• impact melting
• re-entry heating
• Transantarctic Mountains microtektites
• uptake
• volatilization

I conducted an in-depth geochemical study of Australasian microtektites (microscopic impact glass particles) from the Transantarctic Mountains (AUS/TAM), Victoria Land, Antarctica, through petrographic and chemical analyses. The Australasian tektite strewn field is the largest (~15% of the Earth's surface) and the youngest (~0.8 million years old) of the five strewn fields so far know on Earth. It is also the most elusive one: despite its enormous size and young age, its parent impact crater has not been identified yet, although several petrographic and geochemical trends suggest an impact location in Indochina or sourrounding seas. Because of the large amount of tektite material available for reasearch from its entire geographic extension, the Australasian tektite/microtektites strewn field continues to provide important clues on several key aspects of the tektite/microtektite formation mechanism which are stil enigmatic. In this work, I focus on geochemical trends in homogeneous populations of AUS/TAM to contribute to the understanding of chemical fractionation during impact melting and hypervelocity ejection.
The AUS/TAM studied in this work are from Miller Butte (-72.70, 160.25), Victoria Land, Antartica. They belong to the Italian Programma Nazionale delle Ricerche in Antartide (PNRA) collection and are courtesy of the Museo Nazionale dell’Antartide (MNA). Twenty-eight microtektites were selected based on petrographic and morphological features, as well as minimal degree of weathering, based on optical and scanning electron microscopy (SEM). They were divided into six size classes, ranging from 110 µm to 775 µm, in order to study the geochemical behavior of major and trace elements in relation to size. Geochemical data were determined by electron probe microanalysis (EPMA) and laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS).
The twenty eight AUS/TAM have chemical composition consistent with that reported in the literature. Twenty-five belong to normal compostional-type, 2 intermediate-type, and 1 of high-Mg-type. Geochemical data produces important results for the microtektite formation model:
• Significant depletion with a compositinal gap of Na2O (0.26 ± 0.06 (1σ) wt%) and K2O (0.69 ± 0.39 (1σ) wt%)) in the AUS/TAM compared to Australasian microtektites from lower latitude deep sea sediment cores (AUS/DSS) (1 ± 0.53 (1σ) wt% for Na2O and (1.99 ± 1.12 (1σ) wt% for K2O) and AUS tektites from Indochina and Australia (1.20 ± 0.19 (1σ) wt% for Na2O and (2.43 ± 0.24 (1σ) wt% for K2O). A similar depletion is observed in the trace elements Rb (24.40 ± 17.06 (1σ) wt% for AUS/TAM; 100.86 ± 52.97 (1σ) wt% for AUS/DSS; 116.83 ± 13.67 (1σ) wt% for AUS tektites) and Cs (1.44 ± 1.11 (1σ) wt% for AUS/TAM; 5.73 ± 3.70 (1σ) wt% for AUS/DSS; 6.86 ± 1 (1σ) wt% for AUS tektites).
• Inverse relationship between Na2O content (but not K2O) and diameter in AUS/TAM.
• There is an increase in the K2O and Na2O contents from the core to the rim in AUS/TAM, defining U-shaped rim-core-rim profiles. These profiles are particuarly obvious in particles with diameter <300 µm, but become progressively shallower towards concentrations close to the detection limit of EPMA.
From the above results, I infer the following three petrologic constraints:
1. The Na2O and K2O depletion in AUS/TAM is explained by a more pronounced volatilization process in the AUS/TAM. Na and K loss in the microtektite precursor via weathering (leaching) of the target can be ruled out do to the lack of correlation with the similarly mobile element Mg.
2. The inverse relation between Na2O and microtektite size in AUS/TAM is consistent with greater heating and slower cooling rates in larger microtektites. The fact that the similarly volatile element K does not show a clear relation with microtektite size is attributed to the different diffusion rate of the two elements, slower for K.
3. The enrichments in K and Na at the periphery of the AUS/TAM can be explained through volatile uptake the microtektites while still molten. K and Na uptake due to weathering within the Miller Butte sediment traps is ruled out base on the lack of petrographic evidence of glass substitution down to the SEM scale.
Based on the three constraints described above, I envision the following model for the formation of the Australasian microtektites. AUS\TAM, the most distant from the putative impact location in Indochina, are those sourced from the topmost layer of the target, that experienced the highest impact energy, thus the highest ejection velocities, the highest temperatures and thus the stronger volatilazation during impact melting. This account for constrain #1. Physical models in the literature predicts that the size of microtektites decreases with incresing ejection velocity. This implies that the smaller the microtektites, the higher the peak temperatures and the faster the cooling rates. This is in contrast with constrain #2. I thus propose that stonger heating and slower cooling rates in the larger AUS/TAM and vice versa is the record of a second heating event, namely the hypervelocity atmospheric re-entry heating. Such a process accounts not just for constrain #2, but also for constrain #1, determining the volatile compositional gap between AUS/DSS and AUS/TAM – the former ejected close to the hypotetical impact location and the latter, far away from that. The AUS/TAM record a volatile uptake as for constrain #3. This likely occured while AUS\TAM travelled in the impact plume before re-entry. The plume was hot and contained abundant gas produced by mainly by the vaporization/volatilization of the target (of Meso-Proterozoic age according Nd istope model age in the literature). Re-entry heating and cooling is expected to flatten the U-shaped compositional profiles through volatilization more efficiently in larger AUS/TAM than in the smaller ones, consistently with constrain #3.
The geochemical results and petrologic constraints defined through the present study will be instrumental for the numerical modelling of one of the largest Cenozoic impact on Earth. They will be instrumental also for a better understanding of the solild-liquid-gas geochemical and isotopic fractionation process occuring during hypervelocity impacts, including the unexpected occurrence of excess Ar in the AUS/TAM microtektites recently reported in literature.