Geochemistry of aerodynamically distorted Australasian

Geochemistry of aerodynamically distortedAustralasian microtektites: Implications forejecta on Mars and Venus

N G RUDRASWAMI1,* , M SHYAM PRASAD

1, SRIDHAR D IYER1, M PANDEY

1,CHRISTOPH HELO

2 and DIPAK KUMAR PANDA3

1National Institute of Oceanography (Council of ScientiBc and Industrial Research), Dona Paula,Goa 403 004, India.2Institute of Geosciences, Johannes Gutenberg University, 55128 Mainz, Germany.

3Physical Research Laboratory, Navrangpura, Ahmedabad 380 009, India.*Corresponding author. e-mail: [email protected]

MS received 18 September 2020; revised 22 January 2021; accepted 26 January 2021

We reportmicrotektites recovered froma large area of the deep seaCoor (Central IndianOcean) that appear tohave undergone aerodynamic distortion during re-entry into the atmosphere. Considering the geographiclocations, stratigraphic position and chemical compositions these glassy forms belong to the Australasiantektite strewn Beld. The microtektites are elongated to lengths of cms, sometimes Cattened, bent, folded andfused at both ends suggesting that this side could have been the Earth-facing side during the re-entry. Thepresence of Cow lines and distortional features are indicative of high atmospheric pressures experienced by themicrotektites. The location where these microtektites were recovered constitute distal ejecta, and the shapedistortion, that occurred during re-entry of the ejecta, seems to have aAected only a few amongst the extensivecollection of microtektites. Most of the specimens contain lechatelierite inclusions and higher volatile oxides,which are indicative of incomplete homogenizationaftermeltingand lower temperatures of formationvis-�a-visother specimens at the same location. The element distribution patterns in aerodynamically distortedmicrotektites suggest that ablation was similar to normal spherical tektites in which volatile elements arepreserved. In contrast, aerodynamically ablated forms of Australasian ejecta show skin melting where thinlayers of the anterior portions of samplesCowback giving rise to the familiar button shapes.Our observationofdelicate, elongated, Cattened, and viscously deformed specimens is perhaps the Brst to imply that at the distalend of ejecta, each spot in the specimens has undergone different levels of trajectories, heating and ablation.These investigations could have implications to understand ejecta emplacement characteristics on planetarysurfaces that contain appreciable atmospheres such as Mars and Venus.

Keywords. Tektites; deep-sea; microtektites; Mars; Venus.

1. Introduction

The Australasian tektite strewn Beld is the largestof the four known Belds so far from the marinesediments (Folco et al. 2009). The Beld displays

classic distribution patterns of tektites that rangefrom kilogram-sized tektites in the Indochinaregion to the smallest tektites (microtektites)found in the southernmost part of the strewn Beldin Australia (Von Koenigswald 1960; O’Keefe

J. Earth Syst. Sci. (2021) 130:76 � Indian Academy of Scienceshttps://doi.org/10.1007/s12040-021-01589-z (0123456789().,-volV)(0123456789().,-volV)

1976). With an increase in radial distance from thesource area, the strewn Beld also displays dramaticchanges in shapes of the tektites ranging fromblocky and layered Muong Nong-types in thenorth, through the splash forms in theIndo-Malaysian region to the celebrated aerody-namically ablated tektites found commonly inAustralia. Most investigators favour the Indochinaregion as the location for the source crater for thisstrewn Beld (Glass and Pizzuto 1994; Ma et al.2004; Glass and Koeberl 2006; Prasad et al. 2007).Over *100,000 tektites have been found fromAustralia (McNamara and Bevan 2001), andablated tektites deBned as a solid splash intovarious shapes during its re-entry into Earth’satmosphere have been found very commonly andexclusively in Australia (Glass et al. 1996). Incontrast, a few incipient ablated forms are foundamong javanites, one billitonite and one phillip-inite (Chapman 1964) and one specimen fromFlores, west of Java (McCall 2001). However,there are no reports of ablated samples fromIndochina, Malaysia, or China. It was therefore,suggested that the absence of ablated tektites inIndochina could be due to the steepness of the falltrajectory, thereby obliterating all evidence ofablation (Chapman 1964) or the line from Cen-tral Indian Ocean (CIO) through Java andFlores, to Negros Island represents the northernboundary for re-entry and ablation (McCall2001).Additionally, at a distance of about *3,500 km

from the expected crater is the re-entry point of thetektites which underwent a second stage of meltingleading to their ablation (Ford 1988), and this alsomarks the northern boundary for the ablated tek-tites. The northernmost ablated tektite found sofar is the one discovered in the sediments of theCIO (Prasad and Rao 1990). Chapman and Lar-son’s (1963) arc jet ablation experiments have beenfundamental to understand that tektites werethrown up in space as melt droplets, solidiBed ascold, rigid bodies and underwent a second phase ofmelting during their descent through the atmo-sphere leading to skin melting and Cange forma-tion. Further, those authors also establishedconstraints on the entry conditions based on theablation markings on the anterior of tektites.Detailed studies by Delano (1992) suggested thatAustralite Canges can be used as Cight datarecorders. Traditionally, only tektites and thosedominantly from southern Australia are found toshow evidence of ablation.

In more than four decades of tektite research,several thousands of Australasian microtektiteshave been examined (Glass 1972, 1974, 1990, 2003;Glass et al. 1979; Simonson and Glass 2004), andexclusive morphological and compositional studieswere undertaken on microtektites (e.g., Glass 1974;Prasad and Sudhakar 1998; Prasad and Khedekar2003). However, in none of these studies, ablationphenomenon on microtektites has been reported,although, the re-entry conditions for the tektitesand microtektites are expected to be similar –

considering that they belong to the same impactevent. We collected a spade core in the CIO andexamined a large surface area of the ocean Coor(50 9 50 cm sampling area) for Australasianmicrotektites. Of the *2,700 microtektites of[250lm size recovered from this sample, we found asmall percent of specimens to show ablation fea-tures and distortion due to aerodynamic processes.These phenomena on microtektites depict aerody-namic processes that were relatively unknown sofar in the Australasian tektite strewn Beld. Thepresent investigation, coupled with geochemicaland petrographic data of microtektites, revealsaerodynamical features along with spherical tek-tites from the same CIO region and highlights theirdistinctive geochemical behaviour and their entryparameters of the tektites. This study will improvethe database of these aerodynamically distortedmicrotektites, unravel their composition, and mostlikely clarify the nature of entry parameter, sourcematerial and impacting extraterrestrial bodies withwider implications pertaining to different planetaryatmospheres.

2. Sampling method

A 38 cm long spade core sample was collected fromthe CIO during the 62nd cruise of A.A. Sidorenko(April–May 2003), a vessel chartered by theGovernment of India, at the location 17o57.2230Sand 78o01.7110E from a water depth of *4,700 m.This core lies well within the geographic limits ofthe Australasian tektite strewn Beld (Bgure 1). Thesediment type here is red clay, and the sedimen-tation rates are very low. Therefore, it was antici-pated that even a short core length of *38 cmwould penetrate *0.77 Ma horizon of the Aus-tralasian tektite strewn Beld. The dimensions of thespade core (50950 cm) further allowed a largequantity of samples to be collected at a singlelocation.

76 Page 2 of 21 J. Earth Syst. Sci. (2021) 130:76

The core was sub-sampled at two different levels:

(1) The bulk of the sample was sub-sectioned at5 cm intervals, and this was wet sieved using a125 lm mesh size sieve. The [250 lm driedfraction of this material was observed under abinocular microscope, and the microtektiteswere handpicked using a wet brush. In all,from this fraction *2,700 Australasianmicrotektites of diameters [250 lm wereisolated.

(2) An acrylic tube having an internal diameter of*10.5 cm was inserted into this core immedi-ately after the core arrived onboard, and thismaterial was sub-sampled at an interval of *2cm for higher resolution and for identifying thepeak abundance horizon of microtektites.From this fraction, initially, the microtektites

were wet sieved in mesh size of *125 lm. Thedried coarse fraction was further sieved in asieve having *250 lm mesh size, and themicrotektites of [250 lm were handpickedunder a binocular microscope. Subsequently,the 125–250 lm coarse fraction was subjectedto heavy liquid separation, and microtektitesin this size fraction were isolated from the‘sinks’ using a binocular microscope having amagniBcation up to 609.

The depth-wise recovery of microtektites at 2 cminterval (Bgure 2) indicates a peak abundance ofmicrotektites between 30 and 32 cm. This gives asedimentation rate (considering the *0.77 Ma ageof the Australasian tektite event) of *40 cm/106

yr. Given the down core microtektite abundancedata, it appears that either we have just sampled or

Figure 1. Map of the Australasian tektite strewn Beld after Glass and Koeberl (2006) and Folco et al. (2016) along with sitelocation of potential source crater, Transantarctic microtektites, CIO microtektites, and microtektites found on land. The thickline denotes the boundaries of the strewn Beld. Star denotes the present location where ablated microtektites are discovered. Thecircle above the star shows the site for the only ablated tektite found in the CIO (Prasad and Rao 1990). The oval portion inthe Indochina region is the possible location of the source crater as identiBed by several investigators. The darkened portion inthe southern part of Australia denotes locations where aerodynamically ablated Australite tektites have been traditionally found.Australasian microtektites have been found in Antarctica (Folco et al. 2016) in the area shown above, and thus extending thestrewn Beld to *10,000 km in an N–S direction.

