Infrared single mode spectrographs for exoplanet applications
This thesis details the process of designing and building a diffraction limited single-mode fiber (SMF)-fed high resolution spectrograph in the near infrared range. The spectrograph, nicknamed Iranti, investigates infrared applications of new technologies in spectroscopic instrumentation. More specifically, we evaluate spectroscopic technologies applicable for precision radial velocity (PRV) work. The PRV spectrographs of today fit an archetype, typically using multi-mode fibers (MMFs) for throughput, high resolution échelle gratings with cross dispersion, housed in vacuum chambers for stability and calibrated with astrocombs or etalons. With the imminent arrival of extremely large telescopes (ELTs), and the proliferation of extreme adaptive optics (ExAO) systems, current design philosophies are reaching a plateau since they lead to correspondingly large spectrographs and do not take full advantage of the correction offered by adaptive optics (AO).
Instrument builders recognize this limitation and have been designing and building a new class of spectrographs operating at the diffraction limit and based on the use of SMFs. This allows the spectrograph size to be independent of telescope aperture and also offers precision improvements without the temporally variable modal noise associated with MMFs. These SMF spectrograph that work in the infrared and intended for coupling light from 8 meter class and larger telescopes though advanced ExAO systems mostly use complex designs and utilize exotic ceramics and alloys.
Iranti explores a different direction. One of the advantages of a spectrograph that works with diffraction limited feeds is an independence of the spectrograph optical design from telescope aperture. Our instrument has an accessible design that is within the reach of small telescopes, organizations and budgets, but yet does not leave out the performance potential that large observatories seek. Iranti starts with a tried design that is shared with the MMF-fed TARdYS spectrograph at the Tokyo Atacama Observatory. The white pupil setup resembles a miniature version of current PRV workhorses. Since it focuses on investigating applicable technology and concepts, Iranti emphasizes flexibility in reconfiguration for different fiber feeds, spectral coverage, detectors and cross dispersion. The spectrograph is built around an R6 échelle grating with 13.33 lines/mm and working beam sizes of _ 25 mm. The single block collimating parabolic mirror also has a slow beam which allows the spectrograph to be aligned without micrometer adjustments.
The fiber injection on Iranti also uses common ferrule connectors/ultra ferrule connectors (FCs/UPCs), allowing for different fibers to be easily swapped for use on the spectrograph. Among the fibers tested, we try out a multi-core fiber (MCF) that produces seven spectra patterns per order. To ensure enough cross dispersion, particularly to accomodate wide spectral patterns from multiple fibers or MCFs, we use a 333 lines/mm volume phase holographic grating (VPH) as the cross dispersing element.
Due to the white pupil design, we can swap cameras on Iranti depending on the detectors being tested. We use a pair of commercial off the shelf (COTS) 50.8 mm lenses for the primary camera configuration on Iranti to pair with an infrared detector. Notably, this camera system offers similar performance to custom commissioned iterations.
For Doppler measurements, the ratio of planet mass to host star mass and orbital radius determines the magnitude of the velocity signal. In planetary systems around M dwarf stars, planets in the habitable zone impart a radial velocity (RV) on the host star which is in the detectable range of current PRV spectrographs. With the Transiting Exoplanet Survey Satellite (TESS) mission focused on identifying nearby G-, K-, and M-type stars via transiting events, the demand exists for more PRV capable spectrographs to conduct RV followups. Since M dwarfs also have spectral emission peaks in the J band, near-infrared (NIR) spectral coverage would be desirable in a Doppler spectroscopy instrument. For this range, we integrate a suitable infrared detector using indium gallium arsenide (InGaAs) photodiodes into Iranti, but it is not without challenges. This thesis reports on the process of adapting an InGaAs-based detector for astronomical use and characterizing the performance for spectroscopy. The detectors are typically marketed to industry and defense customers for tasks like detection of contaminants on an assembly line or determining water content by imagery. Factoring the capabilities of the detector, we center the primary configuration for Iranti around échelle orders 122 to 116, covering a range of 1210 nm to 1280 nm and expect a resolving power of 244,000 to 232,000.
To address mechanical stability with Iranti, we design effective aluminum mounts with only the necessary facility for adjustments. In many cases we separate the adjustments to one dimension at a time. Iranti also has a system for environmental stability by using insulated enclosures and active temperature control. An important part of the system is in measuring and logging environmental conditions and effects so that adverse stability factors can be correlated if not outright eliminated.
The final part of this thesis goes into detail on the data extraction and analysis of spectra so that stability can be quantified. We handle multiple cases of extracting spectra from single fibers and MCFs. We also evaluate the spectral quality by measuring emission lines from uranium calibration lamps and finally, gauge stability by measuring line width centering and order cross correlation. We subsequently use the data extraction and analysis pipeline to evaluate on sky data from our local site and also with spectra captured from a run at Subaru Observatory through a fiber feed from the Subaru Coronagraphic Extreme Adaptive Optics (SCExAO) instrument.