ASCA AND GRO OBSERVATIONS OF GX-301-2

We observed the X-ray binary pulsar GX 301-2 with ASCA (Astro-D) on 1994 February 13-14 at orbital phase similar to 0.3, when the X-ray emission is in a nonflaring state. We also obtained BATSE/GRO observations of the total high-energy flux concurrent with the ASCA observation, as well as before and after. The BATSE data cover 7.5 orbital periods from 1993 September 20 to 1994 July 28. We folded these data over the orbital period and found that the preperiastron (PP) flare is asymmetric-it rises more slowly than it falls. We also detected a periodic near-apastron (NA) flare. Both the asymmetry of the PP flare and the existence of the NA flare can be explained by an equatorially enhanced stellar wind or a circumstellar disk around the optical companion WRA 977 that GX 301-2 crosses twice per orbit. We measured the GX 301-2 pulse period to be 675.7 +/- 0.1 s, the lowest value observed to date. This demonstrates that the spin-up that started in 1984 continues and is the longest lasting secular spin-up of the known pulsars. The secular spin-up indicates that an accretion disk has formed. The appearance of the NA flare may be related to the secular spin-up if the circumstellar disk also provides a steady source of angular momentum to the neutron star. The ASCA Gas Imaging Spectrometer (GIS) and Solid-state Imaging Spectrometer (SIS) images indicate the presence of two previously unknown soft X-ray sources 3' and 8' from GX 301-2. The brighter source at 8' we identify as a dMe star. Both sources have contributed to prior observations of the low-energy GX 301-2 spectrum. We have combined the nearly simultaneous spectral data from ASCA and BATSE to determine the pulse-phase-averaged 4-60 keV spectrum of GX 301-2. It fits a power-law model with low energy absorption, a narrow iron emission line, an iron absorption edge, and a high-energy cutoff. We measured the SIS0/ASCA iron line energy to be 6.41 +/- 0.02 keV, the energy for neutral iron or with an ionization state no higher than Fe XVII. The iron line equivalent width is 230 +/- 30 eV, with 43 eV < intrinsic width <74 eV (90% confidence). The iron edge is at 7.19 +/- 0.03 keV. The line and edge energies are consistent with the same iron ionization state if the state is no higher than Fe V.