In mid-September of 1999, a multiwavelength campaign was carried out on the coronally active RS Canum Venaticorum binary HR 1099 (K1 IV + G5 V; P = 2.84 days), during commissioning of the High Energy Transmission Grating Spectrometer (HETGS) of the Chandra X-ray Observatory (CXO). The coordinated program involved the Extreme Ultraviolet Explorer (EUVE), the Hubble Space Telescope Imaging Spectrograph (STIS), and the Very Large Array (VLA). The objective was to study the system in quiescence, across as much of the electromagnetic spectrum as practical, and to catch any flares that might occur. The EUVE 80-180 A light curve of HR 1099, covering the period September 13-22, showed only a single impulsive outburst, which occurred at the very end of the 9 day pointing. The 3-25 A bremsstrahlung continuum displayed an overall decay during the 1.5 day Chandra observation on September 14-16, with a few superimposed mild enhancements. The VLA 3.5 cm and 20 cm radio light curves, obtained during a 10 hr period on September 15 immediately before the HST pointing, similarly revealed normal quiescent gyrosynchrotron emission and an absence of impulsive events. In contrast, the 7 hr STIS time series later on September 15 contained two distinct flares. The first was accompanied by intensification of the preexisting broad wings of the medium excitation species (e.g., Si IV λ1393 and C IV λ1548), while the second involved primarily the narrow cores of the lines, and even cooler temperatures. The Fe XXI λ1354 forbidden line showed little response to either flare, consistent with the contemporaneous soft X-ray and EUV light curves. The lack of coronal counterparts to the ultraviolet flares is unusual and suggests that they belong to a separate class of outbursts, sharing some similarities with the transition zone explosive events seen on the Sun. The density sensitive O IV λ1400 multiplet was not affected by either flare. The density sensitive Si III λ1300 multiplet showed little response to the first flare, but a dramatic brightening in the second, likely due to the lower peak temperature of that event. The O IV line ratios were near their low-density limits and suggest ne ~ 1010 cm-3 for the duration of the HST observations. The Si III ratios during the second flare rise indicate ne ~ 7 × 1010 cm-3. The far-UV diagnostics jointly imply electron pressures of neT ~ 2 × 1015 K cm-3, if formed close to their respective ionization equilibrium temperatures. The helium-like triplets of O VII, Ne IX, and Mg XI in the HETGS spectra have forbidden-to-intercombination line ratios consistent with average coronal electron densities of 1011 cm-3 at T ~ 0.3-1 × 107 K over the duration of the Chandra pointing. Fe XXI λ102/λ128 and λ142/λ128 ratios from EUVE suggest ne 1012-1013 cm-3, near 107 K. Thus, the coronal electron pressures could be as much as several orders of magnitude larger than those of the lower atmosphere. We constructed time-resolved spectra from the HETGS event lists and measured the centroids of the three brightest X-ray lines of HR 1099 in 60 minute bins. In high-S/N Ne X λ12.1, we believe that we can see the changing radial velocity of the K1 IV star over the half an orbital cycle recorded by Chandra. The amplitude of the velocity change is only ~100 km s-1 over the 1.5 day interval, a small fraction of the 300-500 km s-1 resolution of the HETGS at 12 A. Ne X λ12.1 also displayed a transient blueshift of ~60 km s-1 immediately prior to the first STIS flare, during which the far-UV lines briefly exhibited blueshifts of similar magnitude. The apparent displacement of Ne X appears independently in the -1 and +1 orders of the medium energy band (MEG), but there was no confirmation in the redundant high energy band (HEG), nor in other X-ray lines. Fe XXI λ1354 showed a blueshift about 30 minutes later at the beginning of the second HST orbit, but only at the 20 km s-1 level. Analogous HETGS time series of the brightest lines of comparison star Capella (α Aurigae; G8 III + G1 III) exhibited steady behavior, consistent with measurement uncertainties, without the transient Doppler bursts seen in HR 1099. Although the reality of the Ne X Lyα blueshift is in doubt, there is no question that the Chandra HETGS velocity scales are stable and free from large systematic errors.
Read full abstract
Nov 1, 2023
Luis C Fernandez + 3
Open Access