The Infra-Red Telescope (IRT) on board the THESEUS mission
Diego Götz, Stéphane Basa, Frédéric Pinsard, Laurent Martin, Axel Arhancet, Enrico Bozzo, Christophe Cara, Isabel Escudero Sanz, Pierre-Antoine Frugier, Johan Floriot, Ludovic Genolet, Paul Heddermann, Emeric Le Floc'h, Isabelle Le Mer, Stéphane Paltani, Tony Pamplona, Céline Paries, Thibaut Prod'homme, Benjamin Schneider, Chris Tenzer, Thierry Tourrette, Henri Triou
TThe Infra-Red Telescope (IRT) on board the THESEUSmission
Diego G¨otz a , St´ephane Basa b , Fr´ed´eric Pinsard a , Laurent Martin b , Axel Arhancet a , EnricoBozzo c , Christophe Cara a , Isabel Escudero Sanz e , Pierre-Antoine Frugier a , Johan Floriot b ,Ludovic Genolet c , Paul Heddermann d , Emeric Le Floc’h a , Isabelle Le Mer a , St´ephane Paltani c ,Tony Pamplona b , C´eline Paries b , Thibaut Prod’homme e , Benjamin Schneider a , ChristophTenzer d , Thierry Tourrette a , and Henri Triou aa AIM, CEA-Irfu/DAp, CNRS, Universit´e Paris-Saclay, F-91191 Gif-sur-Yvette, France b Laboratoire d’Astrophysique de Marseille, UMR 7326,CNRS, Universit´e d’Aix Marseille, 38,rue Fr´ed´eric Joliot-Curie,Marseille, France c Department of Astronomy, University of Geneva, Chemin d’Ecogia 16, 1290 Versoix,Switzerland d Institut f¨ur Astronomie und Astrophysik, Abteilung Astronomie, Universit¨at T¨ubingen, Sand1, D-72076 T¨ubingen, Germany e European Space Agency, ESTEC, Keplerlaan 1, 2201 AZ, Noordwijk, The Netherlands
ABSTRACT
The Infra-Red Telescope (IRT) is part of the payload of the THESEUS mission, which is one of the two ESA M5candidates within the Cosmic Vision program, planned for launch in 2032. The THESEUS payload, composedby two high energy wide field monitors (SXI and XGIS) and a near infra-red telescope (IRT), is optimized todetect, localize and characterize Gamma-Ray Bursts and other high-energy transients. The main goal of theIRT is to identify and precisely localize the NIR counterparts of the high-energy sources and to measure theirdistance. Here we present the design of the IRT and its expected performance.
Keywords:
Gamma-Ray Bursts, infra-red telescopes and instrumentation
1. INTRODUCTION
THESEUS (Transient High Energy Sky and Early Universe Surveyor) is one of the candidate M5 missions withinthe ESA Cosmic Vision program (see Amati et al. this volume). It is currently in phase A (until mid 2021), and ifTHESEUS is selected, it will be launched 2032. The THESEUS mission is being designed as a multi-wavelengthobservatory, whose scientific goal is to increase the discovery space of the high-energy transient phenomena andespecially Gamma-Ray Bursts over the entirety of cosmic history, and especially at redshifts larger than 5.5. Thepayload to be carried by THESEUS is composed by the XGIS (X and Gamma-ray Imaging Spectrometer) codedmask telescopes, operating in the 2 keV–10 MeV energy range, the SXI (Soft X-ray Imager) focusing telescopes,operating in the 0.3–5 keV energy range and the near-Infra-Red Telescope (IRT) sensitive in the 0.7–1.8 micronband. The main goal of the IRT is to identify, accurately localize ( ≤ (cid:48)(cid:48) ) and measure the distance of the NIRcounterparts (redshift determination to an accuracy of 10% or less) of the high-energy sources discovered by theXGIS and SXI. In addition, the IRT will be used to characterize the afterglows, through spectroscopy for a partof them (spectroscopic redshift, neutral hydrogen absorption, presence of metals), and as a multi-purpose agileNIR observatory in space through the implementation of a dedicated Guest Observer (GO) programme, and aTarget of Opportunity programme, with special emphasis on multi-messenger and time-domain astrophysics.The IRT has thus been designed in order to implement imaging capabilities over a wide field of view (15 (cid:48) × (cid:48) ), and moderate resolution spectroscopy (R ∼ (cid:48) × (cid:48) ). The IRTresponsibility is shared among ESA (telescope, thermal control, and detector procurement) and a consortiumlead by France, in collaboration with Switzerland and Germany, that will deliver the IRT instrument. E-mail: [email protected] a r X i v : . [ a s t r o - ph . I M ] F e b . THE IRT TELESCOPE The telescope optical concept is under responsibility of ESA to guarantee any opto-thermo-mechanical conceptsdeveloped by the different prime contractors can interface with the single design of the IRT instrument.The IRT telescope is a focusing three-mirror Korsch configuration with an off-axis (0.884 ◦ ) field. Betweenthe secondary and the tertiary mirrors of the Korsch, there are two additional optical components: a field stopat the intermediate focus and a flat folding mirror, see Fig. 1, to facilitate accommodation in the spacecraft.Note that in the absence of the at folding mirror, the three powered mirrors of the Korsch (M1, M2 and M3) arecoaxial, the common axis being coincident with the axis of radial symmetry of each mirror. The mirror shapesare conics of revolution. The aperture stop is at the primary mirror. Its image, the exit pupil, in the convergingoutput beam is the interface with the IRT instrument. Table 1 gives a summary of the telescope key figures. Table 1. Main characteristics of the IRT telescope.
