Advertisement

Petrology

, Volume 27, Issue 1, pp 95–107 | Cite as

Comparison of the Compositions and Microstructures of Terrestrial and Lunar Impact Glasses: Samples from the Zhamanshin Crater and Luna 16, 20, and 24 Missions

  • T. A. Gornostaeva
  • A. V. MokhovEmail author
  • P. M. Kartashov
  • O. A. Bogatikov
Article
  • 7 Downloads

Abstract

The paper presents pioneering data on the comparative study of impact glasses from the Zhamanshin crater and lunar regolith (delivered by the Luna 16, 20, and 24 probes). The data were acquired using analytical techniques of ultrahigh spatial resolution. Many of the melt and condensate impact glasses, both terrestrial and lunar, are similar in inner structure and composition, which were controlled primarily by the physics of the impacts and similar compositions of the targets.

Keywords:

impact glasses Zhamanshin crater Moon 

INTRODUCTION

During its earliest evolutionary stages, the Earth was intensely bombarded by asteroids. In the course of this process, the planetary material of the Earth was practically completely modified by impacts, which caused its primary differentiation. Impact processes enormously affected the surfaces of this and other planets, and hence, studying these processes provides information of principal importance. Ultrahigh pressures and temperatures that were instantaneously generated during the impacts lead to the partial or complete evaporation of the target material, its partial high-temperature remelting coupled with selective evaporation, and subsequent quenching of the melt, a process that produced glasses of various composition, hosting crystal seeds of mineral phases. Because of this, impact processes can provide information on processes that reworked the rocks under extremely high temperatures and pressures.

The chemical composition of lunar glasses was studied to determine characteristics of rocks on the Moon’s surface (Reid et al., 1972), first of all, its regolith at the sampling sites (Wentworth and McKay, 1988). In the course of these studies, glasses were identified whose composition did not correspond to the usual types of lunar rocks and were noted for anomalously low silica concentrations. These glasses were found in all samples of lunar regolith delivered by both Soviet probes and American astronauts (Fredriksson et al., 1970; Ivanov, 1975). The SiO2 deficit in lunar glasses was then explained by the selective evaporation of this component during impact melting. For example, a Luna 20 sample of lunar regolith was found out to contain glass whose SiO2 concentration was 34.8 wt % and which was, according to (Dowty et al., 1973), residue after the evaporation of the target material at its impact melting. Chondrite-like objects (chondroids) found in Luna 16 and Luna 20 lunar regolith samples (Ivanov, 1975) were also distinguished for their anomalously low silica concentrations. This led to the conclusion that notable silica amounts were lost during impact processes on the Moon.

Simultaneously with the studies of lunar soil, impact evaporation processes were explored experimentally. The experiments were made with lunar rocks and compositionally similar terrestrial basalts (Naughton et al., 1972; Yakovlev, 1972, 2003; Markova et al., 1986). These studies have demonstrated that impact melts undergo selective evaporation: volatile components (Na, K, Si, and Fe) are preferably evaporated, whereas Ca, Al, and Ti enrich the residual material. Such evaporation experiments carried out with acid rocks (Horz et al., 1983; Dikov et al., 1993; Yakovlev et al., 2005) have confirmed the above tendencies. Analysis of the composition of melt glasses in lunar (Warren, 2008) and terrestrial (Yakovlev et al., 2005) impact structures has also shown that the residual melt is typically enriched in low-volatility elements (Ca, Al, and Ti), which can be evaporated at much greater temperatures (Horz et al., 1983; Yakovlev et al., 1997).

The succession of the evaporation temperature of major components from acid and mafic melts is generally consistent with the succession of the volatilities of individual oxides: К2O, Na2O, FeO, SiO2, MgO, CaO, A12O3, and TiO2 (Markova et al., 1986; Yakovlev et al., 1972). It has also been experimentally demonstrated that the vapors are dominated by silica (Yakovlev et al., 2011) because of the preferable evaporation of silicon as compared to other components: according to experimental data, more than 30% of the initial silicon concentration is evaporated (Ivanov, 1975).

Experimental data indicate that, similar to evaporation, condensation is also a selective process: first highly silicic condensate is formed, and the ensuing expansion and cooling of the cloud are associated with the condensation of material with lower silica concentrations and higher contents of volatile components (Wood and Hashimoto, 1993; Petaev and Wood, 1998; Yakovlev et al., 1972).

The condensate phases themselves can occur in the form of both individual globules and their agglomerations, which are peculiar three-dimension “cobwebs” of tree-shaped structures (Lushnipov et al., 1991; Rietmeijer et al., 2006).

METHODS

Our research was carried out using transmitting and scanning analytical electron microscopy (TEM and SEM, respectively) on a JEM-2100 electron microscope equipped with an IETEM INCA-350 energy-dispersive spectrometer (EDS) and a JSM-5610LV electron microscope with an INCA-450 EDS system.

The main technique applied in preparing samples for SEM was the manufacturing of polymer and/or epoxy pellets with the samples and their cutting and polishing. The polished pellets with samples to study were placed into tubular aluminum holders and fixed in them with Wood’s alloy. All polished samples were sputter coated with carbon. This was done to facilitate interpreting and recalculating the X-ray energy-dispersive spectra, because this element has a single K-series peak, which occurs in the low-frequency region and can be readily eliminated when the spectra are quantitatively processed. A true quantitative analysis is assumed to be determining concentrations whose total corresponds to the real total of the analyzed elements. When EDS analyses are recalculated, they are quite often normalized to 100 wt %. Strictly speaking, this analysis cannot be considered to be quantitative because the sample may contain elements that cannot be analyzed by EDS (H, Li, Be, and perhaps, others), and hence, the measured analytical totals can be not 100 wt %. In this situation, a result normalized to 100 wt % distorts the actual concentrations of the elements. However, the total concentrations of elements that cannot be analyzed by EDS in our glass samples were obviously no higher than a few tenths of a percent and commensurable with the cumulative analytical errors, and hence, the analyses can be regarded as quantitative. With regard for these considerations and to speed up the analytical procedure of the glasses, we normalized the analyses to 100 wt %.

In the course of the analysis, we used standard reference samples for both the shapes of the peaks and the concentrations of elements.

Some of the samples were studied in chips. A chipped fragment was glued to the electron-microscope stage with conductive double-face adhesive scotch tape and sputter coated with carbon. Because classic qualitative analysis cannot be applied to a sample of irregular shape, we utilized a previously proven technique that involved sample rotation around the detector (Mokhov, 2009; Kartashov et al., 2010).

Samples for TEM were prepared by suspension techniques. A preparatorily powdered fragment of the sample was placed into a sterile test tube. The test tube with a glass particle and a small amount of distilled water (of analytical grade purity) was put into an ultrasonic dispergator to disintegrate the sample. The small fragments were locally extracted with an ultrasonic extractor that involved an optical microscope and ultrasonic needle (Gornostaeva et al., 2011). The ultrasonically dispersed sample was applied with a droplet of the suspension to a copper grid with an underlying collodion film and was dried in a dewatering box at a temperature of 35°C.

