Applications of Integrated Photonic Spectrographs in astronomy

Abstract

One of the problems of producing instruments for extremely large telescopes (ELTs) is that their size (and hence cost) scales rapidly with telescope aperture. To try to break this relation alternative new technologies have been proposed, such as the use of the Integrated Photonic Spectrograph (IPS). Due to their diffraction-limited nature, the IPS is claimed to defeat the harsh scaling law applying to conventional instruments. In contrast to photonic applications, devices for astronomy are not usually used at the diffraction limit. Therefore, to retain throughput and spatial information, the IPS requires a photonic lantern (PL) to decompose the input multi-mode light into single modes. This is then fed into either numerous arrayed waveguide gratings (AWGs) or a conventional spectrograph. We investigate the potential advantage of using an IPS instead of conventional monolithic optics for a variety of capabilities represented by existing instruments on 8 m telescopes and others planned for ELTs. To do this, we have constructed toy models of different versions of the IPS and calculated the relative instrument sizes and the number of detector pixels required. This allows us to quantify the relative size/cost advantage for instruments aimed at different science requirements. We show that a full IPS instrument is equivalent to an image slicer. Image slicing is a beneficial strategy for ELTs as previously demonstrated. However, the requirement to decompose the input light into individual modes imposes a redundancy in terms of the numbers of components and detector pixels in many cases which acts to cancel out the advantage of the small size of the photonic components. However, there are specific applications where an IPS gives a potential advantage which we describe. Furthermore, the IPS approach has the potential advantage of minimizing or eliminating bulk optics. We show that AWGs fed with multiple single-mode inputs from an PL require relatively bulky auxiliary optics and a 2D detector array which significantly increases the size of the instrument. A more attractive option is to combine the outputs of many AWGs so that a 1D detector can be used to greatly reduce the number of detector pixels required and provide efficient adaptation to the curved output focal surface.