NIRPS (Near Infra Red Planet Searcher) is a new ultra-stable infrared ( YJH) fiber-fed spectrograph that will be installed on ESO’s 3.6-m telescope in La Silla, Chile. Aiming at achieving a precision of 1 m/s, NIRPS is designed to find rocky planets orbiting M dwarfs, and will operate together with HARPS (High Accuracy Radial velocity Planet Searcher). In this paper we describe NIRPS science cases, present its main technical characteristics and its development status.
Since 1st light in 2002, HARPS has been setting the standard in the exo-planet detection by radial velocity (RV) measurements. Based on this experience, our consortium is developing a high accuracy near-infrared RV spectrograph covering YJH bands to detect and characterize low-mass planets in the habitable zone of M dwarfs. It will allow RV measurements at the 1-m/s level and will look for habitable planets around M- type stars by following up the candidates found by the upcoming space missions TESS, CHEOPS and later PLATO. NIRPS and HARPS, working simultaneously on the ESO 3.6m are bound to become a single powerful high-resolution, high-fidelity spectrograph covering from 0.4 to 1.8 micron. NIRPS will complement HARPS in validating earth-like planets found around G and K-type stars whose signal is at the same order of magnitude than the stellar noise. Because at equal resolving power the overall dimensions of a spectrograph vary linearly with the input beam étendue, spectrograph designed for seeing-limited observations are large and expensive. NIRPS will use a high order adaptive optics system to couple the starlight into a fiber corresponding to 0.4” on the sky as efficiently or better than HARPS or ESPRESSO couple the light 0.9” fiber. This allows the spectrograph to be very compact, more thermally stable and less costly. Using a custom tan(θ)=4 dispersion grating in combination with a start-of-the-art Hawaii4RG detector makes NIRPS very efficient with complete coverage of the YJH bands at 110’000 resolution. NIRPS works in a regime that is in-between the usual multi-mode (MM) where 1000’s of modes propagates in the fiber and the single mode well suited for perfect optical systems. This regime called few-modes regime is prone to modal noise- Results from a significant R and D effort made to characterize and circumvent the modal noise show that this contribution to the performance budget shall not preclude the RV performance to be achieved.
Radial velocity instruments require high spectral resolution and extreme thermo-mechanical stability, even more difficult to achieve in near-infra red (NIR) where the spectrograph has to be cooled down. For a seeing-limited spectrograph, the price of high spectral resolution is an increased instrument volume, proportional to the diameter of the primary mirror. A way to control the size, cost, and stability of radial velocity spectrographs is to reduce the beam optical etendue thanks to an Adaptive Optics (AO) system. While AO has revolutionized the field of high angular resolution and high contrast imaging during the last 20 years, it has not yet been (successfully) used as a way to control spectrographs size, especially in the field of radial velocities. <p> </p>In this work we present the AO module of the future NIRPS spectrograph for the ESO 3.6 m telescope, that will be feed with multi-mode fibers. We converge to an AO system using a Shack-Hartmann wavefront sensor with 14x14 subapertures, able to feed 50% of the energy into a 0.4" fiber in the range of 0.98 to 1.8 <i>μm</i> for M-type stars as faint as I=12.