When the particles were analyzed by TEM, their X‑ray spectra always showed fluorescence-excited peaks of the copper grid. These peaks were subtracted from the spectra in the course of their processing, and the presence of copper in the sample can be inferred from the relations between the copper peaks Kα/Lα. The quantitative analysis was conducted using the integrated program package underlain by the Cliff–Lorimer method (Cliff and Lorimer, 1972). In this method, the relative intensities are evaluated with an obligatory assumption that the total of the concentrations of elements is 100% to eliminate variations in the intensities of the analytical lines due to differences in the mass thicknesses of the analyzed particles, and no corrections for matrix effects are introduced. Inasmuch as matrix effects nevertheless start to show up after a certain mass thickness is reached, these effects are taken into account by manually introducing the thickness and density values of the object. The correction coefficients K for this technique were calculated from microprobe analysis of glasses of similar composition preliminarily analyzed by SEM.

It should also be taken into account that the high-energy beam of a transmitting electron microscope locally significantly heats the samples and thus triggers the removal of volatile components (for example, Na and sometimes also K) from it. If the total is normalized to 100% by the Cliff–Lorimer method, this leads to a significant overestimation of the concentrations of other elements, particularly Si. Consequently, TEM analyses of glasses commonly show somewhat overestimated Si concentrations as compared to conventional SEM analyses of the same glasses. To preclude this, we had experimentally evaluated the removal rates of volatiles from glasses, determined how the concentrations of these elements depend on the counting time, and calculated coefficients for recalculating the concentrations of the elements at the start of spectrum recording. The user coefficients K were corrected with regard for these dependences.

MATERIALS

This study was carried out with impact glasses (irghizites and zhamanshinites) from the Zhamanshin crater in Kazakhstan (48°24′ N, 60°58′ E). Samples for this study were taken from P.V. Florensky’s collection and were provided for us by courtesy of the staff of the Petrographic and Ore Museum at IGEM RAS. Zhemanshinite is extensively vesicular glassy or partly recrystallized rock found as blocks of irregular shape up to 50 cm across, with some of the blocks resembling volcanic scoria. Irghizite is black glassy material with a glossy surface and fluidal structure, which has a shape of droplets and solidified splashes ranging from 1–3 mm to 2–4 cm. Larger particles often coalesce with smaller glass beads when in midair. These terrestrial rocks were compared with impact glasses from lunar regolith from the Mare Fecunditatis (Luna 16 samples), Mare Crisium (Luna 24 samples), and an isthmus between them (Luna 20 samples).

CLASSIFICATION OF IMPACT GLASSES

Various current classifications of terrestrial impact glasses include their systematics based on color and SiO2 concentration. However, it was found out that the composition of some impact glasses is not correlated with their color. For example, Delano et al. (1981) have studied the composition of impact glasses that were identified as “yellow” according to their color in thin sections and determined that the composition of the glasses varies (for example, the TiO2 concentration varies from 1.9 to 12.9 wt %, and the SiO2 concentrations varies within a broad range), and these glasses do not define a single group. It was also determined that the color of melt glasses (zhamanshinites) from the Zhamanshin crater is not uniformly correlated with their composition. For example, blue zhamnshinites, which are noted for high CaO concentrations (up to 7 wt %), have a variable composition according to (Koeberl, 1988).

According to literature data and our earlier results (Gornostaeva et al., 2017), the color of impact glasses does not provide unambiguous genetic information. Moreover, the color of glasses is not their impartial property, and hence, it is reasonable to use it only during provisional studies of the glasses to conveniently label them, as we have done in the course of this study.

The compositional gradation of terrestrial impact glasses, first of all, according to their SiO2 concentrations, was applied for further searches for analogues among the target rocks. Lunar glasses were also classified based on their comparison with lunar rocks (Engelhardt and Stengelina, 1981). However, the composition of some of the found glasses did not correspond to any types of lunar rocks known so far. These are, for example, green glass spherules in Apollo 15 samples and orange spherules in Apollo 17 samples (Arndt and Engelhardt, 1987). Studies of the orange glasses have shown that no rocks pf analogous composition had ever been found on the Moon (Roedder and Weiblen, 1973). It was proved that the green glasses are not genetically related to mare basalts, which are bedrocks at the sampling sites of these glasses (Ma et al., 1978).

It was suggested that some impact glasses have certain source rocks based on the losses during the evaporation of some volatile components. One of the shining examples of these glasses are HASP (High-Aluminum Silica-Poor) ones, which are a refractory residue after the evaporation of impact melt. The term HASP was coined in (Naney et al., 1976) to describe the following compositional range of impact glasses (wt %): 30–34 SiO2, 32–36 A12O3, and ~20 CaO. HASP glasses compose up to ~50% of the fine fraction of the lunar soil (Keller and McKay, 1992b). These glasses were found in many Apollo samples (Vaniman, 1990; Wentworth and McKay, 1988), samples delivered by Soviet space missions, and in lunar meteorites (Warren and Kallemeyn, 1995). According to (Warren, 2008), the origin of HASP glasses implies losses at the evaporation of 1/4 to 1/3 of the original SiO2 concentration (the average SiO2 concentration in the lunar highland crust is 45 wt %). However, all glasses of feldspathic composition with A12O3 > 16 wt % and SiO2 < 40 wt %, for which no SiO2 evaporation was suggested, and high-Al glasses with A12O3 > 45 wt % and SiO2 < 27 wt % were also classed with HASP (Vaniman, 1990). Moreover, ultra HASP were discovered (Keller and McKay, 1992b), which are glasses containing 65 wt % A12O3 and as little as 5 wt % SiO2. Generally speaking, the HASP group comprises glasses with >16 wt % Al2O3 and <40 wt % SiO2 (Vaniman, 1990).

RESULTS AND DISCUSSION

HASP-Type Glasses in Lunar Regolith

In the course of this study, some lunar regolith samples studied under a SEM were determined to have a composition corresponding to HASP. These glasses were found in all of the studied Luna 16, 20, and 24 samples. The detected particles were of either clastic or spherical shapes, which is consistent with literature data (Vaniman, 1990).

The composition of these HASP glasses broadly varies, first of all, in Ca concentration. Some of the particles were poor in Ca and had Mg concentrations up to 10 wt %, i.e., were analogous to those previously found in Apollo 16 samples (Naney et al., 1976).

The HASP glasses found using SEM were commonly not individual particles but were components of glass aggregates. As an example, the regolith particle shown in Fig. 1 hosts a glass fragment of HASP composition (Fig. 2). Thereby the particle itself is agglutinate of glasses of different composition with inclusions of iron spherules.

Fig. 1.

Lunar regolith particle with fragments of HASP glass.

Fig. 2.

Energy dispersive spectrum of a fragment of HASP glass.

Our TEM study also allowed us to identify glasses whose composition corresponded to HASP. Practically all of the glasses hosted crystalline inclusions up to 0.2 µm across. Figure 3 shows a crystalline pyroxene inclusion in glass of HASP composition, as follows from its analysis and its electron microdiffraction pattern.

Fig. 3.

TEM image of a crystalline pyroxene inclusion in a particle of HASP glass from Luna 24 material.

HASP-Type Glasses in Zhamanshinites

When studying the zhamanshinites under a SEM, we detected glasses whose composition was close to those of lunar HASDP glasses. As an example, Fig. 4 shows a glass fragment in which a contact between HASP glass (labeled with 1) and typical melt glass of zhamanshinite (labeled 2) is seen. Figure 5 shows the energy dispersive spectrum of the HASP glass, and Fig. 6 displays the spectrum of glass 2.

