zeolites are observed down to 3854 m and are represented by Ca-varieties (laumontite with minor heulandite) only in this zone. Below 4200 m, the hydrothermal mineralization is rarely to be found and represented by thin (no more than 1–2 mm thick) veinlets.

Table 8 Representative EMP analyses of main minerals from the chlorite-anhydrite zone (VDW section)

Along the periphery of the central uplift, the vertical zoning is much less appeared. In places, a superposition of characteristic minerals from different zones (e.g., zeolites, smectites and epidote, actinolite) is recorded there. In addition, a reverse zoning is observed in some boreholes where chlorites replaces smectites downward in the section, whereas zoning of zeolite distribution fits in general into the patterns described above (Fig. 22).

Fig. 22

Distribution of hydrothermal minerals in a vertical section of the southern part of the central uplift of the Puchezh-Katunki impact structure (drill cores 40, 63, 754). Dashed lines indicate sporadic occurrence of a mineral. Compared with the Vorotilovo Deep Well (VDW), the inverse distribution of clay minerals is established. The distribution of zeolites and associated minerals is similar to the VDW section

Among post-impact alteration minerals, sulfides are paid attention due to its economic significance in some impact structures (Johansson 1984; Masaitis 1989; Grieve and Masaitis 1994). Within shocked basement rocks, some primary (pre-impact) sulfides occur; these are mostly pyrite and subordinate pyrrhotite, but in rare places, sphalerite, chalcopyrite, pentlandite, and millerite are documented. Post-impact sulfides are represented by three assemblages: (a) chalcopyrite-sphalerite-galena-pyrite association in laumontite-anhydrite veins that are common at the depth from 1400 to 3200 m in the VDW section; (b) marcasite-pyrite at marginal parts of analcime-calcite veins occurring at the top of authigenic breccia and overlying suevite and lithic allogenic breccia; (c) pyrite with minor marcasite and very rare pyrrhotite within the matrix of suevite and coptomict gritstone. Thus, pyrite is a throughout sulfide for all hydrothermal associations. Correspondingly, three pyrite generations occur distinguished by localization, crystal habit, chemical composition and sulfur isotope ratio.

• (1) Disseminated and vein pyrite from authigenic breccia (down of 1500 m in the VDW section) is characterized by: (a) prismatic and tabular crystal habit; (b) yellow color of grains; (c) high (>1) Co/Ni ratio; (d) δ³⁴S values (from −0.5 to +3.3‰ CDT) are typical for basic mantle-derived igneous rocks (Seal 2006) to inherit more probably the sulfur isotope composition of pre-impact sulfides from the Precambrian basement rocks (Table 9).

  Table 9 Isotopic compositions of sulfur, oxygen, and carbon in some hydrothermal alteration minerals from the Puchezh-Katunki impact structure

• (2) Pyrite from veins cutting allogenic breccia, suevite, and intensely deformed and shocked target rocks at the top of the central uplift, is represented by aggregates of fine (<0.3 mm) rhombidodecahedrons and cuboctahedrons of bronze tint; Co/Ni ratio is low (<1). This pyrite is enriched by the light sulfur isotope (δ³⁴S varies from −8.6 to −21.6‰ CDT; Table 9).

• (3) Pyrite in basal crater lake sediments (coptomict gritstone) is characterized by cubic or, more rarely, by pentagon-dodecahedron habit and of no more than 0.1 mm in size; in addition, anhedral bronze-tinted grains occur. Pyrite is highly enriched by the heavy sulfur (δ³⁴S = 33.4‰ CDT) (Table 9).

Both vertical zonation of post-impact sulfide associations and high variability in the composition and crystal habits indicate to the significant variation of crystallization conditions within the hydrothermal cell while distinction of sulfur isotope composition is evidence of different sources of fluids.