J. Earth Syst. Sci. (2021) 130:76 Page 3 of 21 76

were very close to the *0.77 Ma stratum of theAustralasian impact event. With the availabledata, it is estimated that the abundance ofmicrotektites of[125 lm diameter at this locationis *5/cm2 (Prasad et al. 2007). The microtektitesfrom this core have all the morphological featuresattributed to Australasian microtektites by earlierinvestigators (Glass 1974; Prasad and Sudhakar1999; Prasad et al. 2007).

3. Analytical techniques

Twenty-one microtektites mounted on epoxy werepolished and analyzed using a Cameca SX-100electron probe microanalyzer (EPMA) equippedwith four wavelength-dispersive X-ray spectrome-ters (WDS) at the Physical Research Laboratory(PRL), Ahmedabad, India. The analyses wereperformed using an accelerating voltage of *15kV, an electron beam current of *15 nA, and adefocused beam size of *10 lm. The sample

counting times maintained were 20 s for peak and10 s for the background. Na is the Brst element tobe measured, followed by other major elements, toavoid loss of this volatile element. Mineral stan-dards were used for instrumental calibration andchecked against USNM 2213 (Jarosewich et al.1980), a synthetic tektite standard specially madeby Corning Glass company. Online program PAPwas used for raw data reduction. Multiple analysesof 5–10 spots were analyzed on each microtektite.In addition to the above, from the *2,700

microtektites isolated, we found that some con-tained features which could have been generateddue to ablation or aerodynamic distortion duringre-entry. These features were examined with aJEOL JSM-5800-LV SEM at the National Instituteof Oceanography, Goa, India using Energy-disper-sive X-ray Spectroscopy (EDS) that is attachedwith the system, and with USNM-2213 as a stan-dard. Subsequently, nine of these specimens (sixspherical and three elongated, bent ones) weremounted in epoxy, polished and analyzed with anelectron microprobe under the same operatingconditions as mentioned above.The water content of the specimens was quan-

tiBed at the University of Mainz, Germany, using aFourier-transform infrared spectrometer (FTIR)coupled with an optical microscope system. Thesamples were carefully embedded into epoxyand grounded to a thickness of 150–200 lm. Thewafers were polished with a *1 lm diamondsuspension. The analysis was carried out inthe transmission mode, and a KBr-beamsplitterwas used. A background was collected before eachmeasurement. After linear baseline subtraction,the total water concentration c(H2O) was deter-mined from the intensity of the absorbanceband at 3,600 cm–1, characteristic for the funda-mental O–H stretching vibration, using the Lam-bert–Beer law:

c H2Oð Þ ¼ 1802� A3550

d � q� e3550;

to with A denoting the peak height, i.e., theabsorbance, d the sample thickness, q the densityand e the linear molar absorption coefBcient,respectively. Density was calculated based on themethod of Lange and Carmichael (1990), and themolar absorption coefBcient of 68 L mol–1 cm–1 fordacites was obtained from Yamashita et al. (1997).Raman spectroscopy analyses were conducted on

a Jobin Yvon Horiba confocal instrument. Polishedspecimens were excited with a 532 nm green laser

Figure 2. Down core distribution of Australasian microtek-tites ([125 lm size) from the sediment core AAS 62/56 in theCIO. The peak of microtektite abundance appears at 30–32 cmwhich is the horizon corresponding to the 0.77 Ma Aus-tralasian tektite strewn Beld event.


through a 509 objective. Exposure time was sev-eral tens of seconds with usually three repetitionsto improve the signal-to-noise ratio. Measurementswere performed using a 300 l/mm grating, a 400lm confocal hole and a slit of 100 lm. Theinstrument was calibrated using a silicon standard.The data collected within a single spectral win-dow were subjected to wavenumber-dependentintensity correction to the spectra based on Long(1977).

4. Results

4.1 Aerodynamic features of microtektites

Thirty-seven specimens with indications of aero-dynamic features were found to co-occur with theAustralasian microtektites. Therefore, strati-graphically they are expected to be from the samehorizon as the Australasian tektites. Morphologi-cally, these specimens are club-shaped and elon-gated Blamentous, Cattened ones that are eitherstraight or bent at the centre, and some are foldedwhere both the ends meet in the form a loop(Bgures 3–5). The description, shape and size of thestudied microtektites are given in table 1. All thespecimens considered to be ablated are eithercolourless or have a light yellowish colour, exceptthe larger and Catter specimens are either colour-less, glassy or a shade of pale yellow. All specimenshave Cow lines on their surfaces with most of themretaining their spherical shape (Bgures 3–5). Fur-ther, the Cow lines generally culminate at oraround the equatorial region and on some speci-mens they run along the length of the particle.Whereas, on the aerodynamically distorted tektitesfound in Australia, it is easy to identify the anteriorand the posterior sides of the specimens (Prasadand Rao 1990; Glass et al. 1996). It is however,difBcult to assign either the anterior or posteriorsides to the club-shaped, elongated needles or bentspecimens as observed in this study.Interestingly, however, numerous Cow lines are

observed on only one side of all of these specimens(Bgures 3c and 5). For example, the Cow lines on aclub-shaped sample veer around a lechatelieriteinclusion and terminate at the narrow end of thespecimen (Bgure 3c–f). Some of the specimens areelongated to several mm, giving them a Blamentousappearance while others are elongated, club-shapedor folded in the centre and with the ends weldedtogether in a loop. The longest sample observed is

*1 cm (Bgure 5). Some specimens are folded at theend, giving an appearance of a hockey-stick, orpossess a bulge at the centre (these tend to be bow-shaped) (Bgure 5c). All the specimens invariablyhave numerous directional Cow lines on their sur-faces. These Cow lines are multiple, very closelyspaced and are parallel to the longer dimension ofthe samples. Two of these specimens havelechatelierite grains exposed on their surfaces(Bgures 3c and 5a).

4.2 Chemical composition of microtektites

In this study, all the 37 unpolished microtektitesanalyzed using an SEM-EDS have Cow lines withmany having spherical shapes and some unusualdistorted forms. It should be noted even whencalled ‘microtektites’ they have Cow lines featureson their surface with spherical shape unless other-wise speciBed. Further, polished sections of 21spherical microtektites from the same sample werealso analyzed with an electron microprobe. Basedon MgO (wt.%) content as deBned by Glass et al.(2004), we divided our microtektites into twogroups: normal and intermediate type. We havefour spherical intermediate-type microtektites thathave MgO[6 and\10 wt.%, while the rest of thespherical shape microtektites are normal-type withMgO \6 wt.% (table 2). The current study doesnot have any representation of bottle-green or highMgO ([10 wt.%) microtektites.In addition, nine distorted specimens that were

analyzed with an electron microprobe indicatethese to be of normal-type based on their MgOcontent (table 3). We compared the compositionof the spherical and distorted specimens with thoseof Australasian microtektites (table 4). All themicrotektites, regardless of their shapes, havechemical compositions well within the rangeassigned for Australasian microtektites (Glass1990; Glass and Koeberl 2006), whereas, thespherical microtektites display a broader range inSi, Al and Fe contents and comparatively lower Naand K contents (tables 2 and 4). In the distortedmicrotektites, the chemical composition range isalso well within the range of the Australasianmicrotektites (tables 3 and 4). The number ofspherical microtektites analyzed accounts for amore comprehensive compositional range, how-ever, these have lower Na (0.2–0.6 wt.%), and K(0.6–2.1 wt.%) contents, when compared withaerodynamically distorted microtektites that have


higher Na (0.8–1.5 wt.%) and K (1.9–2.8 wt.%)content. The water content of the two analyzedtektites is extremely low (0.018 and 0.007 wt.%)which is unusual if the SiO2-rich melts were ofvolcanic origin, but the values are within the rangeof water compositions observed in tektites (Bernanand Koeberl 1997).Interestingly, the anterior side of the lone ablated

tektite from the Indian Ocean was found to havehigher Na and K contents as compared with theunablated posterior part of the specimen (Prasad

and Rao 1990). This was contrary to expectationsespecially considering that the ablated anteriorportion has undergone melting and heating whichshould result in lowered contents of volatile ele-ments (Glass et al. 1996). In our study, the chemicalcompositions of all the ablated specimens have amuch narrower range of Na and K, and all of themconform to the ‘normal’ sub-class of microtektites incomposition (table 3). Their enhanced soda-potashcontents indicate that these specimens haveencountered lesser heating than the other associated