Fig. 4.

BSE image (SEM) of a contact between glasses of two types: (1) HASP glass and (2) melt glass of zhamanshinite.

Fig. 5.

Energy dispersive spectrum of HASP glass (glass 1) shown in Fig. 4. Zhamanshinite. SEM

Fig. 6.

Energy dispersive spectrum of HASP glass (glass 2) shown in Fig. 4. TEM. Zhamanshinite. SEM

At the spatial resolution of TEM, we have found out that samples of the brown zhamanshinite contain high-Al glasses with <35 wt % SiO2, which corresponds to HASP glasses. These glasses abound in crystalline inclusions of various composition. The energy dispersive spectra of these glasses show major lines of oxygen, aluminum, silicon, and calcium. Some of the spectra display notably intense peaks of iron, whereas the peaks of magnesium, phosphorus, sulfur, and titanium (whose presence is explained by the occurrence of nanometer-sized inclusions in the glasses) are less intense. The intensities of iron and titanium peaks are enhanced by glass-hosted inclusions, whose composition corresponds to magnetite and titanomagnetite. The copper and carbon peaks are excited by the copper grid and collodion film.

The quantitative analyses of the HASP glasses found by TEM in the brown zhamanshinites have yielded the following average concentrations (wt %) of major components: 0–0.3 MgO, 31–38 Al2O3, 33–35 SiO2, 18–25 CaO, 0–2 TiO2, and 5–9 FeO. Similar to the lunar glasses, these glasses were homogeneous (as seen at the SEM spatial resolution), did not host any inclusions, and contained Na and K (up to 5 wt %). These glasses were found out to host numerous large (up to 200 nm) inclusions of feldspar nanocrystals. These nanocrystals influence the results of the analysis (at this resolution level) and “introduce” elements typical of feldspar, for example, Na and K. As follows from the foregoing, SEM analyses of HASP distort information on the actual composition of these glasses because of the presence of nanometer-sized inclusions, which are found by TEM. Hence, the actual composition of these glasses can be determined only at the spatial resolution of TEM.

It follows that terrestrial impacts produced practically exact analogues of lunar HASP glasses, and hence, this term can be applied to terrestrial impactites enriched in Al and depleted in Si and containing relatively much refractory elements.

The parental rocks of lunar HASP glasses were interpreted differently with regard for the partial evaporation of their elements, for example, Na and Si. For example, it was demonstrated with the application of cluster analysis that HASP glasses were produced at the evaporation of anorthosite gabbro because, according to (Naney et al., 1976), their composition is intermediate between continental basalt (anorthosite gabbro) and plagioclase glasses. Some researchers believe that HASP glasses can be produced by the impact melting of anorthosite (Norris et al., 1993), whose atomic ratios Ca : Al : Si = 1 : 2 : 2. According to another interpretation, the parental rocks of these glasses were lunar soil and/or KREEP basalt breccia (Papike et al., 1997).

The exact limits of the concentrations of major components in these glasses, and hence, their identification, are still uncertain (Norris et al., 1993; Adcock et al., 1997), and the term HASP itself is referred to any lunar glasses enriched in Al and depleted in Si, which is most commonly caused by evaporation as a result of an impact event (Adcock et al., 1997).

At the same time, the SiO2 losses cannot be predicted either exactly or even approximately. For example, laser-induced Si evaporation from glasses of lunar regolith are in certain instances greater than 30 wt % of the initial concentrations (Ivanov and Florensky, 1975). According to other experiments, the silica mass losses due to evaporation of all HASP glasses vary from approximately 20 to 50% (Yakovlev et al., 2009). This broad range is explained by the authors as controlled by the different heat histories of each of the individual particles because of the spatiotemporal variations in the temperature. This is the likely reason for the significant variations in the composition of our analyzed glasses, as is most evident at the TEM level of spatial resolution.

It should be taken into account that the composition of the glasses is controlled not only by selective evaporation during the impact but also by the intense stirring of the heterogeneous melts and the possible later onset of liquid immiscibility (Zolensky and Koeberl, 1991; Pratesi et al., 2002). Hence, direct analogies between impact glasses and their allegedly parental rocks can hardly be drawn.

Tectite-Like Glasses on the Earth and Moon

Along with HASP glasses, terrestrial glasses referred to as tektites have no analogues among the target rocks. Tectite-like glasses were found in lunar glasses from Apollo 12 samples (O`Keefe, 1970). This publication reports the following average composition (wt %) of these glasses: 0.7 Na2O, 6 MgO, 12 Al2O3, 61 SiO2, 2 K2O, 6.3 CaO, 1.2 TiO2, and 10 FeO. Based on similarities between the composition of these glasses and terrestrial tektites, the latter were thought (O’Keefe, 1976) to be lunar tectites brought to the Earth with a comet. An analogous interpretation was also suggested for the origin of irghizites (Izokh, 1986; Skublov and Tyugai, 2004).

In the course of this research, we have found glasses whose composition was close to tektites in Luna 16 regolith samples. Figure 7 displays a fragment of tektite-like glass, whose composition (wt %) is 13.4 Al2O3, 74.8 SiO2, 2.7 K2O, 3.2 CaO, and 6.0 FeO. The fragment is hosted in a matrix of glass of other composition (wt %): 5.0 MgO, 1.3 Al2O3, 51.9 SiO2, 6.1 CaO, 0.8 TiO2, 1.0 MnO, and 34.0 FeO.

Fig. 7.

BSE image (SEM) of a regolith particle. The arrow points to an inclusion of glass of tektite composition. Luna 16 material.

The composition of this tektite-like glass is comparable with the composition of glasses found in lunar soil (O`Keefe, 1970). This composition even better correlates with the composition of terrestrial tektites, whose average composition (wt %) is 68–82 SiO2, 10–15 Аl2O3 (Florensky and Dikov, 1981; Otmakhov et al., 2006; Magna et al., 2011; and others) and the composition of our irghizite samples (wt %) ~75.6 SiO2, ~0.5 TiO2, ~10 Al2O3, ~3.0 MgO, ~2.8 CaO, ~5.3 FeO, ~0.8 Na2O, and ~2.0 K2O (Gornostaeva et al., 2016).

Hence, regardless of the location of the impact (on either the Earth or the Moon), the evaporation and melting of the target material are associated with the transfer of much of major components (such as Na, K, Fe, Si, Mg, Al, and Ca) to the gas cloud and melt. At a high-energy impact, the temperatures and pressures are obviously higher than those sufficient to evaporate these elements. At the same time, the temperature and pressure gradients only insignificantly depend on the localization of the impact and external parameters (Zeldovich and Raizer, 2008). Correspondingly, melts with such sets of chemical elements will produce glasses some of which are similar in composition and inner structure to those in other impact structures. These considerations explain the reasons for similarities between the compositions of some glasses in various impact structures on both the Earth and the Moon, and this is confirmed not only by our study but also by literature data (Izokh and Le Dyk An, 1983; Skublov and Tyugai, 2004). Lunar tektite-like glasses, irghizites, and HASP glasses from the Zhamanshin crater and lunar soil seem to be glasses of this type.