The estimation of input and output of chemical components during the hydrothermal transformation showed that chemical alteration trends are very similar for lower (chlorite-anhydrite) and upper (smectite-zeolite) zones of the hydrothermal column (Masaitis and Pevzner 1999): an addition of Ca and Mg and depletion of alkali and silica are observed for both ones. The principal differences between the zones are the less important depletion of silica in the smectite-zeolite zone, and the inversely correlated behavior of Al and Fe there. Aluminium is added in the smectite-zeolite zone, but iron, in contrast, in the chlorite-anhydrite zone. The summary effect includes significant addition of Ca, Mg, Sr, La, and Y, minor addition of siderophile elements (with increasing intensity in the Co-Cr-Ni series), and also Mo, Cu, Zn, and V. This is correlated with a loss of K, Na, Si, Ti, P, Al, Mn, Ga, Pb, and W, whereas, some of the lithophile elements (Nb, Zr, Yb, and Sc) are inert. Iron increases in the lower zone of the hydrothermal column together with other siderophile elements, but it is depleted in the upper zone. The balance between gain and loss of components is consistent with the newly formed mineral associations, e.g. the zeolite-bearing parageneses form from elements leached out of basement rocks during smectite and chlorite formation.

Thus, a geochemical effect of the post-impact hydrothermal alteration consists in the concentration of weak basic elements and the depletion of alkali and of high-valency amphoteric elements, which precipitate in cavities as zeolites. From a geochemical point of view, the impact-induced hydrothermal alteration may be compared with sub-alkali metasomatism.

The uniform mineral associations and behavior of chemical elements in the whole volume of the hydrothermal cell provide evidence for rather constant properties of the fluids during impact-induced hydrothermal transformations of the rocks. This conclusion is supported by the analysis of the mineral facies in the distinct zones of alteration. Most informative in this context is the diagram of mineral parageneses in relation to the chemical potentials of Na and H₂O (Fig. 23). This figure presents a solution of a unary multisystem for the six most typical hydrothermal minerals, i.e. smectite, chlorite, Ca–Na-zeolite, laumontite, calcite (anhydrite), and Ca–Fe-garnet. Calcium, Si, and Fe + Mg were chosen as virtual components.

Fig. 23

Parageneses of post-impact hydrothermal mineral associations in relation to the chemical potentials of water (approximately 1/T) and sodium (Naumov 2005). The figure presents a solution of a unary multisystem for the six most typical hydrothermal minerals. Calcium, Si, and Fe + Mg are chosen as virtual components. The diagram has been constructed in accordance with topologic regulations, which have been proposed by Schreinemakers and described in detail, by E-an Zen (1966). The diagram has nine stable univariant lines and two stable invariant points. The numbering of fields is a matter of convention. Vertical-hatched areas indicate mineral assemblages that are typical for this field alone, and oblique-hatched areas, typical for two adjacent fields. Compositions of minerals have been accepted by convention as follows: smectite (Sm): Na₀.₄(Fe,Mg)₆Si₇·9H₂O, chlorite (Chl): (Fe,Mg)₉Si₆·8H₂O, Ca–Na-zeolite (Ze): NaCa-Si₇·7H₂O, laumontite (Lm): CaSi₄·3.5H₂O, calcite (Cc) or anhydrite (Anh): Ca, and Ca–Fe-garnet (Grt): Ca₃(Fe,Mg)₂Si₃

When calculated, 4 invariant points of 6, and 6 univariant lines of 15 turned out to be metastable. The generally observed replacement of mineral parageneses in the vertical direction in impact craters corresponds to a vector passing across fields I-V-VIII-IX in Fig. 23. Taking into account that the water potential is completely controlled by temperature (H₂O ~ 1/T) at constant pressure, the obtained results indicate only very restricted variations in chemical parameters in the whole hydrothermal cell. Similarly, from T-μCO₂ diagrams for calcic minerals (Ryzhenko 1981), the precipitation of either calcite or anhydrite from solutions is governed primarily by the temperature at low PCO₂ values: calcium carbonate is stable at lower than 110–150 °C, whereas calcium sulfate precipitates at higher temperatures. Thus, it is probable that the thermal structure plays the decisive role in the evolution of the mineral associations in space and time.

The generalized order of mineral crystallization is as follows:

• in smectite-zeolite zone: saponite/montmorillonite, laumontite/chabazite–calcite, heulandite, analcime–apophyllite, calcite, Fe-montmorillonite, calcite;

• in chlorite-anhydrite zone—salite–epidote–chlorite–calcite, anhydrite, laumontite.

For both zones, these sequences confirm the uniform decrease of the temperature of mineral formation during the hydrothermal process.