Telescope type Off-axis KorschEntrance pupil diameter 700 mmM1-M2 distance 675 mmExit pupil diameter 36 mmCollecting area > Wavelength range 700–1800 nmThroughput >
80 %Pixel scale 0.6 arc secFocal length 6188 mmThe optical scheme of the telescope is sketched in Figure 1. The optical design will implement two separatedfield of views, one for photometry with a minimal size of 15 ×
15 arc min (potentially extended to 17 ×
20 arcmin), and one for spectroscopy of 2 × ∼
400 at 1.1 µ m) slit-less spectroscopy in the 0.8-1.6 microns range. Figure 1. IRT Optical scheme. M1, M2, M3 are under ESA/Prime responsibility. The exit pupil represents the opticalinterface with the IRT Instrument provided by the consortium.
The different observation modes will be implemented through the design of the IRT Camera (IRT–CAM) thatincludes a filter wheel, carrying the different optical filter, as well as a grism, which will allow for spectroscopy,2ee Fig. 2. The IRT-CAM will consist mainly in a structure (IRT–STR), an isostatic mount (IRT-ISM), a filterwheel assembly (IRT–WA, see section 2.1), a calibration unit assembly (IRT–CUA), and a focal plane assembly(IRT–FPA), hosting the detector and its associated cold electronics. The whole structure will be covered withthermal isolating blankets. The detector currently envisaged for the IRT FPA is a Teledyne H2RG, sensitive inthe 0.7–2.5 microns wavelength range.
Figure 2. Left: Full optical scheme of the IRT, including IRT-CAM and optical bench. Right: exploded view of theIRT-CAM and its subsystems (without insulating blankets).
The IRT wheel assembly is a rotating mechanism that ensures the positioning in the beam of the various opticalcomponents. The mechanism has eight positions (one grism, five filters, one open position (window) and oneclosed position). It consists of two sub-assemblies: the wheel structure and the actuator assembly. These aresketched in Figure 3. The wheel is made of invar to ensure an excellent mechanical and thermal stability ofthe optical components in the whole IRT temperature range. Invar is also used for the bottom part of theactuator assembly due to constraints at the mechanism mounting interface on the IRT–STR. The actuatorassembly comprises a SAGEM stepper motor, the shaft, and the angular ball bearings. The shaft is made ofstainless steel 15-5PH for a good thermo-elastic compatibility with the ball bearings and the motor rotor. Ahirth coupling, using tapered teeth oriented towards a central point, is implemented between parts made fromdifferent materials (invar and stainless steel). The hirth coupling teeth mesh together, stiffening the mechanismand allowing relative radial motions between the invar and stainless steel parts due to the CTE mismatch. Thisensures that the angular position between these parts is kept within the required tolerances in the range oftemperatures of interest for the IRT operations. An Eddy current contact-less positional sensor is also includedin the actuator assembly in order to provide an angular reference position.
3. INTERFACES AND RESOURCES
The optical interface of the IRT–CAM and the telescope is at the telescope exit pupil corresponding to thefilters positions in Fig. 2. The IRT instrument consists of the IRT–CAM and of the IRT Data Handling Unit(IRT–DHU). Both the IRT–CAM and the IRT–DHU will be provided by the Instrument Consortium to thePrime Contractor through ESA as Customer Furnished Items (CFIs). The integration of the IRT–CAM andIRT–DHU with the IRT telescope onto the spacecraft will be the responsibility of the Prime contractor. TheIRT–CAM is accommodated onto the IRT main bench that also supports the telescope.The cooling of the IRT Camera is ensured thanks to two Cold Fingers (CF1 and CF2) connected to a cryo-cooling system (the cryo-cooler(s) will be provided by the Prime Contractor and the thermal interface to the3 igure 3. Filter wheel assembly. Actuator assembly in blue and wheel structure in red. instrument consists in the cold fingers). Figure 4 shows the intended position for CF1 and CF2: CF2 will be onthe IRT–WA and CF1 on the IRT–FPA.
