The bulge is the oldest component of the Milky Way .
Since numerous simulations of Milky Way formation have predicted that the oldest stars at a given metallicity are found on tightly bound orbits , the Galaxy ’ s oldest stars are likely metal-poor stars in the inner bulge with small apocenters ( i.e. , R _ { \mathrm { apo } } \lesssim 4 kpc ) .
In the past , stars with these properties have been impossible to find due to extreme reddening and extinction along the line of sight to the inner bulge .
We have used the mid-infrared metal-poor star selection of \citet schlaufman2014 on Spitzer/GLIMPSE data to overcome these problems and target candidate inner bulge metal-poor giants for moderate-resolution spectroscopy with AAT/AAOmega .
We used those data to select three confirmed metal-poor giants ( [ \mathrm { Fe / H } ] = -3.15 , -2.56 , -2.03 ) for follow-up high-resolution Magellan/MIKE spectroscopy .
A comprehensive orbit analysis using Gaia DR2 astrometry and our measured radial velocities confirms that these stars are tightly bound inner bulge stars .
We determine the elemental abundances of each star and find high titanium and iron-peak abundances relative to iron in our most metal-poor star .
We propose that the distinct abundance signature we detect is a product of nucleosynthesis in the Chandrasekhar-mass thermonuclear supernova of a CO white dwarf accreting from a helium star with a delay time of about 10 Myr .
Even though chemical evolution is expected to occur quickly in the bulge , the intense star formation in the core of the nascent Milky Way was apparently able to produce at least one Chandrasekhar-mass thermonuclear supernova progenitor before chemical evolution advanced beyond [ \mathrm { Fe / H } ] \sim - 3 .