Inferred from summarized data on alteration mineral distribution, the hydrothermal zonation is not caused by the composition of the target, nor by shock and post-shock thermal metamorphism. The latter is responsible for textural disordering of rocks and controls only the intensity of the hydrothermal alteration. The post-impact thermal field of an impact crater is the main factor controlling mineral formation. The temperature of crystallization of hydrothermal minerals decreases upward as well as outwards from the crater center. The given mineral zones correspond to a medium (chlorite-anhydrite), and a low (smectite-zeolite) temperature facies. The criteria for determining the boundary between both facies is the transition from chlorites to smectites in the temperature range of 160–180 °C, according to data for modern volcanic areas (e.g., Kristmannsdottir 1985; Slovtsov and Moskaleva 1989; Rychagov et al. 1994; Robinson and Santana de Zamora 1999, and others). Temperature assessments for the equilibrium association Ca-heulandite—and laumontite range from 150 to 190 °C (Cho et al. 1987). The given values are considered as maximum temperatures for the hydrothermal transformations in the smectite-zeolite zone. The association of high-silica zeolites with dioctahedral smectites in coptomict gritstones corresponds to temperatures of 80–130 °C (Rychagov et al. 1993), at which montmorillonite is stable in a near neutral environment. Thus, the temperature interval for mineral formation in the smectite-zeolite zone ranges from <100 to 200 °C, and reaches 200–350 °C in the chlorite-anhydrite zone. The latter value is derived from the presence of ferriferous garnets that form in geothermal systems at temperatures between 275 and 350 °C. The absence of corrensite indicates both relatively high temperature gradient in the smectite-zeolite zone and low water-rock ratio. Below, the temperature gradient is apparently very small. The lower Ni/Co ratio in pyrite from the smectite-zeolite zone is evident for lower crystallization temperature and lower sulfur fugacity as compared with chlorite-anhydrite zone.

The chemical parameters of the mineral formation vary within narrow interval of pH and, as a rule, correspond to weakly alkaline and near-neutral (pH = 6–8) environments. It is a sequence of the specific state of host rocks, which consist mainly of shock-derived aluminosilicates and diaplectic glasses, that are readily amendable to leaching. The intense leaching of this material providing the above-mentioned properties of solutions; besides, this is responsible for the oversaturation of solutions with silica, which, in turn, provides favorable conditions for the growth of Fe-smectites and zeolites. Both these minerals dominate in the upper zone of the hydrothermal column. Regular variations in zeolite compositions (increasing Si/Al ratio) and smectites (decreasing Al^IV contents) point to some drop in the pH values of the environment upwards. The vertical zoning of zeolite distribution was caused both by the temperature gradient and the drop of partial CO₂ pressure as manifested by the crystallization of Ca–Na-zeolites in association with calcite. The lateral zoning was apparently due to spatial differentiation of acid-basic properties of solutions, which controlled the development of high-silica (heulandite, erionite) either low-silica (Ca-chabazite, analcime) varieties.

The question for the source of the hydrothermal solution can be solved on the basis of a general model for impact-induced hydrothermal circulation (Masaitis and Naumov 1995). This model implies the superficial origin of solutions. Another part of the fluids originates in the shock-induced devolatilization of minerals. The most probable source of Ca, Mg, sulfur, and carbon dioxide are rocks in the peripheral part of the circulation cell where carbonate rocks and evaporites frequently occur in the megabreccias, which fill the ring trough. A preliminary estimate of the isotope composition of sulfur, oxygen, and carbon in some of the hydrothermal minerals (Table 9) does not contradict this suggestion. The relatively light sulfur isotope composition of anhydrite associated with pyrite at deep levels of the section corresponds to the sulfur composition of Permian evaporites in the region, i.e., δ³⁴S from 6.4 to 14‰ CDT (Vinogradov 1980). These data can be explained by isotope fractionation of sulfate sulfur with a coefficient of 1.007 that corresponds to the temperature of ca. 300 °C.

Stable isotope values in pyrite and calcite from the smectite-zeolite zone confirm unambiguously the superficial source of fluids. All analyzed calcite samples have moderately depleted δ¹⁸O values (less than +26‰, V-SMOW), which are appropriate to normal fresh- or seawater. Carbon isotope values span in a range from strongly (organic matter influenced) and moderately depleted (typical hydrothermal) to slightly enriched (normal marine) carbonates (from −26 to −7‰, V-PDB) (Versh et al. 2006) Mantle or metamorphic derived carbon source of CO₂ can be excluded. Thus, the isotope composition of oxygen and carbon in calcites from veins in the near-roof part of the authigenic breccia strongly suggests a low-temperature origin from a heterogeneous carbon source, which probably includes both carbonates and organic matter.