Figure 4. Left: Representation of the IRT instrument cold fingers (orange colour). Right: IRT Instrument electricalarchitecture.
Taking into account the scientific requirements in terms of thermal background, we defined the followingtemperatures for the IRT-CAM: the ISA and the WA shall be kept at 160K (CF2), while the detector assembly(DA) and the FPA shall be kept at 120K (CF1), thanks to the cryo-cooler(s).To maintain the temperature stability of the camera (at IRT–FEE (Front-End-Electronics), IRT–FPA, andIRT–WA levels), we will consider, at instrument level, to use temperature probes as well as heaters that shallbe controlled by a dedicated electronic (function within the Detector Control Electronics IRT–DCU and FilterWheel control electronics IRT-WCU (TBC)). As a baseline, we consider that the cryocooler will provide thespecified temperature and thermal power at the cold finger (CF1 and CF2) and that the fine control of theinstrument temperature is ensured by the IRT using heaters (opened or closed loop). Concerning the powerdissipation at CF1 and CF2, it is expected to be in the range 1.9–2.8 and 2.8–6.7 W, respectively.Figure 4 depicts the overall electrical architecture of the IRT Instrument. It includes a warm electrical sub-4ssembly (DHU) and cold electrical sub-systems: the IRT–FPA, the IRT–WA and the IRT–CUA. The IRT DHU(Data Handling Unit) will have digital interfaces with the S/C and the IRT–FPA, and analog interfaces with theFPA (bias and signals), with the WA (motor control) and with the CUA. A primary power interface with theS/C will complete the electrical interfaces.The main dimensions of the IRT Camera are depicted in Figure 5. The mechanical interface to the opticalbench is a hexapod, each of its six feet is bolted with three screws. The IRT DHU main dimensions are 210 mmx 228 mm x 220 mm (not shown).
Figure 5. IRT Instrument main dimensions.
Given the design and interfaces presented above the total mass and power budget for the IRT, excludingharnesses, and including typical phase A margins is reported in Table 2, where we also report the telemetrybudget. The latter is based on the fact, that a series of images in different filters are needed to acquire theGRB afterglow position and the photometric redshift (see later). For a complete sequence up to the photometricredshift we estimate that at least 252 Mbytes of data need to be transmitted to ground. If we include the GRBcharacterization mode between 25 Mbytes and 1.2 GBytes per GRB are needed. For the GO mode, a minimalamount of 80 kbits/s are needed.
Table 2. IRT–CAM resources.
Element Mass (kg) Average Power (W) Maximum/AverageRequired Telemetry (Gb/day)IRT–CAM 32.5 36.2IRT-DHU 5.6 30.1Total Instrument 38.2 66.3 40/13.5
4. OPERATIONS AND IN-FLIGHT CALIBRATION
The IRT operation sequence after a GRB trigger is summarized in Figure 6. Once a GRB is detected by theSXI and/or the XGIS, a slew is requested to the platform in order to place the GRB error box within the IRTphotometric FOV. Then, when the SC is stabilized, the IRT enters the follow-up mode, where a 150 s exposurefor each of the available filters (I, Z, Y, J and H) is acquired; the depth of each image shall exceed 20.4 mag(AB). Thanks to these images and an on-board catalogue (based on Gaia and Euclid surveys), the IRT shallbe able to autonomously identify the GRB afterglow candidate, compute its coordinates (to a better than 5 arcseconds accuracy) and its photometric redshift (expected accuracy better than 10%) and send this informationto ground. Then, as a function of the identified source flux, the IRT enters either the characterization mode,which includes spectroscopy mode for 1800 s (if the source is brighter than 17.5 mag (H, AB)) followed by 18005f deep imaging, or directly the deep imaging mode for 3600 s. In order to activate the spectroscopic mode, thesatellite needs to perform a small slew to put the afterglow positions within the 2 × ×
200 pixels centred on three sources (one is thetarget, the other two astrometric and photometric references) and not the entire FoV frames will be transmittedto ground. IRT can also be operated in calibration mode, see below..
Figure 6. IRT science operations after a GRB trigger.
