Far-infrared and submillimetre observations of the Crab nebula

We present far-infrared (FIR) and submillimetre images of the Crab nebula, taken with the ISOPHOT instrument onboard the Infrared Space Observatory (ISO) and with the Submillimetre Common-User Bolometer Array (SCUBA) on the James Clerk Maxwell Telescope. The ISOPHOT observations were made in the bands centred at 60, 100 and 170 mum, with FWHM of equivalent area Gaussian beams of 44, 47 and 93 arcsec, respectively. The 850-mum SCUBA image was processed using a maximum-entropy method algorithm and has a Gaussian FWHM of 17 arcsec. The 60- and 100-mum images show clear excess of emission, above that expected from an extrapolation of the synchrotron spectrum of the Crab nebula from lower frequencies, as previously seen from IRAS observations. The superior angular resolution of the ISOPHOT images reveal that about half this excess is attributable to two peaks, separated by 80 arcsec. We also present spectra taken using the Long-wavelength Spectrometer onboard ISO, which show that the FIR excess is not due to line emission. The lower-resolution 170-mum image does not show any excess emission, but is possibly fainter, particularly in the NW, than expected from an extrapolation of the lower-frequency synchrotron emission. These findings are consistent with a picture in which the FIR excess is due to emission from a small amount of warm dust (in the ranges 0.01-0.07 and 0.003-0.02 M-circle dot for astronomical silicate and graphite, respectively) which radiates predominantly at 60 and 100 mum but not at 170 mum, and which is superimposed on a synchrotron spectrum that gradually steepens towards shorter wavelengths throughout the far and medium infrared spectral range. The dust geometry is consistent with a torus of diameter approximate to0.8 pc created by the (red supergiant) supernova progenitor prior to its explosion, superimposed upon a broadly distributed component which may be supernova condensates in the filaments. The dust-to-gas ratio in the filaments is comparable to the interstellar value. Therefore, even if the condensates seen at the present epoch could ultimately escape the remnant without being destroyed, the surrounding interstellar medium will not be significantly enriched in dust. Our upper limit of approximate to0.02 M-circle dot on the total mass of carbon in the form of graphite is consistent with the inference from the gas-phase carbon abundances that there has been no significant enrichment of the filaments in carbon nucleosynthesis products from the progenitor. To study spectral index variations in the radio synchrotron emission we compared the 850-mum image with a 20-cm Very Large Array (VLA) image. These images are very similar, implying that there is little variation in spectral index across the face of the remnant between these wavelengths. As seen previously, there are some spectral variations near the centre of the remnant. However, given the good agreement between the integrated flux density at 850 mum and the extrapolated synchrotron spectrum, together with the different epochs of the 850-mum and 20-cm images, we do not see the need for the second radio synchrotron component from the remnant, which has previously been proposed.