Figure 3. (a) S1P5: Spherical ablated microtektite. A bald spot on the top is seen, which could be the stagnation point.Close-spaced Cow lines are visible on the top half of the specimen. (b) S1P1: Spherical ablated microtektite. The Cow of materialis seen to moving away from the top of the specimen – all Cow apparently stops near the equatorial region of the specimen.(c) S2P1: Club-shaped, suppressed microtektite with close-spaced Cow lines on one side of the specimen generated during re-entry. The Cow lines veer around a large, blocky, exposed lechatelierite particle near the narrow end of the specimen.(d) MagniBed picture of the front end or the anterior part of the specimen from where the spiraling, close-spaced Cow lines seemto emanate. (e) Another view of the above club-shaped specimen showing its suppressed shape. (f) MagniBed picture of thelechatelierite grains which could be seen in Bgure (c and f) above, Cow lines have gone around these exposed inclusions.


microtektites. The microtektites with their uniqueappearance exhibit major and minor chemicalcompositions similar to the Australian andTransantarctic microtektites (Bgure 6). Whencompared to glass micrometeorites from deep-seasediments and Antarctica, they seem to be muchenriched in Si and depleted in Mg, setting themapart from each other (Bgure 7).From the FTIR graph on aerodynamically dis-

torted tektites, it can be observed that prominentpeaks are present at around 3,600 cm–1, which ischaracteristically attributed due to fundamentalO–H stretching vibration. The peak is sharp and

more prominent in IyPh˙1-avg as compared toIyPh˙2-avg (Bgure 9).

4.2.1 Occurrence of Lechatelierite

Lechatelierite (or fused quartz) grains are foundexposed on two non-spherical ablated specimensand in the polished sections of three ablated spec-imens (Bgures 5, 8 and tables 1, 3). Lechatelieriteis considered as one of the deBnitive qualities oftektites and is the commonest inclusion in tek-tites (O’Keefe 1976). Barnes (1963) observed

Figure 4. Spherical microtektites showing signs of aerodynamic ablation. (a) S1P4: Oval, suppressed microtektite with Cow lineson the exposed surface. These Cow lines are observed on only this side of the specimen. (b) MagniBed picture of the Cow linesshown in Bgure (a). (c) S1P3: Oval-shaped microtektite with thicker Cow lines, the Cow lines are seen only in this side of thespecimen. (d) MagniBed picture of the Cow lines as shown in Bgure (c) above. The other side of the specimen shows some incipientdevelopment of Cange. (e) S1P6: Circular Cow lines all over the specimen with material thrown back at the left side. Consideringthis feature it is assumed that the right side of the specimen was Earth facing during re-entry. (f) MagniBed image of the Cowlines shown in Bgure (e).


lechatelierite inclusions on the posterior surfacesof australites around which Cow structure wasobserved. In splash-form tektites, lechatelierite isseen to have long tails, drawn-out parallel to thestriae and often contorted and the length may beup to half mm (Barnes 1958). Kinnunen (1990)identiBed lechatelierite inclusions of *2–500 lmsize in indochinites. Interestingly, in moldavites,lechatelierite occurs as *1 mm size isometricparticles or as long Cat Bbers formed due to shearCow. The width and thickness of the Bbers usu-ally vary from hundredths to a few tenths of mmand are up to a few centimetres in length (Trnkaand Houzar 2002). On the non-spherical speci-mens in this study, lechatelierite is exposed andthe Cow lines veer around the lechatelieriteinclusions (Bgures 3e, f and 5a). The size of thelargest lechatelierite (quartz) blocky inclusion ina teardrop-shaped microtektite is 80 lm and themicrotektite has a length of *1.4 mm(Bgure 3c–f). The other microtektite hostinglechatelierite is Cattened, bent, inverted, elon-gated and has a long dimension of *8.9 mm andcontains directional Cow lines all over the

specimen (Bgure 5a). Three such blocky inclu-sions are found in this specimen.Lechatelierite inclusions, also observed in the

exposed interiors of spherical specimens (Bgure 8a,b) are seen as black areas within a diffuse dark,silica-rich matrix in the backscattered X-ray ima-ges. The silica-rich inclusion has a dimension of*74950 lm and the lechatelierite inclusions havesizes of 17 and 15 lm (Bgure 8b). There appears tobe a variation in the SiO2 contents of these high-Si-grains (table 3), which implies incomplete fusionwith the host microtektite. Among Australasianmicrotektites, similar grains have been observed muchsouth of CIO location among the TransantarcticMountain microtektites (Folco et al. 2009) as well asin the locations closer to the impact site, much northof the present location (Cassidy et al. 1969; Folco et al.2009).Cassidy (1964) suggested that the liquidus tem-

perature of an australite tektite is 1347±15�C andthat cristobalite is the Brst phase to crystallizeduring cooling. The ‘lechatelierite’ inclusions foundin some of the elongated microtektites (e.g.,Bgure 5a) show the Cow lines to veer over or around

Figure 5. Some of the largest non-spherical specimens were distorted by the ablation process during re-entry. (a) S3P1: Thisspecimen is bent, overturned and stuck together from both ends. This plastic deformation apparently has taken place duringimmense stress conditions during re-entry. Numerous close-spaced Cow lines are seen throughout the specimen. Lechatelierite:either one particle broken into three or three separate particles are seen near the broader portion of the specimen (see inset for amagniBed picture of lechatelierite), the Cow lines due to atmospheric interaction seem to veer around the lechatelierite particles.(b) S3P2: Elongated, somewhat tapering and platy specimen with numerous close-spaced Cow lines and (c) S3P3: Elongated,bent, plastically deformed specimen with numerous Cow lines on its surface.


Table 1. Description of ablated microtektites in AAS 62/56 ([250 lm fraction).

Sl. no.

Size

(mm) Description Remarks

1 2.9 Elongated, Cattened, folded and jointed at the ends. Close-spaced

parallel Cow lines in the ‘inside’ portion of the fold. Few Cow lines on

the ‘outside’ as well

Figure 8(e)

2 4.18 Boomerang-shaped, with a bulge at the center of the specimen. Close-

spaced parallel Cow lines throughout the specimen

Parallel Cow lines

3 0.22 Flattened sphere. Circular Cow lines on the ‘anterior’ of the specimen,

Cow lines absent on the ‘posterior’. Incipient Cange development

S1P1: Contains high-silica

inclusion (Bgures 3b and 8a)

4 0.33 Flattened oval. Close-spaced, meandering circular, Cow lines on a

major part of the anterior. Posterior devoid of Cow lines

S1P4 (Bgure 4a)

5 0.3 Flattened sphere. Few, thick circular Cow lines on the anterior.

Posterior devoid of Cow lines

S1P3: Lechatelierite and high

silica diffuse, inclusion

(Bgures 4c and 8b)

6 1.42 Club-shaped. Somewhat Cattened, parallel Cow lines seen on only one

side of the specimen along the length beginning from a whorl at the

head

S2P1, Lechatelierite exposed

(Bgure 3c–f)

7 0.25 Sphere. Circular Cow lines spreading from a part of the anterior with a

small projection and a rough area unaffected by Cow is seen here

S1P2 : High silica diffuse

inclusion

8 8.86 Elongated, Cattened, folded and jointed throughout. Thick, close

spaced parallel Cow lines on the ‘anterior’ extending throughout the

specimen.

S3P1, Lechatelierite exposed

(Bgure 5a)

9 7.37 Boomerang-shaped with a thickened center. Parallel Cow lines

emanating from the center and extend to the ends of the specimen

S3P3 (Bgure 5c)

10 0.3490.29 Slightly oval. Stagnation point visible and circular Cow lines emanating

from the specimen up to the equatorial region. Posterior shows

incipient Cange development.

S1P5 (Bgure 3a)

11 10.19 Flat, Caky, elongated, broken at the ends. Continuous, close spaced

parallel Cow lines on the specimen

S3P2 (Bgure 5b)

12 0.83 Rounded, folded in a loop and jointed at the ends. Close-spaced parallel

Cow lines at the folded ‘interior’ part of the specimen, fewer Cow lines

on the outside.

S2P3 (Bgure 8c)

13 0.98 Flattened, irregular, broken with possible Cow lines on one side of the

specimen

S2P2 (Bgure 8d)

14 0.27 9 0.31 Oval. Clear-Cange development, circular regular Cow lines on most part

of the specimen

S1P6 (Bgure 4e)

15 0.35 Bun-shaped, oval. Close-spaced circular Cow lines Flow lines

16 0.28 Oval, suppressed sphere with circular Cow lines on one side of specimen Flow lines

17 0.35 Oval bun shaped. Many close-spaced circular Cow lines on one side of

specimen

Flow lines

18 3.75 Elongated, folded in the middle with many close-spaced parallel Cow

lines

Flow lines

19 1 Sphere with a large bubble on the posterior. Flange formation is seen

with numerous circular Cow lines on the ‘anterior’

Flow lines

20 1.3 Club-shaped with longitudinal, parallel Cow lines on one side of

specimen

Flow lines

21 0.21 Bun-shaped with close-spaced circular Cow lines on one side of

specimen

Flow lines


the inclusions in the CIO collection would there-fore, indicate a temperature of the formation belowthis temperature – since the inclusions seem to beintact and non-homogenized during the secondphase of melting.

