Stellar Astrophysics from TESS Lighcurves of Eclipsing OBA stars?

OBA Stars physics from eclipsing light-curves?

Background and Goal

Massive stars, according to conventional wisdom, frequently are in close binaries, of comparable mass (size). First strand: massive star physics is not well constrained, and massive binary structure/evolution physics rests on small samples (10's).  Second strand: detailed lightcurve modelling of eclipsing binary systems (given TESS-like quality) can yield very detailed information 

about Mass (ratios), radius (ratios), Temperature (ratios), etc..

Starting Point and Sample

Eleonora Zari devised a large (500k) sample of plausible OBA stars candidates (G < 16), for spectroscopic follow-up with SDSS-V. Multi-epoch BOSS spectra. The stars were selected to have (intrinsically) blue colors, albeit possibly subject to dramatic reddening, and a likely M_K < 0.

But their variability can be assessed with the usual trick of backing the rms lightcurve variability pout of the GDR2 "errors" (in G, BP and RP).

OBA star candidates in the variability vs color-variability plane. Among blue, luminous stars there are essentially two categories of variables (rms G-band > 0.05): pulsators (RRL & Cepheids), with BP/RP variability ~1.6, and (presumed) eclipsing binaries (with BP/RP~1). The latter are the objects of desire here, and selected as in the Figure below.

The rms of the light curve for eclipsing systems is a funny thing: it dramatically favours systems where the eclipses are both deep and occupy a substantive fraction of the orbital period: very close, equal mass/radius binaries are the posterchildren..


Their sky distribution suggests that they are young (massive?) stars:

mostly in the Milky Way, but also in the LMC; the sample if candidate close-eclipsing-binaries contains 1350 targets.

Lightcurve analysis

Their apparent magnitude distribution suggest that > 500 of them should be in the TESS lightcurve 'comfort range'.

But their sky distribution implies crowding that might affect the light curve quality.

Does it makes sense to check for how many of them TESS light curves are available? And then fit them?