Spectroscopic follow-up of compact object binary candidates from Gaia DR3: White dwarfs, neutron stars, black holes, and the parallax zeropoint

Astrometry and radial velocities (RVs) from Gaia DR3 yielded orbits for hundreds of thousands of binary systems, including several samples proposed to contain black holes (BHs), neutron stars (NSs), and white dwarfs (WDs). We present results of a systematic spectroscopic follow-up program targeting these objects. Beginning with a sample of 227 sources, we used a combination of archival data and many-epoch spectroscopic follow-up to characterize more than 200. We obtained 1292 high-quality RVs over a period of four years using the TRES and FEROS spectrographs, achieving a typical precision of 50 m/s and at least 10 RVs for 60 sources. We use these data to test the Gaia orbital solutions and tighten constraints on orbital parameters and component masses. Joint fitting of astrometry and RVs allows us to directly constrain flux ratios, verifying that undetected companions are genuinely dark. We find that ~60% of the astrometric candidates indeed host compact objects, including the two known Gaia BHs, 27 NS candidates, and a dozen massive WDs. We show that tight WD+WD binaries may masquerade as NSs within this sample. The spectroscopic candidates have lower purity: ~50% have spurious solutions, and a majority of the rest are post-mass-transfer binaries or hierarchical triples. Joint astrometry+RV fits of binaries with dark companions yield direct, parallax-independent distance measurements. Using 40 such systems, we measure the Gaia DR3 parallax zeropoint for astrometric orbital solutions. We find \(Z=-0.0362\pm0.0053\) mas, consistent with the single-star zeropoint for sources of similar color and magnitude. These results will guide the selection of cleaner candidate samples from Gaia DR4, where a longer observing baseline will enable discovery of many more compact object binaries. RV follow-up will remain important for confirming individual systems, particularly those with extreme parameters.