Calibration observations will be performed at regular intervals (the frequency depends on the nature of theinformation required) to obtain the data needed to produce the calibration files and to monitor the good healthof the instrument. To monitor the detector health (bad pixels, flat field, linearity) an on-board calibration unit(CUA) will be used. On the other hand, in order to be compliant with the photometric accuracy requirements( <
5. IRT EXPECTED PERFORMANCE
The main IRT characteristics in terms of expected scientific performances is summarized in Table 3.In order to precisely estimate the photometric capabilities of the IRT, we have developed a simulation tool,that takes into account the detector performance (readout noise, dark current, pixel size, . . . ), the operating con-ditions (temperature, photometric aperture), the satellite planned pointing (expected zodiacal light backgroundand out of field stray-light), the observation conditions (source characteristics, exposure and individual frameduration), the filter characteristics, and the optical imperfections of the system. The latter include the satel-lite jitter and drift, which together with the detector readout noise are the main limitations to the instrumentsensitivity.Taking all these effects into account, we can realistically estimate the expected limiting sensitivity of the IRT.Our simulations indicate that we are always compliant with the requirements of Table 3, both for the ground6 able 3. IRT Performance.
IRT Characteristic ValuePhotometric sensitivity per filter I: 20.9 (goal:21.3)(AB, in 150 s, SNR = 5) Z: 20.7 (goal: 21.2)Y: 20.4 (goal: 20.8)J: 20.7 (goal: 21.1)H: 20.8 (goal: 21.1)Photometric accuracy 5%Expected photo-z accuracy < < < ≥ Figure 7. Reconstructed photometric redshift vs. injected ones for a Monte Carlo simulation of the IRT detected GRBsat the end of the photo-z sequence (based on ). The yellow stripe represents the 10% accuracy region. Detailed simulations have been performed also for the IRT spectroscopic mode. The following steps havebeen implemented to provide a preliminary evaluation of the performances of the IRT spectrometric mode:1. A preliminary analysis based on a simplified geometric model gives access to the linear dispersion that hasto be implemented in order to reach the specified Resolving powers (R). This has been performed at opticaldesign level. 7. Once dispersion is set, we simulate parts of the spectra in a process involving, dispersion, geometricalaspects, diffraction, platform stability, straylight/backgrounds, etc. This analysis gives access to the pixeldependent Instrument Spread Function (ISF), which gives the spectral content of the signal susceptible tobe collected by each pixel. Then, the per-pixel noise can be computed to finally derive the SNR. This isdone for a few columns of pixels (typically 10 to 20) distributed over the spectral axis, and a few contiguouspixels in the spatial axis (typically 3).3. Derive the actual resolving power.4. Derive the processed SNR/R, i.e. after summing the contiguous rows of pixels, and stacking individualexposures done during a sequenceA typical output of these steps is shown in Fig. 8.
Figure 8. Top left: Actual resolving power as a function of wavelength. Bottom left: Linear dispersion in physical unitson the detector plane: spectrum spreads almost linearly over 18.6 mm. Right panel: Source signal and background signalover three columns and combined SNR with the contribution of each column color coded.
Given the longer exposure of 1800 s (with individual frames of 60 s), the satellite stability plays a strongrole in the spectral performance. In order to assess the IRT spectroscopic performance in the most realistic way,we used several S/C simulated drift patterns, expected for the THESEUS orbital conditions and the expectedattitude control system performance. We also weighted our results in terms of expected zodiacal backgroundover the foreseen THESEUS sky pointings. The source and background signals have been estimated for eachdetector pixel, and some margin has been taken for imperfect stacking of the single frames (note that in order tostack the individual 60 s frames we used the “0th” spectral order as a guide). We could in this way estimate theexpected SNR of each spectral bin (summed over three detector lines), and this as a function of the resolvingpower for each wavelength. With this performance we expect to be able to determine the spectroscopic redshiftof GRBs (also for those at z <
6. CONCLUSIONS
We presented the phase A design of the IRT Telescope, which is part of the payload of the THESEUS M5project. We have shown that we the current design we are able to meet the required scientific performance, and8ence, if THESEUS is selected, IRT will play a key role in detecting, localizing and measuring the distance ofcosmological Gamma-Ray Bursts in providing essential information to all those facilities interested in the deepUniverse science the ’30s.
Acknowledgments
This work has been partially funded by the French Space Agency (CNES). D.G. acknowledges financial supportby LabEx UnivEarthS (ANR-10-LABX-0023 and ANR-18-IDEX-0001).
REFERENCES [1] Corre, D.,