5. Discussion

5.1 Aerodynamically distorted Australasianmicrotektites

Presuming that the studied particles are notmicrotektites or tektites by virtue of the cm-size ofsome of the elongated specimens, then the otherpossibility to be considered is the particles arePele’s hair to which they bear a physical resem-blance. Pele’s hairs and tears are products of

volcanic activity found mostly in basaltic volcanicsites (Heiken 1972; DuDeld et al. 1977; Mouneet al. 2007 and references therein), while Davis andClague (2006) recovered pyroclasts (North ArchVolcanic Field, north of Oahu) consisting of glassspheres and Pele’s hair-like fragments that formedfrom mildly explosive eruptions in water depths[4500 m. Some of our more Blamentous specimensbear a physical resemblance to Pele’s hairs, but thelatter is of brown sideromelane glass with 50 wt.%SiO2, 10–11 wt.% CaO, 7–10 wt.% MgO and\0.5wt.% K2O (DuDeld et al. 1977). Therefore, con-sidering the geographic location, stratigraphicposition and major oxide compositions, Ramanspectra and lack of water, all of which are reliableindicators, the studied specimens are indeed Aus-tralasian microtektites (Bgures 9 and 10). This

Table 1. (Continued.)

Sl. no.

Size

(mm) Description Remarks

22 0.35 Bun-shaped, circular Cow lines on one side of specimen Flow lines

23 0.7 Teardrop-shaped with longitudinal, parallel Cow lines on one side of

specimen. Specimen Cattened longitudinally

Flow lines

24 1.3 Elongated teardrop with parallel Cow lines longitudinally over one side

of the specimen

Flow lines

25 1.05 Broken, club-shaped, slightly Cattened with Cow lines longitudinal and

parallel to each other.

Flow lines

26 5 Boomerang-shaped, curved, elongated specimen with longitudinal,

close-spaced, parallel Cow lines and a bump in the center of the

specimen

Flow lines

27 3.95 Club-shaped, with longitudinal, parallel Cow lines Flow lines

28 2.23 Club-shaped with a projection. Flow lines parallel, close-spaced and

along the longer dimension of the specimen

Flow lines

29 3.5 Club-shaped with a bend at one end of the specimen. Flow lines are

close-spaced, parallel to each other and along the longer dimension of

the specimen

Flow lines

30 0.55 Irregular/club-shaped with Cow lines parallel to each other along the

longer dimension of the specimen

Flow lines

31 1.4 Elongated, folded in a loop and jointed at the ends. Flow lines parallel

to the longer dimension of the specimen

Flow lines

32 0.31 Slightly Cattened sphere. Circular, close-spaced Cow lines on the

anterior and incipient Cange development

Flow lines

33 0.63 Flattened, straight piece of microtektite material with close-spaced,

parallel Cow lines

Flow lines

34 7.5 Elongated, thin, club-shaped and bent slightly (plastic deformation),

with parallel Cow lines more dominant on one side of specimen

Flow lines

35 0.33 Slightly Cattened sphere. Circular, close-spaced Cow lines on the

anterior and incipient Cange development

Flow lines

36 5.38 Elongated, rod-like with Cow lines along the longer dimension of the

specimen

Flow lines

37 1.7 Club-shaped, Cattened slightly and bent (plastic deformation) with

parallel, close-spaced Cow lines along the longer dimension of the

specimen

Flow lines


argument is further strengthened by the presenceof lechatelierite in some of the specimens.Microprobe analysis of the Pele’s tears from

Masaya, an active basaltic shield volcano locatedin Nicaragua, in general, show compositions similarto Pele’s hair; but from the outer to inner parts ofthe tears a strong chemical gradient is noticedhaving SiO2 is *78 wt.%, with low concentrationsof the other major oxides, and Na2O and K2O arenearly equal and *1 wt.% (Moune et al. 2007).These tears are unlike the CIO microtektites,studied here, that have a more homogeneouscomposition and with enhanced K2O vis-�a-visNa2O (table 4). Similarly, the CIO microtektites donot correspond to the silicate microspherulesrecovered from the plume of Mt. Etna that aresmaller (generally \ 2 lm) and have a variableSiO2 content (47–98 wt.%; Lefevre et al. 1986). Thelikeness of the CIO basin particles to keatite (SiO2

75–88 wt.%), a Si-rich grain in volcanic eruptionclouds (Rietmeijer 1988) is ruled out.To leave no iota of doubt, the CIO microtektites

were compared with extraterrestrial glass cosmicspherules (similar to that of Van Ginneken et al.

2018), collected from Antarctica and deep-seasediments. These spherules have undergoneextreme heating as displayed by distinct composi-tion with much higher Si value (Bgure 7) similar tothose observed earlier by Folco et al. (2010). Thepresent work of the aerodynamically distortedparticles shows amazing geochemical similaritiesthat have major elemental composition fallingwithin the range of Australasian microtektites(Bgure 6). Apart from this, the Transantarcticmountain microtektites also have a range similar tothose particles in the present study indicating thatall the geochemical behaviour of these specimensare undeniably microtektites. The majority ofoxide such as K2O, MgO, Al2O3, TiO2, CaO, andFeO vs. SiO2 overlap (Bgure 6). The volatile ele-ments do not show any speciBc trends as themicrotektites seem to be as depleted in Na2O andK2O similar to that seen in Australian microtek-tites. One of the reasons is that Na volatilization inthe microtektites in this study is incompletebecause probably the specimen may have been inspace for a shorter amount of time before theycooled faster and retained some sodium back. The

Table 2. Major element compositions of unablated microtektites from AAS 62/56.

# Diameter (lm) SiO2 Al2O3 FeO MgO CaO K2O Na2O TiO2 MnO Total

Intermediate type

P23 220 64.62 16.69 6.27 7.46 3.13 0.69 0.41 0.89 0.13 100.36

P13 230 65.96 15.69 5.79 6.17 3.85 0.80 0.48 0.89 0.14 99.85

P25 220 66.03 15.60 5.48 6.95 3.15 0.80 0.46 0.86 0.12 99.50

P24 330 66.70 14.17 4.85 7.38 3.91 0.62 0.33 0.82 0.10 98.90

Normal type

P21 305 67.13 13.82 7.74 4.82 1.97 1.01 0.61 1.14 0.15 99.49

P27 220 63.42 18.48 5.37 5.60 4.85 0.57 0.25 1.01 0.12 99.69

P19 232 67.23 17.29 4.97 3.40 4.09 1.36 0.51 0.94 0.09 99.91

P16 330 69.08 15.30 5.88 3.19 3.41 1.83 0.61 0.82 0.09 100.21

P32 305 69.82 13.36 5.34 5.66 3.47 0.87 0.41 0.89 0.15 99.97

P30 366 69.90 12.96 6.59 4.47 3.03 1.60 0.56 0.75 0.12 99.98

P32A 425 70.50 13.62 4.71 5.64 3.04 0.75 0.38 0.80 0.10 99.76

P18 195 70.80 15.50 3.86 3.32 3.57 1.07 0.42 0.92 0.08 99.56

P17 256 70.91 14.19 4.65 4.45 3.22 0.96 0.37 0.82 0.09 99.71

P22 512 71.11 12.66 6.00 3.28 3.07 2.14 0.59 0.69 0.11 99.69

P28 425 71.42 13.89 5.37 2.46 2.92 1.77 0.52 0.79 0.11 99.26

P12 732 71.71 12.59 5.82 4.20 2.80 1.47 0.44 0.71 0.10 99.89

P33 500 72.21 14.52 3.96 3.41 3.55 1.05 0.37 0.86 0.09 100.04

P15 270 72.50 13.24 5.35 2.45 2.80 1.73 0.38 0.75 0.10 99.31

P27A 560 72.61 14.36 3.51 3.06 3.47 1.09 0.36 0.86 0.07 99.40

P14 295 73.77 12.02 4.65 3.58 2.68 1.42 0.45 0.74 0.10 99.44

P26 295 74.17 13.97 3.31 2.96 3.30 1.00 0.33 0.84 0.08 99.96

The classiBcation of l microtektites is based on MgO content as deBned by Glass et al. (2004). The current study does not haveany high Mg-type microtektites.


chemical similarity of the Australasian microtek-tites with the present study except for the alkalielements could be related to the entry parametersrather than the impact event and source crater.Presently, none of the Transantarctic Mountainsmicrotektites as large as *800 lm has shown anyevidence of ablation and unlike the microtektitesfound in deep-sea sediment cores, although theyare well-preserved from weathering by the coldfrozen environment (Folco et al. 2009). The shapeand size of microtektites further reinforce the factthat they cannot be deposited further away fromthe source location like the TransantarcticMountains and restrict the extension of theAustralasian strewn Beld to the CIO region.

