First Attempt at Spectroscopic Detection of Gravity Modes in the Long‐Period Pulsating Subdwarf B Star PG 1627+017

In the first spectroscopic campaign for a PG 1716 variable (or long-periodpulsating subdwarf B star), we succeeded in detecting velocity variations dueto g-mode pulsations at a level of 1.0-1.5 km/s.The observations were obtainedduring 40 nights on 2-m class telescopes in Arizona, South Africa,andAustralia. The target,PG1627+017, is one of the brightest and largest amplitudestars in its class.It is also the visible component of a post-common envelopebinary.Our final radial velocity data set includes 84 hours of time-seriesspectroscopy over a time baseline of 53 days. Our derived radial velocityamplitude spectrum, after subtracting the orbital motion, shows three potentialpulsational modes 3-4 sigma above the mean noise level, at 7201.0s,7014.6s and7037.3s.Only one of the features is statistically likely to be real,but allthree are tantalizingly close to, or a one day alias of, the three strongestperiodicities found in the concurrent photometric campaign. We furtherattempted to detect pulsational variations in the Balmer line amplitudes. Thesingle detected periodicity of 7209 s, although weak, is consistent withtheoretical expectations as a function of wavelength.Furthermore, it allows usto rule out a degree index of l= 3 or l= 5 for that mode. Given the extremeweakness of g-mode pulsations in these stars,we conclude that anything beyondsimply detecting their presence will require larger telescopes,higherefficiency spectral monitoring over longer time baselines,improved longitudecoverage, and increased radial velocity precision.Comment: 39 pages, 9 figures, 4 tables, ApJ accepted. See postscript for full abtrac