Comparison of hard X-ray spectra obtained by spectrometers on Hinotori and SMM and detection of “superhot” component

Abstract

We report the results of comparison between hard X-ray spectra in solar flares obtained by broad-band spectrometers aboard Hinotori and SMM satellites. The energy ranges of the Hinotori spectrometer were known only with growing uncertainty because of background X-rays from particle-induced radioactivity in the NaI crystal, which gradually outweighed X-rays from the radioactive calibration source ( 109Cd). Within uncertainty brought about by assuming the typical energy of the background X-rays, calibrations are consistent with constant energy ranges, i.e. same as prelaunch values. Using these energy ranges, we find that spectra by the Hinotori spectrometer are usually consistent with those by the SMM spectrometer for flares in 1981. On the contrary, flares in 1982 persistently show 20–50 % higher flux by Hinotori than by SMM. If this discrepancy is entirely attributable to errors in the calibration of energy ranges, the errors would be ∼10 %. There are other factors which may increase the uncertainty in determining a spectrum. Despite such a discrepancy in absolute flux, in the decay phase of one flare, spectra revealed a hard X-ray component (probably a “superhot” component) that could be explained neither by emission from a plasma of ∼2×10 7 K nor by a nonthermal power-law component. Indeed, imaging observations during this period show nearly cospatial hard X-ray emission with soft X-ray emission, in contrast with earlier times at which hard and soft X-rays come from different places. Contrary to the temperature diagnosis using FeXXVI lines, hard X-ray spectral data give temperatures ∼7×10 7 K for the superhot component.