Given the process by which the tektites areproduced, these contain extremely low H2O con-tents (\0.02 wt.%) (Beran and Koebrel 1997) andthis is true for microtektites as well (Glass et al.1997). This has been one of the crucial charac-teristics to distinguish them from volcanic glass.In the present study, Brstly, the measured watercontent of *0.005 to 0.01 wt.% is well within thedeBned range of tektites. Secondly, the viscouslydistorted form of the specimens requires meltviscosities low enough to allow for intense defor-mation that cannot be achieved at low waterconcentrations and temperature realistic for amagmatic system. Also, the FTIR peaks areslightly shifted from the ideal value that is 3,660cm–1 or 2.73 lm towards higher wavelength orlower wavenumber which can be interpreted dueto the addition of alkalis or other crystal BeldeAects aAecting O–H stretching vibration(Gilchrist et al. 1969; Beran and Koebrel 1997).Glass et al. (1997) suggested that some tektitefragments and microtektites recovered from mar-ine sediments can be highly hydrated or palago-nitized, depending on the composition of the glass.Besides, the Raman spectra acquired for aerody-namically distorted particles typically depictedfeatures similar to tektites with two broad bandsat 400–600 and 800–1100 cm–1 (Bgure 10). Gen-erally, Raman spectra of the glasses are charac-terized high wavenumber region (800–1200 cm–1)and mid wavenumber region (400–650 cm–1) andcan be seen in our analysis also. The Broad Ramanbands at 800–1100 cm–1 can be attributed to theeAect of non-bridging oxygen present in the silicatetrahedron, however, observations at 400–600cm–1 is mainly governed by mixed vibration acrossSi–O–Si bridging bonds and may also be aAectedby the alkalis present in the system. In most of theT

able

3.Major

elem

entcompo

sition

sofablatedspecim

ens.

No.

Shape

Size(lm)

SiO

2Al 2O

3FeO

MgO

CaO

K2O

Na2O

TiO

2MnO

Total

Rem

arks

S1P1

Sphere

219

67.11

13.72

5.98

3.57

3.26

2.20

1.02

0.79

0.14

98.03

Norm

almicrotektite

Diffuse

grain

50952

84.56

8.04

2.83

1.64

1.39

2.64

0.72

0.44

0.04

102.30

Highsilica

inclusion

Diffuse

grain

51950

89.20

4.38

1.63

0.67

0.56

2.09

0.54

0.23

0.01

99.31

Highsilica

inclusion

S1P2

Sphere

246

69.60

13.75

5.18

2.86

3.14

2.81

1.48

0.79

0.09

99.70

Norm

almicrotektite

Diffuse

grain

20

77.91

9.94

3.20

1.76

1.94

2.66

1.06

0.58

0.10

99.15

Highsilica

inclusion

S1P3

Sphere

301

71.07

13.09

4.96

2.98

2.69

1.94

0.76

0.76

0.11

98.36

Norm

almicrotektite

Diffuse

grain

74959

82.10

9.33

3.54

1.85

1.74

1.94

0.58

0.48

0.13

101.69

Highsilica

inclusion

Sub-hedral

17

100.69

0.07

0.10

––

0.06

––

–100.92

Lechatelierite

Sub-hedral

15

100.18

0.05

0.24

0.03

0.02

0.06

0.02

––

100.60

Lechatelierite

S1P4

Sphere

3079330

68.00

13.99

5.89

3.08

2.86

2.46

1.06

0.79

0.09

98.83

Norm

almicrotektite

S1P5

Sphere

489

68.52

14.65

5.48

3.06

3.15

2.65

1.34

0.80

0.10

99.76

Norm

almicrotektite

S1P6

Sphere

440

67.63

13.92

6.10

3.65

3.14

2.30

1.05

0.79

0.11

98.70

Norm

almicrotektite

S2P1

Club-shaped

1420

67.88

14.02

6.01

3.45

3.63

2.34

1.13

0.82

0.10

99.39

Norm

almicrotektite

S2P2

Platy,glassyfragment

984

66.65

14.35

6.60

4.07

3.70

1.89

0.84

0.83

0.12

99.06

Norm

almicrotektite

S2P3

Folded

andjointed

826

68.52

13.41

6.24

3.71

3.01

2.34

1.24

0.76

0.10

99.34

Norm

almicrotektite


cases Al3+ substitutes to Si4+ and the substitutioncharge is balanced by alkalis, and thus Al3+ mainlygoverns the structural framework of the tetrahe-dral. Pure silica glass is mainly characterized by abroad and intense band around. In our opinion, it isnot due to the concentration of Aluminum 435cm–1. The intensity is gradually decreased due tothe addition of alkalis in the framework. Aluminummay also aAect the spectra and band 800–1100cm–1 by entering the structural framework of thetetrahedra. The shift in wavenumber of the spectradue to Al3+ may be attributed to the decrease in(Al/Si)–O force constant with an increase in Al/Siratio (Rossano and Mysen 2012).In summary, a volcanic origin for the CIO glass

distorted specimens is negated since a volcanicsource, if any, has to be local and coincident withthe Australasian tektite event and should haveproduced glassy material which is of tektite-likecomposition. Chemical composition and otheranalyses rule out the possibility of these glasseshaving a volcanic origin. Moreover, there are noreports of similar glassy particles produced by thevolcanoes in the vicinity of the CIO basin such asBarren, Java, Sumatra, etc. Additionally, lechate-lierite inclusions are more common in microtektitesand tektites rather than in terrestrial volcanicglasses (O’Keefe 1976). The presence of lechate-lierite particles and the absence of microlites orcrystallites in addition to their compositions alsorule out a volcanic origin and indicate that the CIOspecimens are typical microtektites.

5.2 Ablation phenomenon

Tektites in the southern part of the Australasiantektite strewn Beld have undergone the second

phase of melting because of which the outer skin ofcold, rigid tektite bodies was pulled back, forming aCange. Such specimens have been found exclusivelyin Australia (O’Keefe 1976) except for a loneCanged button specimen in the Indian Ocean(Prasad and Rao 1990), which are the well-knownaerodynamically ablated tektites (Canged aus-tralite buttons). Chapman and Larson (1963) car-ried out extensive arc-jet ablation experimentsunder variable conditions on tektites and otherglasses that had variable sizes and physical prop-erties. These classic experiments have led to ourunderstanding of the ablation phenomenon. Thoseauthors have demonstrated that at low stagnation-point pressures tektite glass ablates without pro-ducing any ring waves. This feature was alsoobserved in the Indian Ocean tektite, which did notpossess ring waves on its anterior side (Glass et al.1996). However, with increasing stagnation-pointpressure (closer to 1 atm) ring waves appear, andthe relative spacing between these ring waveswould decrease, which would further depend on theentry trajectory of the specimen (Chapman andLarson 1963). Some of the CIO specimens bearsuch evidence of severe aerodynamic stress. Weobserve centimeter size, Cattened samples(Bgure 5) of tektite compositions which bear aresemblance to experimentally generated glassdrops (Chapman and Larson 1963) distorted byaerodynamic pressures. The sphere-shaped speci-mens (Bgures 3 and 4) all have multiple, very close-spaced ring waves. These ring waves are only a fewlm apart considering that we are observingmicroscopic specimens of 200–350 lm diameter.Furthermore, under much severe aerodynamicpressures, glass tends to get distorted into Cattenedand much-elongated shapes where mm-sized

Table 4. Range and average of oxides of present samples compared with typical compositions of Australasian microtektites ofCassidy et al. (1969), Glass et al. (2004) and Glass and Koeberl (2006).

Wt.%

Range of

normal

microtektites

Average of

spherical

microtektites

Range distorted

microtektites

Average of

distorted

microtektites

Australasian

microtektite

composition

SiO2 63.42–74.17 69.6 66.65–69.60 68.33 49.6–77.0

Al2O3 12.02–18.48 14.47 13.09–14.65 13.88 7.5–22.1

FeO 3.31–7.74 5.21 4.96–6.60 5.83 3.0–8.1

MgO 2.45–7.46 4.47 2.86–4.07 3.38 1.9–17.1

CaO 1.97–4.09 3.3 2.69–3.70 3.18 1.0–5.8

K2O 0.57–2.14 1.17 1.89–2.81 2.33 0.1–3.7

Na2O 0.25–0.61 0.44 0.76–1.48 1.10 0.2–2.8

TiO2 0.69–1.01 0.85 0.76–0.83 0.79 0.5–1.0

*Spherical microtektites are normal/intermediate type based on MgO content.


droplets tend to get Cattened and stretched(Chapman and Larson 1963).Earlier studies reported that more than 90% of

the specimens are close to spherical with someunusual shape have not been reported in literaturewidely probably because researchers were skepti-cal of its veracity. Some of the elongated speci-mens from the CIO have similarities to thosesamples which have been ablated in the labora-tory under negligible aerodynamic forces (e.g.,Bgure 5c); however, the glasses used in theseexperiments had different viscosity (Chapman andLarson 1963). The heating experiment conductedon the glasses produced various shapes dependingon viscosity, velocity, aerodynamic properties, etc.