The principal scheme for the formation of the hydrothermal mineralization is given in the following. Superficial aqueous fluids infiltrate hot rocks of the central uplift and impactites filling the crater, reacting with shock-derived aluminosilicates and impact glass and forming Fe–Mg clay minerals. These fluids become weakly alkaline due to the selective removal of silica, aluminia, and alkalis, plus under neutral environment also of Ca. This is provided by the widespread presence of shock-disordered aluminosilicates and diaplectic glasses. A constant super-saturation of solutions by silica creates favorable conditions for the formation of Fe-smectites and zeolites—the main impact-generated hydrothermal minerals. During ascent in the thermogradiental field, the hydrothermal solutions become more acidic as a result of both the absorption of OH⁻ ions, as rock-forming silicates are replaced by smectites and chlorites, and because of the increase of the CO₂ solubility due to the decrease of the temperature. This is an important factor against the fall of pressure, which can cause dissolution of hydrocarbonate complexes that lead to boiling of solutions near the surface. Due to the thermal gradient in the authigenic breccia zone, the newly formed mineral associations also show a vertical and horizontal distribution. Silica, Al, K, and Na are removed, but accumulate partly in the upper zone of hydrothermal *cell by precipitation of zeolites.

Thus, the distribution of hydrothermal mineralization in the Puchezh-Katunki structure is satisfactory explained by the model of a near-surface circulation system, confined to the crater center and controlled by the thermal energy induced by the impact event. The characteristics of the hydrothermal mineralization are typical for impact structures (Naumov 1996, 2002, 2005). The zoning of the hydrothermal mineralization resembling an “abyssal” heat source actually results from the thermal evolution of the impact crater. Many features of the hydrothermal mineral associations at the Puchezh-Katunki, e.g., mineral associations, composition of minerals, temperature zoning, are in keeping with features of hydrothermal associations in volcanic areas. This similarity is the result of the similar thermodynamic and hydrological setting in the near-surface part of the Earth’s crust.

4 Thermal Evolution of the Impact Structure

The intense appearance of thermal transformations and hydrothermal alteration in the central uplift of Puchezh-Katunki structure is paid a special attention because it enables to simulate the thermal evolution of a large impact crater based on observed mineralogical effects in addition to speculative physical modelling.

The pre-impact thermal field of the target (crystalline basement and overlying sedimentary cover) was controlled by a geothermal gradient characteristic for stable platform regions. At present, this gradient ranges from 15 to 20° km⁻¹ and was most likely the same at the time of the impact in Jurassic.

The impact event changes abruptly the stable thermal regime. A subsequent thermal evolution of the impact structure should be considered with regard to successive stages of cratering.

• 1. The rocks constituting the present central uplift underwent a shock pressure of no more than 40–45 GPa; more intensely compressed rocks were evaporated, melted and ejected. Thus, residual temperature (immediately after the unloading behind the shock wave front) for rocks, which remained in solid state exceeded an initial (pre-impact) temperature for ca. 900 °C. Taking into account that at the moment of compression these rocks occurred at a depth up to 6–8 km, their post-shock temperatures could be as high as 1000–1100 °C. This temperature was decreasing sharply downward in accordance with shock-wave attenuation; at depths below 10 km, the post-shock temperature gradient became insignificant. Post-shock isotherms under transient crater floor are suggested to be hemispherical following in general the configuration of isobaric surfaces.

• 2. Additional heating in the subsurface zone of the excavation crater was due to injections of impact melt impregnated with fragments overheated much more than the liquidus temperature (up to 1500–1600 °C or even higher). This heating followed immediately the shock wave and radial displacements of rock masses. As the distribution of melted material in injected masses was rather irregular, in some places these masses consisted entirely of heated fragments enclosed in thin-crushed material showing (as well as the fragments) signs of viscous flow. This additional heating was irregularly spaced, local, and relatively short-term as far as the volume of injected masses was much less relative to the volume of the heated massif. It is quite probable that the heating was somewhat more intense owing to internal friction in the material during its instantaneous movement between rock blocks.