Condensate Glasses on Earth and Moon

Globular condensate glasses, which were produced by the coalescence of individual globules into their aggregates were found (although in small amounts) in regolith samples delivered by three Soviet space missions. Along with the aggregates, the samples contain single condensate globules up to 200 nm. Such globules were found in both lunar regolith (Fig. 8) and zhamanshinite samples (Fig. 9).

Fig. 8.

TEM image of an amorphous spherule of condensate glass. Luna 24 material.

Fig. 9.

TEM image of a condensate spherule in zhamanshinite.

Lunar regolith contains rarely found less silicic globular glasses containing such volatile components as Na and K. These glasses are usually recrystallized and contain poly- and single-crystal mineral phases, for example, pyroxenes (Fig. 10). Analogous glasses were found in the melt component of zhamanshinites (Fig. 11).

Fig. 10.

TEM image of an aggregate of globular Na-bearing recrystallized lunar glasses. Luna 16 material. The inset shows a fragment of a pyroxene crystal.

Fig. 11.

TEM image of an aggregate of globular Na-bearing recrystallized lunar glasses from zhamanshinite. The inset shows a fragment of a pyroxene crystal.

Condensate glasses were also found when irghizites were studied by TEM (spatial resolution ~5 nm). These condensate glasses of globular morphology are typically highly silicic (>82–85 wt %) and host nanometer-sized inclusions.

In addition to globular shapes, the condensate is sometimes found as thin films on the surface of particles and envelops these particles in both lunar and terrestrial samples.

For example, a pyroxene crystal in a regolith sample (delivered by Luna 24) was found out to be coated with a thin (approximately 200 nm) film. Its analysis shows that it contains, in addition to Si and O, only trace amounts of Al. Condensate films of the type were found out to cover many nanocrystals and single particles in our samples studied by TEM. Many of the found particles of native molybdenum in lunar samples were coated with high-silicon condensate films (Fig. 12). The latter sometimes contains, along with Si an O, also volatile elements, such as Na and K. The thickness of the condensate film broadly varies because of multiple impacts, which caused the multilayer inner structure of the film.

Fig. 12.

TEM image of native molybdenum nanocrystals in a condensate film. Luna 24 material.

Analogous condensate covering on nanocrystals were found in terrestrial impact glasses. In an irghizite sample, we found an aggregate of rutile crystals coated with a very thin nanofilm of high-K glass, which contained no elements other than Si and O (Fig. 13). The enlarged fragment (Fig. 13, inset) shows that the thickness of this film is as small as 10 nm. A zhamanshinite sample was found out to contain a srebrodolskite (Ca2\({\text{Fe}}_{2}^{{3 + }}\)O5) crystal coated with a thin film of analogous high-Si condensate (Fig. 14).

Fig. 13.

TEM image of rutile crystals coated with a glass nanolayer in irghizite. The inset shows an enlarged fragment of the image.

Fig. 14.

TEM image of a large inclusion of srebrodolskite Ca2Fe2O5 coated with a condensate film in zhamanshinite.

The condensate glasses were found in close association with melt glasses, as was discovered in the course of our research and those of other scientists (Dikov et al., 2000). The ratios of Mg and Si isotopes measured in HASP (Herzog et al., 2012) provided grounds to suggest that they may include condensation products of a gas–plasma cloud.

Our TEM study led us to discover a significant compositional heterogeneity of the lunar glasses. In addition to melt glasses, the samples also included high-Si glass that contained only 2–7 wt % Al2O3, with this glass most likely formed by condensation. In some instances, these glasses occur in association with melt glasses.

Many researchers believed that irghizites (Florensky and Dikov, 1981; Margolis et al., 1991) and tektites (Engelhardt et al., 1987) are of condensate genesis. As has been demonstrated by calculations, the condensate droplets produced by any impact cannot be larger than a few hundred nanometers in diameter and cannot grow to a few centimeters under any naturally occurring physical parameters (Zeldovich and Raizer, 2008; Yakovlev et al., 2003; Lushnipov, 1991; Rietmeijer et al., 2006; Gornostaeva et al., 2018).

The minimal size of primary condensate segregations is, according to experimental data, close to 6 nm (Rietmeijer et al., 2006). According to our data, the minimal size of individual spherules in the lunar condensate glasses is about 5–10 nm, and roughly the same size is also typical of condensate glasses from the Zhamanshin crater (Gornostaeva et al., 2016).

The maximal size of the individual condensate globules in synthesized condensate varies from 100 to 400 nm (Rietmeijer et al., 2006). Lunar (Fig. 8) and terrestrial (Fig. 9) condensate glasses in our samples contain spherules of comparable size of 100–300 nm.

One of the typical features of the lunar glasses produced by the condensation of an impact cloud is their high silica concentrations (Keller and McKay, 1992a, 1997; Christoffersen et al., 1994; Warren, 2008). Theoretical considerations concerning the process of evaporation and its experimental studies indicate that the dominant component of the impact-generated vapor is silicon and its volatile compounds (such as SiO), and this is reflected in the composition of the condensate, including its high silica contents regardless of the composition of the target rocks (Ivanov, 1975).

Keller and McKay (1992a) described nanospherules containing 89 wt % SiO2 (on average), with some of the spherules containing as much as 93 wt % SiO2 and small Al admixtures. The high silica concentrations and spherical shapes of the spherules are, according to these authors, indicative of their generation by the condensation of a gas cloud. We have found very closely similar spherules in both lunar regolith and zhamanshinite samples.

It is, however, pertinent to mention that the material of the condensate glasses never consists of SiO2 alone. For example, even the silica richest condensate glasses in our irghizite and zhamanshinite samples never consist of 100% SiO2 but contain 2–3 wt % Al2O3. The homogeneous condensate glass spherules found in the course of our TEM study of lunar regolith samples contained small Al2O3 admixtures (1–3 wt %). The glasses sometimes also contained trace concentrations (within the sensitivity of the method) of other elements. It should be mentioned that this is also typical of the synthetic condensate: experimental data indicate that the silicic condensate contains low concentrations of Al and Fe (Rietmeijer and Karner, 1999).

The spherical shapes and high SiO2 concentrations (81–97 wt %) of GASP (Gas-Associated Spheroid Precipitate) high-Si condensate glasses <10 µm found in lunar samples (Warren, 2008) were the reasons for attributing them to condensates. However, as was demonstrated above, GASP spheroids approximately 10 µm across cannot be produced by condensation. Moreover, we have demonstrated that irghizites are melt droplets coated with condensate glasses (a few nanometers thick) (Gornostaeva et al., 2016). The GASP glasses found in the matrix of HASP glasses were likely produced by an analogous mechanism (Warren, 2008). According to Warren (2008), melt droplets that solidified in midair fell back into the still-hot melt. The electron-microscope images presented in this paper show clearly seen GASP globules with coatings. It can be hypothesized that, similar to irghizite, GASP droplets where coated with condensate when flying through the impact cloud, and this was the condensate described in (Warren, 2008) as coating on the spherules. Contrast in the BSE images suggests that the condensate on the GASP spherules is much richer in Fe than the high-Si condensate of the irghizites. This fact may be explained by a higher Fe concentration in the gas–plasma cloud and by that the GASP spherules flew through this cloud at lower temperatures of the condensate window than the temperatures at which the irghizites were formed.