Most of the tektites that have low viscosity of l =0.08 Pa s have shown *95% sphericity with a fewpercent to deformed shape (Bgures 3–5). However,with an increase in viscosity from l = 0.08 to 0.29Pa s and 3.5 Pa s, the percent of tear-shapedincreases dramatically from *5 to 75% (Chapmanand Larson 1963). The experimental evidencedoes reCect the tendency of high viscosity of glassto deformation, but then again the observeddeformation of tektite particles to achieve a shapeother than sphericity is also feasible at a lowviscosity.The preserved shape of the tektites underscores

the importance of heating and quenching. Lowviscosity is needed to deform the dacitic melt of the

Figure 6. The minor and major element composition vs. SiO2 of present work compared to literature data on Australasian andTransantarctic microtektites (Folco et al. 2010). All values reported in the present study are in wt.%. The Australasian andTransantarctic microtektites are shown in a circle and rectangle Blled grey colour, respectively. The distorted and ablatedspherical microtektites are shown in diamond yellow colour. The spherical microtektites with Cow lines features on their surfaceare shown in rectangle red colour.


tektites and such particles resemble basaltic Pele’shair fragments formed in volcanic environments(Bgure 11). However, in comparison to Pele’s hairformed from low-viscosity basalts, the tektites wereformed from SiO2-rich melts with generally muchhigher viscosities. Such melts may achieve lowviscosities either by high water concentrations,which can be ruled out due to the lack of water inthe tektites or by very high temperatures, abovetheir liquidus. For the presently studied tektites,the liquidus temperature is of the order of*1100–1200�C – with corresponding viscositiesbetween 104 and 106 Pa s. Such viscosities seemunreasonable to allow the given. Dropping theviscosities of the tektite melts to values charac-teristic for basaltic systems which are known toproduce Pele’s hair, temperatures of at least*1500�C can be expected (Bgure 11). The factthese very delicate structures were preservedemphasizes the high temperatures of formationfollowed by rapid quenching of the melt throughthe glass transition interval. However, it is perti-nent that despite the unusual shape, the chemicalcomposition is scattered in a small range (Bgure 6).Chapman and Larson (1963) also suggested thatstriae distortions provide information about theentry trajectory, and with a more significant rate ofheating, the thickness of the melt layer would besmaller. Given the above parameters, the presence

of close-spaced striae on all the specimens – dom-inantly only one facet of the specimens and verythin melt layers contained on all the specimenswould perhaps indicate steeper entry angles andhigh entry velocities. So far, the smallest sizedtektite which showed evidence for aerodynamicablation measured *2 mm (Baker 1963; Chao1993). Chao (1993) suggested that these smallspecimens were indeed rigid and relatively coolprior to entry and further, steeper entry velocitieswould lead to higher aerodynamic stresses resultingin fringe waves and Cange Cattening. Therefore,small specimens such as the ones in this studycould have been rigid, cold bodies before re-entry.The largest Australite found so far measures83.797.21954.51 mm (weighing 437 g) and isunablated (McNamara and Bevan 2001). Consid-ering several large unablated specimens (weighing[100 g) found in Australia, McNamara and Bevan(2001) suggested that there could be an upper sizelimit for ablation to take place, beyond which thetektites may behave like meteorite masses whichhave fusion crusts but do not develop Canges. Ourpresent study suggests a possible lower limit forablation to 0.2 mm. Given the tektite mass con-tained in some of the elongated specimens, it couldbe discerned that the original, unablated diametersof some of the CIO specimens could have been afew mm in size.Flow lines that diverge and go around lechate-

lierite inclusions indicate that while the objectswere still Cuid and being deformed, the melt Cowedaround the lechatelierite inclusions. Protrusion ofthe lechatelierite particles above the adjacent sur-face is due to differential solution, and not becauseof ablation. Aerodynamic ablation is a commonphenomenon observed in the southern part of thestrewn Beld, mostly in tektite specimens fromAustralia. If the CIO specimens were ablated sim-ilarly, they should have experienced two phases ofmelting: once during the impact upon which theywere ejected into the air, and subsequently duringre-entry. Such double melting would have reducedthe volatile contents relative to the othermicrotektites. The Na–K contents, on the con-trary, are enhanced in these specimens and hencehave not experienced such high temperatures.Elkins-Tanton et al. (2003) suggested that thesplashing of shock melt produces tektites, i.e., theyare Cuid droplets in motion which solidiBed.Because of their plastic deformation and in view oftheir higher volatile contents, it, therefore, appearsthat CIO specimens are shock melt that has

Figure 7. The ternary diagram of Mg–Si–Fe with comparisonto glass cosmic spherules from deepsea sediments and Antarc-tica; and microtektites from the present study along withAustralasian and Transantarctic microtektites (Folco et al.2010). The region in grey area is that of the cosmic spheruleswhile that marked by dotted red lines is for microtektites.


undergone aerodynamic distortion while in Cight.Ablation experiments (Chapman and Larson 1963)revealed that at low stagnation point pressures,ring waves do not appear on aerodynamicallyablated specimens, whereas at higher pressures thepresence of ring waves is seen. The shapes of theparticles that resemble extrusion of melt intoelongated particles, and distortion of these speci-mens suggests plastic deformation rather thanablation, that too at low temperatures.The question as to why ablated specimens were

not found so far despite focused morphologicalstudies on microtektites (Glass 1974; Prasad and

Sudhakar 1998; Prasad and Khedekar 2003) needsto be addressed. The tektites of variable sizes andalso of differential ablation can be attributed tovariable Cight trajectories followed by the tektites.The trajectory of the studied tektites is dependentupon various parameters such as velocity, size,entry and re-entry angle, viscosity of melt, atmo-spheric density, wind direction, the angle at whichthe target rock was hit by the impactor. Aerody-namically ablated specimens can be found only at afew thousand km from the source area and thedistance where re-entry of the ejecta occurredhas been pegged at *3500 km (Ford 1988). This

Figure 8. (a) S1P1: Polished section of a spherical ablated microtektite shown in Bgure 3(b). The polished section reveals silica-rich inclusions with diffused margins (see table 3 for chemical composition). The left end of the specimen also shows Cow linesgenerated during the process of aerodynamic ablation. (b) S1P3: Polished section of another spherical ablated microtektiteshown in Bgure 4(c). A diffuse inclusion of silica-rich mineral is visible. In the diffuse inclusion, dark inclusions are lechatelieriteas seen in the backscattered electron image of the polished section. Flow lines generated during ablation are also seen on the leftside of the photograph. (c) S2P3: Polished section of non-spherical, elongated, bent and jointed ‘microtektite’ – the internalsection is featureless, homogeneous. (d) S2P2: Polished section of a Cat, elongated specimen – showing homogeneity throughoutthe exposed surface. (e) Large ([1 mm diameter) microtektite showing schlieren which is exposed because of etching during theresidence of the microtektite on the seaCoor. (f) ‘Normal’, unablated microtektite with pitting and no Cow lines as seen on theablated specimens in the present study.


phenomenon could be observed only in the south-ern part of the strewn Beld – south of the point ofre-entry of Australasian impact ejecta. Microtek-tites are found solely in the oceans where the con-ditions for their preservation are conducive,however, at this distance from the source area, themicrotektite abundances for a given area of theseaCoor drop dramatically from *1000 microtek-tites/cm2 closer to the source area to a mere *20microtektites/cm2 away from the source (Glassand Koeberl 2006; Prasad et al. 2007). Themicrotektite abundance at the present location is*5 microtektites/cm2 (Prasad et al. 2007), and

only 37 out of 2700 microtektites investigated dis-play some evidence of ablation. Glass and co-workers have carried out pioneering work over theyears on the Australasian microtektites yet,throughout their investigations, the area of theseaCoor sampled by them, given the logisticsinvolved, was having a cross-sectional area of *4cm2 at each location (Glass 1978). Earlier, sam-pling of *225 cm2 of the CIO sediment seaCoor ledto the discovery of minitektites (Prasad and Sud-hakar 1999). In our study, at a 2,500 cm2 cross-sectional area the seaCoor sampled is an order ofmagnitude higher than any so far, therefore to Bndsuch a phenomenon, one needs to sample a largecross-sectional area of the deep seaCoor especiallyin the distal parts of the strewn Beld.The location from where the CIO specimens were

recovered is at the distal end of the strewn Beld andlocated[3,500 km from the putative source (Pra-sad et al. 2007). This location also falls south of there-entry point of the Australasian ejecta because anaerodynamically ablated Canged button was dis-covered at least 400–500 km north of this samplelocation (Prasad and Rao 1990). The Bnding ofaerodynamically distorted specimens outlines sev-eral factors: the fact that these specimens havehigher soda-potash contents indicate that theyhave not experienced the level of heating andhomogenization that the other microtektites in thislocation have undergone. The aerodynamicallydistorted specimens have been ejected into space

Figure 9. Fourier-transform infrared spectrometer (FTIR) curve of aerodynamically distorted tektites where prominent peaksare present at around 3,600 cm–1. The peak sharpness in IyPh˙1-avg is more noticeable than compared to IyPh˙2-avg.

