Limits from the Hubble Space Telescope on a Point Source in SN 1987A

We observed supernova 1987A (SN 1987A) with the Space Telescope Imaging Spectrograph (STIS) on the Hubble Space Telescope (HST) in 1999 September and again with the Advanced Camera for Surveys (ACS) on the HST in 2003 November. Our spectral observations cover ultraviolet (UV) and optical wavelengths from 1140 to 10266 Å, and our imaging observations cover UV and optical wavelengths from 2900 to 9650 Å. No point source is observed in the remnant. We obtain a limiting flux of Fopt ≤ 1.6 × 10-14 ergs s-1 cm-2 in the wavelength range 2900-9650 Å for any continuum emitter at the center of the supernova remnant (SNR). This corresponds to an intrinsic luminosity of Lopt ≤ 5 × 1033 ergs s-1. It is likely that the SNR contains opaque dust that absorbs UV and optical emission, resulting in an attenuation of ~35% due to dust absorption in the SNR. Correcting for this level of dust absorption would increase our upper limit on the luminosity of a continuum source by a factor of 1.54. Taking into account dust absorption in the remnant, we find a limit of Lopt ≤ 8 × 1033 ergs s-1. We compare this upper bound with empirical evidence from point sources in other supernova remnants and with theoretical models for possible compact sources. We show that any survivor of a possible binary system must be no more luminous than an F6 main-sequence star. Bright young pulsars such as Kes 75 or the Crab pulsar are excluded by optical and X-ray limits on SN 1987A. Other nonplerionic X-ray point sources have luminosities similar to the limits on a point source in SN 1987A; RCW 103 and Cas A are slightly brighter than the limits on SN 1987A, while Pup A is slightly fainter. Of the young pulsars known to be associated with SNRs, those with ages ≤5000 yr are all too bright in X-rays to be compatible with the limits on SN 1987A. Examining theoretical models for accretion onto a compact object, we find that spherical accretion onto a neutron star is firmly ruled out and that spherical accretion onto a black hole is possible only if there is a larger amount of dust absorption in the remnant than predicted. In the case of thin-disk accretion, our flux limit requires a small disk, no larger than 1010 cm, with an accretion rate no more than 0.3 times the Eddington accretion rate. Possible ways to hide a surviving compact object include the removal of all surrounding material at early times by a photon-driven wind, a small accretion disk, or very high levels of dust absorption in the remnant. It will not be easy to improve substantially on our optical-UV limit for a point source in SN 1987A, although we can hope that a better understanding of the thermal infrared emission will provide a more complete picture of the possible energy sources at the center of SN 1987A.