Each compound is characterized by its own evaporation and condensation temperatures. As known from experimental data, K2O and Na2O are the first to evaporate and the last to condense, and the temperatures of these processes are significantly different from the evaporation and condensation temperatures of other elements (Petaev et al., 2014). Low-temperature condensation produces glasses enriched in volatile components. In rare instances, condensate glasses in lunar soil samples are not only highly silicic but also contain volatile components, first of all, K2O and Na2O. For example, condensate glasses of the VRAP (Volatile-Rich Alumina-Poor) type were described in (Keller and McKay, 1992b) as spheroids ranging from 200 to 400 nm and containing 54–70 wt % SiO2, <5 wt % Al2O3, 2–3.5 wt % K2O, and 0.4–1.4 wt % Na2O. The spheroids were typically rich in FeO (20–32 wt %), but this feature is explained by the presence of abundant nanometer-sized Fe inclusions, found in (Keller and McKay, 1992b) in these glasses. These authors believe that VRAP glasses are condensates produced by the condensation of silicate vapors from the melt when the HASP were formed. It was later suggested (Warren, 2008) that high concentrations of volatiles in VRAP glasses might be explained by high concentrations of these elements on the surface of the spheroids when they flew through the low-temperature zone of the gas–plasma cloud. Some experimentally produced condensate glasses (Yakovlev et al., 2000) were also closely similar to VRAP.

This low-temperature process could occur either very late in the condensation window or take place in the pore space at a high concentration of volatile components in the vapors. In view of this, similarities can be found in lunar VRAP glasses and the pore condensate. Such pore condensate was found in the course of this study among melt glasses of zhamanshinite (Fig. 11). It was likely formed in relation to condensation in the close space of a pore, on its walls, from vapors rich in light elements. We have detected analogous condensate in lunar samples (Fig. 10), and it was most probably formed in volcanic (but not impact) glasses, as follows from the fact that the surface layer of lunar soil is depleted in volatile components (Florensky and Nikolaeva, 1984; Ivanov, 2014). However, in contrast to the volcanic condensate, VRAP glasses show obvious evidence of an impact process (Keller and McKay, 1992b; Warren, 2008). Correspondingly, they could only be formed very late during the low-temperature evolution of the condensate window.

It was once believed that amorphous films on the surface of particles are typical exclusively of atmosphere-free cosmic bodies and were formed under the effect of solar wind (SW). Regolith particles are bombarded by SW particles, which penetrate to depths of a few dozen nanometers from the surface. The surface layer of the regolith is thereby amorphized, and thus glasses films are produced, which were first found in Apollo samples (Dran et al., 1970; Bibring et al., 1972). One of the SW operating agents is fluxes of heavy ions, which leave behind tracks in regolith grains (Walker and Yuhas, 1973). One of the possible SW effects is enrichment in light elements, including carbon. Proponents of the solar wind theory overlooked the possibility of the origin of amorphous films under the effect of an impact itself, although theoretical considerations in (Zeldovich and Raizer, 2008) directly suggest that this is the main mechanisms producing these films. This gap was bridged in (Keller and McKay, 1991; Dikov et al., 1998).

At heterogeneous nucleation and growth of the droplets in the course of the condensation process, the growth rates of these droplets should be equal, and the probability of the coalescence of the droplets and their joint layer-by-layer growth during the rest of the condensation time is high. Thereby nanometer-thick films can be formed without retaining the globular character of the elements. This is confirmed for both lunar rocks (Wentworth et al., 1999) and regolith grains (Noble et al., 2001). At a great number of impacts and microimpacts on the Moon and in the absence of weathering processes (which could have destroy these films), the films on regolith particles should be extremely widespread.

In contrast to lunar regolith, whose fragments are often coated with condensate films, the latter are only very rarely found (but nevertheless are) in terrestrial impact glasses. For example, we have found a condensate film on srebrodolskite nanocrystals (Fig. 14) in irghizite and on rutile (Fig. 13) in irghizite.

Condensate films differ from spherical (droplet) condensates, which were formed in the impact cloud, first of all, in structure and composition. Comparison of the composition of condensate films produced in experiments and the composition of films detected on lunar regolith particles (Keller and McKay, 1997) led Yakovlev et al. (2010) to conclude that they were formed at the complete condensation of the vapors, regardless of the individual fugacities of the components. Experiments on the evaporation and condensation of augite have shown that the upper condensate layers contain elevated concentrations of Na (up to 5.7 wt %) (Dikov et al., 2009). The composition of the glass film differs from that of the matrix glass in the presence of highly volatile elements, such as Na and Zn. Studying lunar glasses in Luna 16 lunar soil samples by XPS, Dikov et al. (2002) have found out that the surface layers of these glasses are not compositionally homogeneous and are enriched in volatile components, such as Zn, Na, and C. Hence, the condensate films may contain both volatile (C, Na, K, and Zn) and mildly volatile (Si and Fe) elements. Our study has confirmed that some lunar condensate films contain volatile elements, including Na and K, as well as C, although not all of the films contain these elements.

The reasons for variations in the thicknesses of the films are the energetic parameters of the impacts, the residence time of the particle in the condensation zone, and later processes affecting the particle. The latter may be later impacts and SW bombardment, which can increase the thickness of the film.

Glass condensate films on the particles conserve and preserve these particles and preclude the oxidation of their metallic phases that were formed during impacts on both the Moon and the Earth (Mokhov et al., 2011; Gornostaeva et al., 2014, 2016).

Unlike lunar condensates, terrestrial condensates with volatile components have practically no chance to survive with time because of their destruction and decomposition by weathering. In view of this, the condensate films first found in terrestrial impactites in the course of our study are much more silicic, except only the condensates in closed pores in the zhamanshinites (Gornostaeva et al., 2017).

CONCLUSIONS

Many impact glasses on both the Earth and the Moon are similar in structure and composition, which are controlled primarily by the physics of the impacts and the similar compositions of the targets. This provides grounds to use a common systematics for lunar and terrestrial impact glasses and warrants drawing more analogies between them, which in turn, facilitates more accurate and realistic extrapolations of data on terrestrial samples to lunar ones and, what is particularly important with regard for the deficit in the lunar materials.

Intense processes of the complete melting of target rocks at stirring, evaporation, and condensation of the material result in glasses whose composition only weakly correlates with that of the target rocks and impactor material. The composition of some of the glasses does depend on the composition of the targets, but other glasses have merely some universal composition. It should be taken into account that the composition of the glasses is controlled not only by selective evaporation but also by intense stirring of the heterogeneous melts after the impacts and the possible onset of their liquid immiscibility. It is hardly possible, and cannot be warranted, to use direct analogies of either lunar or terrestrial impact glasses with their target rocks.

The amorphous surface layer on regolith particles is formed not only as a result of interaction between the material and solar wind but can also be formed, both on the Earth and the Moon, at impact events. Condensate glass films on the surface of metallic particles and mineral phases fix and protect them, preclude their oxidation and secondary alterations, including those at long-lasting contact with the atmosphere.