Figure 10. Raman spectra of aerodynamically distorted tek-tites are similar to tektites having two broad bands at 400–600and 800–1100 cm–1.


during the impact similar to the ablated specimensthat have been reported from Australia, however,given their smaller masses did not have sufBcientdynamic ram pressure. After the initial distortion,they lost their momentum due to the high densityof the atmosphere that they encountered whichslowed them down and considering the low tem-peratures in the upper atmosphere the specimenshave cooled rapidly during their descent.It is now seen that this small sampled area of

50950 cm sampled contains: normal microtektites,intermediate microtektites, HMg microtektites,HNa–K microtektites and also aerodynamicallydistorted microtektites. The chemical compositionof the microtektites analyzed displays this range ofsub-classes as described. This indicates that even asmall spot in the distal region of an impact containsejecta that has deBned different trajectories asindicated by their shapes and compositions. In anysample, there is a vast difference in sizes ofmicrotektites ranging anywhere between a fewmicrometres to a few mm (Prasad and Sudhakar1999) all within a few cm2 of the seaCoor that wassampled. Schaller and Melosh (1998) suggestedthat the smaller particles reach terminal speeds athigher altitudes than the large particles. ThiseAectively means that at each location, themicrotektites that are deposited have different

trajectories because of their variable sizes andchemical compositions. This could also be thereason that, while there is a systematic decrease inthe volatile contents of microtektites withincreasing distance from the source crater (Folcoet al. 2010), it is also true that at every locationanalyzed in the strewn Beld, the tektites have awide size range chemical compositions.

5.3 Implications

The above Bndings reveal an insight into the pro-cesses that occur during re-entry, especially ofmicrotektites while ablation of tektites has beenrecorded abundantly. Collisional processesbetween microtektites in Cight have been observedof about*8–10% of the Australasian microtektites(Prasad and Khedekar 2003). Schaller and Melosh(1998) found several Venusian craters and theirejecta distribution. They suggested that theobserved parabolas in the ejecta distribution are aresult of the eAect of winds on the Venusian sur-face, which they also believed has a similarity tothe ejecta distribution primarily similar to those oftektite strewn Belds on the earth. Ablated featuresidentiBed on microtektites are new phenomenathat help to comprehend the processes during

Figure 11. Viscosities calculated for some selected tektite compositions, in comparison to the viscosity of a basaltic melt thatforms Pele’s hair (Moune et al. 2007). Viscosities are calculated from Giordano et al. (2008). Liquidus temperatures werecalculated using the MELTS-software package (Gualda et al. 2012; Ghiorso and Gualda 2015). The MELTS-program was usedfor the aerodynamically distorted tektites along with Pele hair’s composition. The viscosity is significantly higher in the distortedtektites compared to Pele’s hair.


atmospheric interaction and re-entry that are nowknown to aAect microtektites and seem like apossibility especially on planets with substantialatmospheres.Microtektites that are formed by melt disruption

and condensation of rock vapour caused by impactand ejecta can only be distributed widely by eventsthat have enough energy to penetrate the atmo-sphere which plays a substantial role in heatingand transporting target materials in impacts ofsub-km and km projectiles. However, apart fromthe nature of atmosphere, velocity and angle of theimpactor, the type of bedrock constrain the qualityand distribution of ejected material. Lorenz (2000)compared the formation and distribution environ-ment for microtektites on Mars, Venus and Titanwith that of Earth. Mars has a tenuous atmosphereand thus impact events are more frequent and canoccur for smaller and less energetic events also.However, Venus has a high-density atmosphereand thus more energetic impact with high entryvelocity is required as compared to Mars. Due tothe dense atmosphere of Venus the craters that aregenerally\30 km in diameter are meager, unusu-ally shallow, and multiple or complex and irregularin shape because of catastrophic fragmentation andCattening of the projectile by the atmosphere. Asthe Moon has low gravity and the atmosphere isabsent retention of high-speed impact melts isdifBcult and most of the impact melts are found tooccur on crater Coors. Aerodynamic ablation is acommon phenomenon observed in the southernpart of the Australasian strewn Beld, and this maybe attributed due to the oblique collision of theimpactor. The specimens which are suggested hereto be indicative of aerodynamic distortion arefound to co-occur with the Australasian microtek-tites. Therefore, stratigraphically they belong tothe same horizon as the Australasian microtektites.This implies that at the distal end of the ejecta,there may be specimens that have undergone dif-ferent levels of ablation. Differential ablation at thesame location may be due to different trajectories,different entry angles and sizes.According to Gilchrist et al. (1969) sometimes

water is low along edges where atmospheric heatingis severe. However, microtektites in our case maybe homogenous due to their small size and rela-tively uniform heating eAect. Furthermore, watercontent is usually low along the anterior side of theparticle. Watt et al. (2011) found that tektites fromAustraliasian strewn Belds have higher water con-tent and may have been formed in a wet or aqueous

environment as compared to those tektites associ-ated with other strewn Belds. They also suggestedthat celestial bodies such as Moon where theatmosphere is negligible, the rocks that meltedduring impacts will not undergo buAering byvolatile cloud and in spite of the high water contentof bedrock which has impacted the microtektitesare significantly devoid of water. Thus nature ofthe atmosphere must be taken into considerationwhile studying tektites produced during high-velocity impacts. Although, we have focused ourpaper mainly on the geochemical aspect of the CIOmicrotektites, we believe that the interpretationscould be extended to those planets that have anappreciable atmosphere. This will help researcherswho have modelled the tektite trajectory by usinganalog starting material. Till date very few exper-imental works has been done to understand theablated tektites as these generally belong to splashform of tektites and have been inCuenced by rollingand distortions (Elkins-Tanton et al. 2003). How-ever, ablated microtektite forms are of utmostimportance and are rare to be found especially inoceanic sediments. Our Bnding will be an impetusfor further analog experimental studies and providea basis for experimental studies to understand thetrajectory and entry parameters of tektites.

6. Conclusion

Sampling a large area (2,500 cm2) area of the CIOseaCoor helped to recover *2,700 microtektites([250 lm diameter) in the distal area of the Aus-tralasian tektite strewn Beld. The ablated speci-mens (37 specimens out of 2,700 microtektitesrecovered from this site) all have Cow lines orstriations on their surfaces. The microtektitespossess two different forms: spherical or variants ofa sphere and, highly elongated, Blamentous forms.The latter forms display shapes that are bent,plastically deformed and jointed at both ends,straight and platy, etc. These specimens containhigher soda-potash contents when compared withthe associated unablated Australasian microtek-tites in the same sediment core. This leads us toconsider that the aerodynamically distorted speci-mens have not experienced the amount of heatingthat the associated microtektites have undergone.Compositionally, the CIO microtektites are other-wise indistinguishable from the ablated specimens.Therefore, by their geographic location,


stratigraphic position and chemical composition,the CIO specimens are similar to the Australasianmicrotektites. The specimens have experienceddifferent atmospheric entry conditions than thoseundergone by the Canged ablated specimens foundcommonly in Australia. These specimens help tounderstand the atmospheric interaction and abla-tion during re-entry, especially concerning sub-mmmicrotektites. Furthermore, at the studied site(17�57.2230S and 78�01.7110E) which is south of there-entry point deBned for aerodynamic ablation ofAustralasian tektites, we also Bnd aerodynamicallydistorted specimens that have high Na–K contents.These aspects of the cm-level diversity of the ejectaare also important from a planetary perspective aswell. For example, the distribution of ejecta onVenus is observed to have many similarities to thatof tektites on the Earth.

Acknowledgements

We are grateful to all the participants of the cruiseAAS-62 for their help during sample collection,Vijay Khedekar for help during the SEM observa-tions. This work is sponsored by the Ministry ofEarth sciences-PMN and the Indian Space ResearchOrganization PRL-PLANEX project. We acknowl-edge the reviewers for their comments that helped toimprove the presentation of the paper and Dr Cha-lapathi Rao for encouraging us to carry out therevisions. This is NIO’s contribution 6670.

Author statement

NGR, MSP, SDI, MP, CH and DKP have allsupported equally to the main objective behind themanuscript, writing and analysis. MSP and SDIhave supported with sample collection and initialobservations.

References

Baker G 1963 Disc-, Plate-, and bowl-shaped australites;Meteoritics 2 36–49.

Barnes V E 1958 Properties of tektites pertinent to theirorigin; Geochim. Cosmochim. Acta 14 267–278.

Barnes V E 1963 Detrital mineral grains in tektites; Science142 1651–1652.

Beran A and Koebrel C 1997 Water in tektites and impactglasses by Fourier transformed infrared spectrometry;Meteorit. Planet. Sci. 32 211–216.

Cassidy W A 1964 Experimental testing of hypotheses for theorigin of tektites; Geochim. Cosmochim. Acta 28999–1000.

