^aDipartimento di Scienze della Terra e dell’Ambiente, Università degli Studi di Pavia, Italy
^bIGG-CNR Unità Operativa di Padova, via Gradenigo, 6, Padova, Italy
^cDipartimento di Scienze della Terra, Università di Torino, Italy
^dCrisDi, Interdepartmental Centre for the Research and Development of Crystallography, Torino, Italy e Bayerisches Geoinstitut, Bayreuth, Germany
^fInt’l Research School of Planetary Sciences, Università G. d’Annunzio, Chieti, Italy
¹Present address: Dipartimento di Geoscienze, Università degli Studi di Padova, Italy.

High resolution single-crystal X-ray diffraction (HR-SCXRD) and Mössbauer spectroscopy of the intracrystalline cation distribution have been performed on augitic core-crystals from a Miller Range nakhlite (sample MIL 03346,13) with approximate composition of En₃₆Fs₂₄Wo₄₀. The Mössbauer data on the single-crystal yielded a very low Fe³⁺ content [Fe³⁺/Fe_total – 0.033(23) a.p.f.u.] that, together with the Electron microprobe analysis (EMPA) and the X-ray structural data allowed us to obtain the accurate cation site distribution and the Fe²⁺–Mg degree of order. This leads to a closure temperature (T_c) of 500 with a standard deviation of ±100 °C that would correspond to a slow cooling rate, which is in disagreement with petrologic evidence that indicates that this sample originates from a fast cooled (~3–6 °C/h) lava flow.
In order to clarify this discrepancy we undertook (i) a SC-XRD study of an augite (~En₄₉Fs₉Wo₄₂) from a pyroxenite (TS7) of Theo’s flow, a 120-m-thick lava flow regarded as a terrestrial analogue of MIL 03346; (ii) an annealing experiment at 600 °C on a crystal from exactly the same fragment of MIL 03346. SC-XRD data from TS7 augite yields a T_c = 600(20) °C, consistent with the cooling rate expected at 85 m below the surface. This T_c is higher, although similar within error, to the T_c = 500(100) °C obtained for MIL 03346; thus suggesting relatively slower cooling for MIL 03346 with respect to TS7. The annealing experiment on the MIL 03346 crystal clearly showed that the degree of order remained unchanged, further confirming that the actual T_c is close to 600 °C.
This result appears inconsistent with the shallow depth of origin (~<2 m) assumed for MIL 03346, further supporting the discrepancy between MIL 03346 textural and petrologic evidence of fast cooling and the above mentioned T_c results obtained for augite. Therefore, a tentative scenario is that, soon after eruption and initial quench and while still at relatively high-T (~600 °C), MIL 03346 was blanketed with subsequent lava flows that slowed down the cooling rate and allowed the augite Fe²⁺–Mg exchange reaction to proceed.