Notes

ACKNOWLEDGMENTS

The authors thank Prof. P.V. Florensky (Gubkin State Oil and Gas University) and Dr. M.K. Sukhanov (Institute of the Geology of Ore Deposits, Petrography, Mineralogy, and Geochemistry, Russian Academy of Sciences) for providing samples for this study. This study was conducted under government-financed project “Study of the Composition and Structure of Mineral Materials with Techniques of High Spatial Resolution”.

REFERENCES

  1. 1.
    Adcock, C.T., Spilde, M.N., and Papike, J.J., Automated HASP glass search using the electron microprobe, 38th Lunar Planet. Sci., 1997, vol. 28, pp. 1151–1152.Google Scholar
  2. 2.
    Arndt, J.V. and Engelhardt, W., Formation of Apollo-17 orange and black glass beads, Proc. XVII Lunar Planet. Sci. Conf. Part 2. J. Geophys. Res.: Solid Earth, 1987, vol. 92, no. B4, pp. 372–376.CrossRefGoogle Scholar
  3. 3.
    Bibring, J.P., Duraud, J.P., Durrieu, L., et al., Ultrathin amorphous coatings on lunar dust grains, Science, 1972, vol. 175, pp. 753–755.CrossRefGoogle Scholar
  4. 4.
    Christoffersen, R., McKay, D.S., and Keller, L.P., Grain rims on ilmenite in the lunar regolith: comparison to vapor deposits on regolith silicates, 35th Lunar Planet. Sci. Conf. Proc., 1994, vol. 25, pp. 259–260.Google Scholar
  5. 5.
    Cliff, G. and Lorimer, G.W., Quantitative analysis of thin metal foils using emma-4, the ratio technique, Proc. 5th European Congress on Electron Microscopy, Institute of Physics, London, 1972, pp. 140–141.Google Scholar
  6. 6.
    Delano, J.W., Lindsley, D.H., and Rudowski, R., Glasses of impact origin from Apollo-11, -12, -15, and -16: evidence for fractional vaporization and mare/highland mixing, Proc. 12th Lunar Planet. Sci. Conf., 1981, pp. 339–370.Google Scholar
  7. 7.
    Dikov, Yu.P., Gerasimov, M.V., Yakovlev, O.I., and Wlotzka, F., The correlation of alkalis and aluminum during high-temperature volatilization of albite and nepheline, 34th Lunar Planet. Sci. Conf. Proc. Abstracts, 1993, pp. 403–404.Google Scholar
  8. 8.
    Dikov, Yu.P., Ivanov, A.V., Wlotzka, F., et al., High enrichment of carbon and volatile elements in the surface layers of Luna 16 soil sample 1635: result of comet or meteorite impact?, Earth Planet. Sci. Lett., 1998, vol. 155, no. 3, pp. 197–204.CrossRefGoogle Scholar
  9. 9.
    Dikov, Y.P., Huth, J., Wlotzka, F., and Ivanov, A.V., HASP glasses in Apollo-17 orange soil sample 74220, 31st Lunar Planet. Sci. Conf., 2000, vol. 31, pp. 1110–1111.Google Scholar
  10. 10.
    Dikov, Yu.P., Ivanov, A.V., Wlotzka, F., Galimov, E.M., and Wanke, G., The nature of volatiles in the lunar regolith, Solar Syst. Res., 2002, vol. 36, no. 1, pp. 1–11.CrossRefGoogle Scholar
  11. 11.
    Dikov, Yu.P., Gerasimov, M.V., Yakovlev, O.I., and Ivanov, A.V., Valence state of iron in a condensate from the Luna 16 regolith, Petrology, 2009, vol. 17, no. 5, pp.429–438.CrossRefGoogle Scholar
  12. 12.
    Dowty, E., Keil, K., and Prinz, M., Major-element vapor fractionation on the lunar surface: an unusual lithic fragment from the Luna 20 fines, Lunar Planet. Sci. Conf., 1973, vol. 21, no. 1, pp. 91–96.Google Scholar
  13. 13.
    Dran, J.C., Durrieu, L., Jouret, C., and Maurette, M., Habit and texture studies of lunar and meteoritic materials with a 1 MeV electron microscope, Earth Planet. Sci. Lett., 1970, vol. 9, no. 5, pp. 391–400.CrossRefGoogle Scholar
  14. 14.
    Engelhardt, W.V. and Stengelina, R., Normative composition and classification of lunar igneous rocks and glasses, II. Lunar glasses, Earth Planet. Sci. Lett., 1981, vol. 52, no. 1, pp. 55–66.CrossRefGoogle Scholar
  15. 15.
    Engelhardt, W.V., Luft, E., Arndt, J., et al., Origin of moldavites, Geochim. Cosmochim. Acta, 1987, vol. 51, no. 6, pp. 1425–1443.CrossRefGoogle Scholar
  16. 16.
    Florensky, P.V. and Dikov Yu.P., Genesis of tektites: reason of their composition and structure, Geokhimiya, 1981, vol. 6, pp. 809–819.Google Scholar
  17. 17.
    Florensky, K.P. and Nikolaeva, O.V., Volatile components and continental matter of planets, Geokhimiya, 1984, no. 9, pp. 1251–1267.Google Scholar
  18. 18.
    Fredriksson, K., Nelen, J., and Melson, W.G., Petrography and origin of lunar breccias and glasses, Geochim. Cosmochim. Acta. Suppl., 1970, vol. 1, p. 419.Google Scholar
  19. 19.
    Gornostaeva, T.A., Mokhov, A.V., and Gornostaev, A.N., Ultrasound extractor for sample preparation in transmission electron microscope, VII Vseross. konf. po rentgenospektral’nomu analizu (7th All-Russian Conference on X-Ray Analysis), Novosibirsk: 2011, p. 53.Google Scholar
  20. 20.
    Gornostaeva, T.A., Mokhov, A.V., Kartashov, P.M., Bogatikov, O.A., The protective role of glass film over the surface of metallic particles of the lunar regolith, Dokl. Earth Sci., 2014, vol. 459, pp. 1457–1459.CrossRefGoogle Scholar
  21. 21.
    Gornostaeva, T.A., Mokhov, A.V., Kartashov, P.M., and Bogatikov, O.A., Condensate glasses from the Zhamanshin Crater. I. Irghizites, Petrology, 2016, vol. 24, no. 1, pp. 1–20.CrossRefGoogle Scholar
  22. 22.
    Gornostaeva, T.A., Mokhov, A.V., Kartashov, P.M., and Bogatikov, O.A., Condensate glasses from the Zhamanshin Crater. II. Zhamanshinites, Petrology, 2017, vol. 25, no. 1, pp. 1–22.CrossRefGoogle Scholar
  23. 23.
    Gornostaeva T.A., Mokhov A.V., Kartashov P.M., Bogatikov O.A. Impactor type and model of the origin of the Zhamanshin Astrobleme, Kazakhstan, Petrology, 2018, vol. 26, no. 1, pp. 82–95.CrossRefGoogle Scholar
  24. 24.
    Herzog, G.F., Delaney, J.S., Lindsay, F., et al., Magnesium and silicon isotopes in hasp glasses from apollo-16 lunar soil 61241, 43rd Lunar Planet. Sci. Conf., 2012, pp. 1579–1580.Google Scholar