Cassidy W A, Glass B P and Heezen B C 1969 Physical andchemical properties of Australasian microtektites; J. Geo-phys. Res. 74 1008–1025.

Chao E C T 1993 Comparison of the Cretaceous–Tertiaryimpact events and the 0.77 Ma Australasian tektite event:Relevance to mass extinction; USGS Bull. 2050, U.S. Govt.Printing ODce, Denver, Colorado, USA, 22p.

Chapman D R 1964 On the unity and origin of theAustralasian tektites; Geochim. Cosmochim. Acta 28841–880.

Chapman D R and Larson H K 1963 On the lunar origin oftektites; J. Geophys. Res. 68 4305–4358.

Davis A S and Clague D A 2006 Volcaniclastic deposits fromthe North Arch volcanic Beld, Hawaii: Explosive fragmen-tation of alkalic lava at abyssal depths; Bull. Volcanol. 68294–307.

Delano J W 1992 Australite Canges as Cight data recorders(abstract); Lunar Planet. Sci. Conf. XXIII., pp. 301–302.

DuDeld W A, Gibson E K and Heiken G H 1977 Somecharacteristics of Pele’s hair; J. Res. U.S. Geol. Surv.5 93–101.

Elkins-Tanton L T, Parmentier E M and Hess P C 2003Magma ocean fractional crystallization and cumulate over-turn in terrestrial planets: Implications for Mars; Meteorit.Planet. Sci. 38(12) 1753–1771.

Folco L, D’Orazio M, Tiepolo M, Tonarini S, Ottolini L,Perchiazzi N, Rochette P and Glass B P 2009 Transantarc-tic Mountain microtektites: Geochemical aDnity withAustralasian microtektite; Geochim. Cosmochim. Acta 733694–3722.

Folco L, Glass B P, D’Orazio M and Rochette P 2010 Acommon volatilization trend in Transantarctic Mountainand Australasian microtektites: Implications for theirformation model and parent crater location; Earth Planet.Sci. Lett. 293(1–2) 135–139.

Folco L, d’Orazio M, Gemelli M and Rochette P 2016Stretching out the Australasian microtektite strewn Beldin Victoria Land Transantarctic Mountains; Polar Sci.10(2) 147–159.

Ford R J 1988 An empirical model for the Australasian tektiteBeld; Australian J. Earth Sci. 35 483–496.

Ghiorso M S and Gualda G A R 2015 An H2O–CO2 mixedCuid saturation model compatible with rhyolite-MELTS;Contrib. Mineral. Petrol. 169 53.

Gilchrist J, Thorpe A N and Senftle F E 1969 Infrared analysisof water in tektites and other glasses; J. Geophys. Res. 741475–1483.

Giordano D, Russell J K and Dingwell D B 2008 Viscosity ofmagmatic liquids: A model; Earth Planet. Sci. Lett.271(1–4) 123–134.

Glass B P 1972 Bottle-green microtektites; J. Geophys. Res.77 7057–7064.

Glass B P 1974 Microtektite surface sculpturing; Geol. Soc.Am. Bull. 85 1305–1314.

Glass BP 1978Australasianmicrotektites and the stratigraphicage of australites; Geol. Soc. Am. Bull. 89 1455–1458.

Glass B P, Swincki M B and Zwart P A 1979 Australasian,Ivory Coast and North American tektite strewn Belds: Size,mass and correlation with geo-magnetic reversals and other


Earth events; Proc. Lunar Planet. Sci. Conf. X,pp. 2535–2545.

Glass B P 1990 Tektites and microtektites: Key facts andinferences; Tectonophys. 171 393–404.

Glass B P and Pizzuto J E 1994 Geographic variation inAustralasian microtektite concentrations: Implications con-cerning the location and size of the source crater; J.Geophys. Res. 99 19,075–19,081.

Glass B P, Chapman D R and Prasad M S 1996 Ablatedtektite from the central Indian Ocean; Meteorit. Planet.Sci. 31 365–369.

Glass B P, Muenow D W, Bohor B F and Meeker G P 1997Fragmentation and hydration of tektites and microtektites;Meteorit. Planet. Sci. 32(3) 333–341.

Glass B P 2003 Australasian microtektites in the South ChinaSea: Implications regarding the location and size of the sourcecrater; Proc. Lunar Planet. Sci. XXXI, pp. 1092–1093.

Glass B P, Huber H and Koeberl C 2004 Geochemistryof Cenozoic microtektites and clinopyroxene-bearingspherules; Geochim. Cosmochim. Acta 68 3971–4006.

Glass B P and Koeberl C 2006 Australasian microtektites andassociated impact ejecta in the South China Sea and theMiddle Pleistocene super eruption of Toba; Meteorit.Planet. Sci. 41 305–326.

Gualda G A R, Ghiorso M S, Lemons R V and Carley T L 2012Rhyolite-MELTS: A modiBed calibration of MELTS opti-mized for silica-rich, Cuid-bearing magmatic systems; J.Petrol. 53 875–890.

Heiken G 1972 Morphology and petrography of volcanic ashes;Geol. Soc. Am. Bull. 83 1961–1988.

Jarosewich E, Nelen J A and Norberg J A 1980 Referencesamples for electron microprobe analysis; Geostandard.Newslett. 4 43–47.

Kinnunen K A 1990 Lechatelierite inclusions in indochinitesand the origin of tektites; Meteoritics 25 181–184.

Lange R L and Carmichael I S E 1990 Thermodynamicproperties of silicate liquids with emphasis on density,thermal expansion and compressibility; Modern Methods ofIgneous Petrology, Berlin, pp. 25–64.

Lefevre R, Gaudichet A and Billon-Galland M A 1986 Silicatemicrospherules intercepted in the plume of Etna volcano;Nature 322 817–820.

Lorenz R D 2000 Production, Acceleration and distributionof microtektites on Earth, Mars, Venus and Titan;Catastrophic Events Conference, 3021p.

Long D A 1977 Raman Spectroscopy; McGraw-Hill, NewYork, 276p.

Ma P, Aggrey K, Tonzola C, Schnabel C, de Nicola P, HerzogG F, Wasson J T, Glass B P, Brown L, Tera F, Middleton Rand Klein J 2004 Beryllium-10 in Australasian tektites:

Location of the source crater; Geochim. Cosmochim. Acta68 3883–3896.

Mc Call G J H 2001 Tektites in the geological record: Showersof glass from the sky; Geol. Soc, London, 256p.

McNamara K and Bevan A 2001 Tektites; Booklet publishedby Western Australian Museum, Perth 1 38p.

Moune S, Faure F, Gauthier P J and Sims K W W 2007 Pele’shairs and tears: Natural probe of volcanic plume; J.Volcanol. Geotherm. Res. 164 244–253.

O’Keefe J A 1976 Tektites and their Origin; Elsevier Publish-ing Co., Amsterdam, 254p.

Prasad M S and Rao P S 1990 Tektites far and wide; Nature347(6291) 340.

Prasad M S and Sudhakar M 1998 Microimpact phenomena onAustralasian microtektites: Implications for ejecta plumecharacteristics and lunar surface processes; Meteorit.Planet. Sci. 33 1271–1279.

Prasad M S and Sudhakar M 1999 Australasian minitektitesdiscovered in the Indian Ocean; Meteorit. Planet. Sci. 34179–184.

Prasad M S and Khedekar V D 2003 Impact microcratermorphology on Australasian microtektites; Meteorit.Planet. Sci. 38 1351–1371.

Prasad M S, Mahale V P and Kodagali V N 2007 New sites ofAustralasian microtektites in the Central Indian Ocean:Implications for the location and size of source crater; J.Geophys. Res. 112 E06007.

Rietmeijer F J M 1988 Enhanced residence of submicronSi-rich volcanic particles in the lower stratosphere; J.Volcanol. Geotherm. Res. 34 173–184.

Schaller C J and Melosh H J 1998 Venusian ejectaparabolas: Comparing theory with observations; Icarus131 123–137.

Simonson B M and Glass B P 2004 Spherule layers – Recordsof ancient impacts; Ann. Rev. Earth Planet. Sci. 32329–361.

Trnka M and Houzar S 2002 Moldavites: A review; Bull. CzechGeol. Surv. 77 283–302.

van Ginneken M, Genge M J and Harvey R P 2018 A new typeof highly-vaporized microtektite from the TransantarcticMountains; Geochim. Cosmochim. Acta 228 81–94.

Von Koenigswald G H R 1960 Tektite studies. II. Thedistribution of the Indo-Australian tektites; Proc. K. Ned.Akad. Wetens, Amsterdam, pp. 142–153.

Watt N, Bouchet RA and Lee C-TA 2011 Exploration of tektiteformation processes through water and metal content mea-surements; Meteorit. Planet. Sci. 46 1025–1032.

Yamashita S, Kitamura T and Kusakabe M 1997 Infraredspectroscopy of hydrous glasses of arc magma compositions;Geochem. J. 31 169–174.

Corresponding editor: N V CHALAPATHI RAO


Documents

Geochemistry of aerodynamically distorted Australasian