Supersoft Sources Optical Novae

Luminous SSSs were recognized as an important new class of intrinsically bright X-ray sources in the LMC by [75]. Their spectra and luminosities are in the range of (105-106) K and 1036-1038 erg s-1, respectively. The first two sources were detected with the Einstein Observatory in the LMC and optical identifications established the binary nature of the sources. ROSAT detected many more SSSs in the MCs (see red sources in the images of Fig. 20.2), some in the Galaxy and several in nearby galaxies, especially also in M 31 and M 33. Their observed characteristics are consistent with those of white dwarfs, which are steadily or cyclically burning hydrogen-rich matter accreted onto the surface at a rate of order 10~7 M0 yrs-1 for timescales of (106-107)yrs. Steady burning can also occur in a post-nova stage for shorter time scales and SSS have been observed in a few classical novae and symbiotic novae in the galaxy, LMC and one in ROSAT observations of M 31. With the improved sensitivity of XMM-Newton and Chandra many more SSS were detected in nearby galaxies (e.g. 18 in M 31 and 5 in M 33 in the XMM-Newton surveys, with Chandra 8 in M 81 (distance 3.6 Mpc), 10 in M 101 (7.2 Mpc)). [61] discovered with Chandra the first three SSS in an elliptical galaxy, NGC 4697 (16Mpc). Several of the SSS are transients. In addition the first supersoft pulsator (865 s) has been discovered with XMM-Newton as a transient source in M 31 [48,73]. For reviews see [40,41].

As mentioned above, it was known that optical novae could contribute to the class of SSS. [51] therefore correlated optical novae detected in M 31 and M 33 with XMM-Newton, Chandra, and ROSAT data from catalogs and archives. It turned out that the majority of SSS in M 31 are novae in their SSS state (21) and even M 33 where the optical nova catalogues are much less complete than for M 31 revealed two X-ray source/optical nova correlations (see Fig. 20.6). This work more than tripled the number of known optical novae with SSS phase. For many of the novae, X-ray light curves could be determined. From the delay of the onset of the SSS phase after the outburst of two of the novae, one could estimate the ejected hydrogen masses to 10~5 and 10~6 M0. As many novae at different time after outburst can be observed in each observation of the M 31 bulge, monitoring of this area will lead to a better understanding of the X-ray emission of novae and nova outburst properties.


O 2 arcmin = 450 pc N2000-03




200 400 600 600 1000

200 400 600 600 1000

Fig. 20.6 Chandra HRCI image of the center area of M 31 on Oct 31, 2001. Circles with 5" radius indicate nova positions. The cross indicates the M 31 center, the aim point of the observation (left). Soft X-ray light curves of novae in M 31 and M 33 that were detected within 1 000 d after the optical outburst. The light curve of the Galactic nova V1974 Cyg that was monitored by ROSAT after its outburst in 1992, is shown for comparison (right) [51]

20.3.3 Supernova Remnants and Supernovae

Observations in our own Galaxy established SNRs as bright sources of both thermal and nonthermal X-rays (see Chap. 17). After a few hundred to thousands of years they mark the location of a supernova explosion heating the interstellar medium. Samples of SNRs have been obtained for different galaxies and studied to probe their evolution and gain information on the ISM of the host galaxy. In X-rays, sources have been identified with SNRs not only in Local Group galaxies but also in galaxies as distant as ~ 10 Mpc. However, in many galaxies the overlap of X-ray detected SNRs with optically and radio selected SNRs is small. This may be caused by distance dependant effects in the sensitivity to detect SNRs in different wavelength regimes [49].

With Einstein and ROSAT, SNRs have been identified mostly in Local Group galaxies. XMM-Newton and Chandra Observatory now resolve details in SNRs of the MCs. In both, M 33 and M 31, XMM-Newton observations identified 21 SNRs and classified an additional 23 using X-ray hardness ratio criteria and correlations with optical or radio SNR candidates [52, 55]. The SNRs cover a luminosity range from 4 x 1034 to 5 x 1036 erg s-1 in the 0.2-4.5 keV band. With the superior point spread function (PSF) of the Chandra Observatory mirror/detector system five SNR could be spatially resolved in M 31 [79, and references therein] and four (plus three marginally) in M 33 [26], respectively. In addition, two X-ray selected M 33 SNR candidates from [55] were identified as SNRs in optical images [26] proving the X-ray selection strategy. With the number of SNRs known it is now possible to compare X-ray luminosity functions of SNRs in different Local Group galaxies. As a result LMC SNRs on average show higher luminosities than SNRs in M 31 and M 33. This effect may be caused by differences in interstellar abundances between the galaxies.

But one does not have to wait for hundreds of years to see X-ray emission from the location of supernova (SN) explosions. About 25 SN were observed in X-rays within days to years after the outburst in nearby galaxies (see [38] for a recent review and Chap. 7). SN are classified as type II or I based on the presence or absence of hydrogen lines in their optical spectra. While most of the X-ray detected SN are of type II, six are of type Ib/c, Ic, or Ic/pec. No type Ia SN - believed to be nuclear detonations of carbon+oxygen white dwarfs when exceeding the Chandrasekhar limit through accretion - has been detected to date in X-rays. Also some gamma-ray bursts and their X-ray afterglow have been connected to SN explosions in distant galaxies (see Chap. 6). Here I only shortly want to mention some results of the very close by SN 1987A (in the LMC, 50kpc) and SN 1993J (in M 81, 3.6Mpc).

SN 1987A in the LMC was the closest SN exploding during the X-ray era. For the first time, hard X-rays (5 x 1037 erg s-1 in 45-100 keV band) due to emission from radioactive decay of the debris could be observed for a SN starting about half a year after the explosion [71] till it dropped below the sensitivity limit of the instruments 2yrs later. Only in 1991, the SN was redetected as a faint soft X-ray source in ROSAT observations (~1034ergs-1 in 0.5-2keV band [3,28]). Its X-ray flux,


XMM/NEWTON observation □

ROSAT PSPC observations ROSAT HRI observations o / Chandra observation A


days (after explosion)


Fig. 20.7 X-ray emission from the supernova SN 1993J in the galaxy M81 was detected by ROSAT only six days after the explosion at the end of March 1993. The supernova location, south of the center of M81, is indicated by the arrow (right half of the image). In an earlier observation in October 1992, no X-ray emission was detected at the supernova position (left half of the image) (left). The light curve of the X-ray luminosity of SN 1993J gives informations on the mass loss rate and wind velocity of the supernova progenitor star (right [80])

now due to circumstellar interactions, has been continuously rising to soft X-ray luminosities of 1036 erg s"1 till mid 2005 [50].

SN 1993J in M 81 exploded very early in the ROSAT mission live time. Soft emission with an absorption corrected luminosity of 3 x 1039ergs" in the 0.12.4 keV band was detected within 6 days (see Fig. 20.7, [81]). According to standard SN models, the SN X-ray light curve is determined by the interaction of the SN ejecta with the stellar wind of the progenitor star. The well sampled X-ray light curve of SN 1993J in this way revealed the pre-supernova evolution of the progenitor (see e.g. [37,80]).

Was this article helpful?

0 0

Post a comment