  25. 25.
    Horz, F., Fechtig, H., Janicke, J., and Schneider, E., Morphology and chemistry of projectile residue in small experimental impact craters, 14th Lunar Planet. Sci. Conf. Proc., 1983, vol. 14, pp. B353–B363.Google Scholar
  26. 26.
    Ivanov, A.V., Role of vaporization in the formation of chemical composition of lunar glasses, Geokhimiya, 1975, no. 8, pp. 1150–1153.Google Scholar
  27. 27.
    Ivanov, A.V., Volatiles in Lunar Regolith Samples: A Survey, Solar System Res., 2014, vol. 48, no. 2, pp. 113–129.CrossRefGoogle Scholar
  28. 28.
    Ivanov, A.V. and Florensky, K.P., The role of vaporization processes in lunar rock formation, 6th Lunar Planet. Sci. Conf. Proc., 1975, vol. 6, pp. 1341–1350.Google Scholar
  29. 29.
    Izokh, E.P., Petrochemistry of rocks of targets, impactites, and tektites of the Zhamanshin astrobleme, Kosmicheskoe veshchestvo i Zemlya (Cosmic Matter and Earth), Novosibirsk: Nauka, Sib. Otd-nie, 1986, pp. 159–203.Google Scholar
  30. 30.
    Izokh, E.P. and Le Dyk, An., Tectites of Vietnam. Hypothesis of Cometary Transportation, Meteoritika, 1983, vol. 42, pp. 158–169.Google Scholar
  31. 31.
    Kartashov, P.M., Mokhov, A.V., Gornostaeva, T.A., et al., Mineral phases on the fracture of a glass particle and in the fines of a Luna 24 regolith sample, Petrology, 2010, vol. 18, no. 2, pp. 107–125.CrossRefGoogle Scholar
  32. 32.
    Keller, L.P. and McKay, D.S., The origin of amorphous rims on lunar plagioclase grains: solar wind damage or vapor condensates, 54th Annual Meet. Meteorit. Soc., 1991, vol. 766, p. 114.Google Scholar
  33. 33.
    Keller, L.P. and McKay, D.S., Impact glasses and vapor condensates in Apollo-11 soil, Lunar Planet. Sci. Conf. Abstracts, 1992a, vol. 23, pp. 673–674.Google Scholar
  34. 34.
    Keller, L.P. and McKay, D.S., Micrometer-sized glass spheres in Apollo-16 soil 61181: implications for impact volatilization and condensation, Lunar Planet. Sci. Conf., 1992b, vol. 22, pp. 137–141.Google Scholar
  35. 35.
    Keller, L.P. and McKay, D.S., The nature and origin of rims on lunar soil grains, Geochim. Cosmochim. Acta, 1997, vol. 61, no. 11, pp. 2331–2341.CrossRefGoogle Scholar
  36. 36.
    Koeberl, C., Blue glass: a new impactite variety from Zhamanshin Crater, U.S.S.R, Geochim. Cosmochim. Acta, 1988, vol. 52, no. 3, pp. 779–784.CrossRefGoogle Scholar
  37. 37.
    Lushnipov, A.A., Negin, A.E., Pakhomov, A.V., and Smirnov B.M., Aerogel structures in gases, Usp. Fiz. Nauk, 1991, vol. 161, no. 2, pp. 113–123.CrossRefGoogle Scholar
  38. 38.
    Ma, M.-S., Schmitt, R.A., Warner, R.D., et al., Genesis of Apollo-15 olivine normative mare basalts: trace elements correlations, 9th Lunar Planet. Sci. Conf. Proc., 1978, vol. 9, pp. 523–533.Google Scholar
  39. 39.
    Magna, T., Deutsch, A., Mezger, K., et al., Lithium in tektites and impact glasses: implications for sources, histories and large impacts, Geochim. Cosmochim. Acta, 2011, vol. 75, no. 8, pp. 2137–2158.CrossRefGoogle Scholar
  40. 40.
    Margolis, S.V., Claeys, P., and Kyte, F.T., Microtektites, microkrystites, and spinels from a Late Pliocene asteroid impact in the Southern Ocean, Science, 1991, vol. 251, no. 5001, pp. 1594–1597.CrossRefGoogle Scholar
  41. 41.
    Markova, O.M., Yakovlev, O.I., Semenov, G.A., and Belov, A.N., Some general results of experiments on evaporation of natural melts in the Knudsen cell, Geokhimiya, 1986, no. 11, pp. 1559–1569.Google Scholar
  42. 42.
    Mokhov, A.V., Analytical electron microscopy in study of ultra-dispersed fraction of lunar soil, I Vseross. molodezh. konf. “Mineraly, stroenie, svoistva, metody issledovaniya” (1st All-Russian Youth Conference on Minerals, Structure, Properties, and Methods of Study), Il’meny: 2009, pp. 42–45.Google Scholar
  43. 43.
    Mokhov, A.V., Kartashov, P.M., Gornostaeva, T.A., Bogatikov, O.A., Native ytterbium of regolith AS Luna-24, Dokl. Earth Sci., 2011, vol. 441, pp. 1692–1694.CrossRefGoogle Scholar
  44. 44.
    Naney, M.T., Crowl, D.M., and Papike, J.J., The Apollo-16 drill core: statistical analysis of glass chemistry and the characterization of a high alumina-silica poor (HASP) glass, Lunar Planet. Sci. Conf. Proc., 1976, vol. 7, pp. 155–184.Google Scholar
  45. 45.
    Naughton, J.J., Hammond, D.A., Margolis, S.V., and Muenow, D.W., The nature and effect of the volatile cloud produced by volcanic and impact events on the Moon as derived from a terrestrial volcanic model, 3rd Lunar Planet. Sci. Conf. Proc., 1972, vol. 3, pp. 2015–2024.Google Scholar
  46. 46.
    Noble, S.K., Pieters, C.M., Taylor, L.A., et al., The optical properties of the finest fraction of lunar soil: implications for space weathering, Meteorit. Planet. Sci., 2001, vol. 36, no. 1, pp. 31–42.CrossRefGoogle Scholar
  47. 47.
    Norris, J.A., Keller, L.P., and McKay, D.S., Impact glasses from the ultrafine fraction of lunar soils, 24th Lunar Planet. Sci. Conf. Proc., 1993, vol. 24, pp. 1093–1094.Google Scholar
  48. 48.
    O’Keefe, J.A., Tektite glass in Apollo-12 sample, Science, 1970, vol. 168, pp. 1209–1210.CrossRefGoogle Scholar
  49. 49.
    O’Keefe, J.A., Tectites and their Origin, New York: Elsevier, 1976.Google Scholar
  50. 50.
    Otmakhov, V.I., Varlamova, N.V., Manankov, A.N., and Lapova, T.V., Physicochemical studies of tektites for cosmic monitoring, Izv. Tomskogo Politekhn. Univ., 2006, vol. 309, no. 5, pp. 40–44.Google Scholar
  51. 51.
    Papike, J.J., Spilde, M.N., Adcock, C.T., et al., Trace element fractionation by impact-induced volatilization-sims study of lunar HASP glasses, Lunar Planet. Sci. Conf. Proc., 1997, vol. 28, p. 1059.Google Scholar
  52. 52.
