SAFARI : a far-infrared imaging spectrometer for SPICA
Sensitive spectroscopy in the infrared wavelength domain allows astronomers to directly assess the physical state and energy balance of cool matter in space, and thus enabling them to study in detailed the various processes that govern formation and evolution of planets, stars and galaxies over cosmic time. After missions that employed warm or a small size telescopes with high levels of background emission, SPICA will provide the next step in far-infrared research with a large cold telescope in combination with instruments employing modern ultra-sensitive detectors. The SPICA far infrared spectrometer SAFARI has been designed to fully profit from the low background of the telescope and from the achieved high sensitivity of modern Transition Edge Sensor (TES) detectors. The grating based concept SAFARI will provide an unprecedented sensitivity of about 5 x 10-20 W/m2 with TES performance as achieved to date – with further TES improvements SAFARI will become even more sensitive.
The baseline SAFARI design uses a beam steering mirror (BSM) that forwards the incoming signal to the dispersing and detection optics. The BSM is used to select sky or calibration signals and forward that to a nominal R~300 (low) resolution optics chain or to a high resolution optics chain. The low resolution is obtained by dispersion through a diffraction grating illuminating a line of TES detectors. For the high resolution mode the signal is fed through a Martin-Pupplet interferometer before entering the grating, yielding spectra with a resolution of R~11000 at 35m to R~1500 at 230m. The full 35-230 μm wavelength range is split in to four different bands, each with its own grating and TES detectors. The baseline design has for each of the bands three separate spatial pixels, to provide background reference measurements, but also to provide some imaging capability.
With SPICA’s cold, 2.5 meter telescope and the baseline TES NEP of 2x10-19 W/√Hz, for the new grating based SAFARI the sensitivity of the R~300 mode will be about 5 x 10-20 W/m2 (5σ, 1hr). With the current design further improvements in TES performance will directly lead to even higher instrument sensitivity – today’s state of the art TES performance is already close to 1x10-19 W/√Hz! With this high sensitivity astronomers will e.g. be able to detect the [OIV] line in relatively average galaxies out to a redshift z~3. Thus the evolution of galaxies can be followed through their most active periods in cosmic time from about 10 billion years ago to what they look like today. Additionally we will be able to observe dust features from even earlier epochs, out to redshifts of z~7-8, thus providing insight into dust formation in the very early phases of the universe.