• 3. Differently directed fast block displacements behind the shock-wave front resulted in some friction heating along boundaries of blocks, up to local melting and pseudotachylite formation.

• 4. Elevation of the central uplift at the stage of early modification of the transient crater was accompanied by intense shifting and heating in a conical zone pinching out downward (Ivanov 2008). This heating is probably responsible for the thermal decomposition of minerals and emergence of thermal residual magnetization generating an intense positive magnetic anomaly over the central uplift. In some places, the selective melting occurred; this local melt could mix with still hot impact melt portions.

• 5. After the central uplift had arisen, a relatively long-term heat anomaly involving a massif of deformed crystalline rocks of conical or ellipsoid shape formed within it. The vertical extension of the massif was up to 5–6 km and the total volume of ca. 1000 km³ (Masaitis and Naumov 1995; Naumov 1996). The temperature was decreasing from 500–600 °C in the center of the massif to 100 °C at its periphery. The next evolution of the thermal anomaly consisted in temperature equilibration within it and simultaneous general cooling, which was also irregular: it was slower in areas where a thicker (a few hundreds of meters) cover of hot deposited ejecta overlaid the uplift. At the summit of the uplift the ejecta blanket was thin or absent, and therefore, oxidation of hot rocks and some additional heating occurred. During this stage, the conductive heat transfer was predominant (except for the uppermost of the authigenic breccia column). When the temperature in the central part of the massif fell to reach the critical point for water vapor to fluid transition (ca. 400 °C and lower in dependence from pressure), the stage of convective thermal flow started.

• 6. The last stage of thermal evolution of the Puchezh-Katunki impact structure consisted in the development of hydrothermal circulation. In fact, this circulation started at peripheral parts of the thermal anomaly where temperatures were lower than the critical water point immediately after its stabilization following the completion of the early modification stage. As the massif was cooling, the front of circulation moved toward its centre. Thus, the stages of convective and conductive cooling coexisted for a short time until the convective cell involved all mass of heated rocks.

Intense fluid circulation is provided by both temperature gradient and high permeability within the circulation cell and occurrence of the high-mineralized shallow basins in annular trough and central pit that supply available water reserves due to downward infiltration through relatively loose allogenic breccias and highly fractured authigenic breccia. In the middle of the central uplift, descending flows gave way to ascending ones. Hot water solutions supplied not only into the authigenic breccia, but also into porous breccias and suevites as well as into layers of redeposited material of the crater lake floor.

Three stages of the regressive hydrothermal process might be distinguished (Masaitis and Naumov 1995; Naumov 2002, 2005; Fig. 24). (a) an initial stage, when the isotherms preserve their original configuration (the temperature decreases from the top downward); (b) a main stage after an inversion of the thermal field; the thermal gradients during this stage ranging from about 30 °C km⁻¹ in the central part of the thermal anomaly to about 100 °C km⁻¹ on its periphery. (c) a final stage, during which thermal gradients are less than 10–30 °C km⁻¹ and hot-water circulation takes place in near-surface area only.

Fig. 24

A general scheme for impact-generated hydrothermal circulation in the Puchezh-Katunki impact crater. Three stages of united regressive hydrothermal process are shown (see text)

The substitution of chlorites by smectites downward in the section is evidence for mineral formation during the initial stage. The creation of the vertical zonation of hydrothermal minerals in the central uplift because of the thermal gradient in the authigenic breccia occurs in the main stage when the circulation system incorporates the maximum rock volume. Finally, zeolite-calcite cement is precipitated within suevites and lithic breccia, and late calcite vein forms.

The time of existence of the hydrothermal system is estimated to be several thousands of years (Naumov 2005). The hot-water circulation hastened the gradual cooling of the impact crater. Finally, the temperature distribution was levelled against the regional geothermal gradient, and the hydrothermal activity decayed.

Thus, it is supposed from the modelling of the thermal evolution of the impact structure that the thermal field in the central part of the sub-crater zone was inverted twice. Firstly, the inversion was caused by the impact-induced origination of a heated rock massif, in which the temperature decreasing downward; secondly, by the preferred cooling of this massif close to surface and reconversion to normal vector direction of the geothermal field.