    Petaev, M.I. and Wood, J.A., The condensation with partial isolation (CWPI) model of condensation in the solar nebula, Meteorit. Planet. Sci., 1998, vol. 33, no. 5, pp. 1123–1137.CrossRefGoogle Scholar
  53. 53.
    Petaev, M.I., Jacobsen, S.B., and Huang, S., Testing models of moon origin: condensation of impact-vaporised bulk silicate earth material, 45th Lunar Planet. Sci. Conf. Proc., 2014, pp. 2316–2317.Google Scholar
  54. 54.
    Pratesi, G., Viti, C., Cipriani, C., and Mellini, M., Silicate-silicate liquid immiscibility and graphite ribbons in Libyan desert glass, Geochim. Cosmochim. Acta, 2002, vol. 66, no. 5, pp. 903–911.CrossRefGoogle Scholar
  55. 55.
    Reid, A.M., Warner, J., Ridley, W.I., and Brown, R.W., Major element composition of glasses in three Apollo-15 soils, Meteoritics, 1972, vol. 7, no. 3, pp. 395–415.CrossRefGoogle Scholar
  56. 56.
    Rietmeijer, F.J.M. and Karner, J.M., Metastable eutectics in the Al2O3–SiO2 system explored by vapor phase condensation, J. Chem. Phys., 1999, vol. 110, pp. 4554–4558.CrossRefGoogle Scholar
  57. 57.
    Rietmeijer, F.J.M., Nuth, J.A., Rochette, P., et al., Deep metastable eutectic condensation in Al–Fe–SiO–H2–O2 vapors: implications for natural Fe-aluminosilicates, Am. Mineral., 2006, vol. 91, pp. 1688–1698.CrossRefGoogle Scholar
  58. 58.
    Roedder, E. and Weiblen, P.W., Apollo-17 “orange soil” and meteorite impact on liquid lava, Nature, 1973, vol. 244, pp. 210–212.CrossRefGoogle Scholar
  59. 59.
    Skublov, G.T. and Tyugai, O.M., Petrochemical model of the formation of tektite-like glasses of the Zhamanshin Crater and their relation with lunar impact genesis, Zap. Ross. Mineral. O-va, 2004, no. 6, pp. 95-117.Google Scholar
  60. 60.
    Vaniman, D.T., Glass variants and multiple hasp trends in apollo 14 regolith breccias, Lunar Planet. Sci. Conf. Proc., 1990, vol. 20, pp. 209–217.Google Scholar
  61. 61.
    Walker, R. and Yuhas, D., Cosmic ray track production rates in lunar materials, Geochim. Cosmochim. Acta. Suppl., 1973, vol. 3, pp. 2379–2389.Google Scholar
  62. 62.
    Warren, P.H. and Kallemeyn, G.W., Que93069: a lunar meteorite rich in HASP glasses, 26th Lunar Planet. Sci. Conf. Proc., 1995, vol. 26, pp. 1465–1466.Google Scholar
  63. 63.
    Warren, P.H., Lunar rock-rain: diverse silicate impact-vapor condensates in an Apollo-14 regolith breccias, Geochim. Cosmochim. Acta, 2008, vol. 72, pp. 3562–3585.CrossRefGoogle Scholar
  64. 64.
    Wentworth, S.J. and McKay, D.S., Glasses in ancient and young Apollo-16 regolith breccias: populations and ultra mg glasses, 18th Lunar Planet. Sci. Conf. Proc., 1988, vol. 18, pp. 67–77.Google Scholar
  65. 65.
    Wentworth, S.J., Keller, L.P., McKay, D.S., and Morris, R.V., Space weathering on the Moon: patina on Apollo 17 samples 75075 and 76015, Meteorit. Planet. Sci., 1999, vol. 34, no. 4, pp. 593–603.CrossRefGoogle Scholar
  66. 66.
    Wood, J.A. and Hashimoto, A., Mineral equilibrium in fractionated nebular systems, Geochim. Cosmochim. Acta, 1993, vol. 57, no. 10, pp. 2377–2388.CrossRefGoogle Scholar
  67. 67.
    Yakovlev, O.I., Dikov, Yu.P., and Gerasimov, M.V., Differentiation caused by impact-induced vaporization during the Earth’s accretion, Geochem. Int., 2000, vol. 38, no. 10, pp. 937–954.Google Scholar
  68. 68.
    Yakovlev, O.I., Dikov, Yu.P., Gerasimov, M.V., et al., Experimental Investigation of Factors Controlling the Composition of Glasses from the Lunar Regolith, Geochem. Int., 2003, vol. 4, no. 5, pp. 417–430.Google Scholar
  69. 69.
    Yakovlev, O.I., Gerasimov, M.V., and Dikov, Yu.P., Experimental melting of obsidian by laser pulses with implications for the analysis of the composition of impact glasses, Geochem. Int., 2005, vol. 43, no. 3, pp. 211–221.Google Scholar
  70. 70.
    Yakovlev, O.I., Gerasimov, M.V., and Dikov, Yu.P., Temperatures of formation of HASP and GASP particles, 40th Lunar Planet. Sci. Conf. Proc., 2009, vol. 40, p. 1261.Google Scholar
  71. 71.
    Yakovlev, O.I., Gerasimov, M.V., Dikov, Yu.P., Compositions of lunar condensates and conditions of their formation on the surface, XVI Ross. soveshch. po eksperimental’noi mineralogii (16th All-Russian Conference on Experimental Mineralogy), 2010, pp. 291–293.Google Scholar
  72. 72.
    Yakovlev, O.I., Gerasimov, M.V., and Dikov, Yu.P., Conditions of condensate rim formation on the surface of lunar regolith particles, Geochem. Int., 2011, vol. 49, no. 10, pp. 953–966.CrossRefGoogle Scholar
  73. 73.
    Yakovlev, O.I., Dikov, Yu.P., Gerasimov, M.V., and Vlottska, F., Vaporization of aluminum from silicate melt, Geochem. Int., 1997, vol. 35, no. 12, pp. 1046–1059.Google Scholar
  74. 74.
    Yakovlev, O.I., Kosolapov, A.I., Kuznetsov, A.V., and Nusinov, M.D., Results of study of fractional evaporation of basalic melt in vacuum, Dokl. Akad. Nauk SSSR, 1972, vol. 206, no. 4, pp. 970–973.Google Scholar
  75. 75.
    Zel’dovich, B. and Raizer, Yu., Fizika udarnykh voln i vysokotemperaturnykh gidrodinamicheskikh yavlenii (Physics of Impact Waves and High-Temperature Hydrodynamic Phenomena), Moscow: FIZMATLIT, 2008.Google Scholar
  76. 76.
    Zolensky, M.E. and Koeberl, C., Why are blue zhamanshinites blue? Liquid immiscibility in an impact melt, Geochim. Cosmochim. Acta, 1991, vol. 55, no. 5, pp. 1483–1486.CrossRefGoogle Scholar

Copyright information

© Pleiades Publishing, Ltd. 2019

Authors and Affiliations

  • T. A. Gornostaeva
    • 1
  • A. V. Mokhov
    • 1
    Email author
  • P. M. Kartashov
    • 1
  • O. A. Bogatikov
    • 1
  1. 1.Institute of the Geology of Ore Deposits, Petrography, Mineralogy, and Geochemistry (IGEM), Russian Academy of SciencesMoscowRussia

Personalised recommendations