- LATEST NEWS
- Instrument Features
- Instrument Overview
- Observing Modes
- Scientific Cases
- MIRADAS contacts at GTC
- Useful Documents
- More Information
Instrument Features
The basic MIRADAS concept is a near-infrared multi-object echelle spectrograph operating at spectral resolution R=20000 over the 1-2.5µm bandpass. MIRADAS selects targets from a 5 arcmin field of view using up to 12 deployable probe arms with pick-off mirror optics, each feeding a 3.7 x 1.2 arcsec field of view to the spectrograph. The spectrograph input optics also include a "slit slicer" which reformats each probe field into 3 end-to-end slices of a fixed 3.7 x 0.4-arcsec format – combining the advantages of minimal slit losses in any seeing conditions better than 1.2-arcsec, while at the same time providing some (limited) two-dimensional spatial resolution. The spectrograph optics then provide a range of configurations providing the observer with the ability to choose between maximal multiplex advantage and maximal wavelength coverage, with several intermediate options, depending upon the needs of the science program. Both "normal" spectroscopy and spectropolarimetry will be possible, thanks to a Polarization Modulator Subsystem which can be inserted into (and retracted from) the beam.
General Parameters
Parameter | Value | Comment |
Target field of regard | 5-arcmin diameter | Each probe arm patrols a 2D workspace within this circular field |
Individual target field of view | 3.7x1.2-arcsec | |
Slit slicer geometry | 3 slices of 3.7x0.4-arcsec each | |
Detector focal plane | 4096x2048 pixels | Mosaic of 2Kx2K HAWAII-2RG |
Spectro‐polarimetry | Linear, circular | Available for single‐object cross‐dispersed mode |
Continuum sensitivity | J=18.0 mag H=17.7 mag K=16.7 mag |
S/N=10 for 1-hour on-source exposure |
Emission line sensitivity | 5x10-18 ergs/cm2/s
(point) 8x10-18 ergs/cm2/s (resolved) |
S/N=10 for 1-hour on-source exposure; resolved source assumes 1 square arcsecond detect cell |
Offered Multiplex Configurations
Configuration | MXS Targets | Instantaneous Bandpass |
SO-Short | 1 | 1.04-1.78 µm |
SO-Long | 1 | 1.34-2.50 µm |
Maximum MXS | 12 | Any SINGLE order from the Table below |
MIRADAS Echelle Orders
Order | Wavelength (µm) | Band | Order | Wavelength (µm) | Band |
14 | 2.3700-2.5000 | K | 24 | 1.4132-1.4718 | Atm. |
15 | 2.2220-2.3820 | K | 25 | 1.3555-1.4095 | Atm. |
16 | 2.0885-2.2245 | K | 26 | 1.3107-1.3493 | J |
17 | 1.9360-2.0860 | K | 27 | 1.2602-1.2988 | J |
18 | 1.8700-1.9700 | Atm. | 28 | 1.2170-1.2531 | J |
19 | 1.7869-1.8534 | H | 29 | 1.1750-1.2100 | J |
20 | 1.6943-1.7608 | H | 30 | 1.1365-1.1703 | Atm. |
21 | 1.6144-1.6809 | H | 31 | 1.1009-1.1291 | J-Io |
22 | 1.5409-1.6044 | H | 32 | 1.0664-1.0937 | J-Io |
23 | 1.4746-1.5355 | H | 33 | 1.0343-1.0607 | J-Io |
34 | 1.0048-1.0303 | J-Io |
Note that additional combinations of multiplex and wavelength coverage are also possible. By selecting different order sorting filters and probe output masks, the observer could choose any combination of target multiplex (NMXS) and wavelength coverage by consecutive orders (Norder), as long as the product NMXS * Norder < 12. For instance, a configuration with 6 MXS targets and a wavelength coverage from 2.0885 to 2.3820 microns (orders 15 and 16) would work.
Instrument Overview
The following figure shows a sketch of the completely assembled instrument. All key elements can be seen, from the flange with which it will be attached to the GTC's Folded Cassegrain E focus, the MXS probe arms (which will also be shown in a separate sketch below), the Decker masks, macro-slicer, filter wheels, and finally the actual spectrograph with the primary echelle grating, collimator, the cross-dispersion gratings, the detector module.



Optical Design Layout
Next plot shows the MIRADAS Optical Concept Block Diagram: Incoming light from the telescope enters the instrument at the top of the figure and is selected and folded by the probe arms in the top two blue-shaded regions. The probe outputs is stacked, sliced, and filtered in the two middle green-shaded regions, after which it enters the cross-dispersed echelle spectrograph, as represented in the lower blue-shaded region.
A pickoff mirror located near the telescope focal plane relays the light down the probe arm, where it encounters a collimating doublet lens. The lens feeds the light through a series of folds in the probe mechanism, which maintain a fixed optical path length while the probe arm is moved to variable target locations in the field of regard. The 4th fold mirror is under the MXS optical bench, and is fixed in its location and orientation. Shortly after the fixed fold comes the cold pupil stop, located at the pupil image created by the collimator doublet. Finally, a re-imaging doublet brings the beam to a focus. At the output focal plane from the MXS system, the individual probes produce fields of 3.7 arcsec × 0.4 arcsec (2.7 mm × 0.8 mm), each of which is separated from the next probe by a 8.5 mm center-to-center distance, creating a sparsely-filled pseudo-longslit at the macro-slicer input.

Components Description
MIRADAS Probe Arms
A key part of the MIRADAS instrument is its multiplexing system (MXS). MIRADAS includes an MXS employing deployable probe arms with pickoff mirrors. The MXS patrol field for MIRADAS is 5 arcmin in diameter, or about 250 mm. The primary function of the MXS probe mechanism is to select targets in the patrol field and relay light from them to the actual spectrograph. This is accomplished using up to 12 independently actuated MXS pick-off probe arms. Each arm patrols a sector of the MIRADAS field, just below the input telescope focal plane. Optics in each arm relay light from a 3.7 arcsec × 0.4 arcsec field.


Decker Masks
The MIRADAS design has different modes or combinations of multiplexing and instantaneous wavelength coverage. The first of these has maximum multiplex (i.e. using all 12 probes) and minimum wavelength coverage (a single order). The second mode uses one probe (single-object mode) and uses about 12 orders to cover almost an octave in wavelength at a single go. Other options can allow medium multiplex modes, with 2 − 3 targets observable with 3 − 4 spectral orders each simultaneously. The actual output geometry of the probes is completely fixed. The selection of multiplex mode shown is achieved using a mask mechanism at the output focal plane, which will pass all probes, “blank off”, “blank off” all but one probe for the single-object mode, or select other probes for medium-multiplex mode.

Polarization Modulator Subsystem
Half-wave plate (HWP) polarization modulator: The angular orientation of the HWP will be selectable using a rotary mechanism. The presence of the HWP will introduce an additional focus shift in the incoming beam from the telescope by ∼ 0.8 mm, but this can be compensated with telescope secondary mirror focusing. Quarter-wave plate (QWP) polarization modulator: The QWP will be insertable into the optical beam immediately “downstream” from the HWP. The angular orientation of the QWP will be selectable using a rotary mechanism. The presence of the QWP will introduce an additional focus shift in the incoming beam from the telescope by ∼ 0.8 mm, but this can be compensated with telescope secondary mirror focusing.
Wollaston Prism Subsystem
The MIRADAS Wollaston prism is located just downstream of the cold stop of one MXS probe arm, thus minimizing its physical size and clear aperture. The Wollaston is carried on an insertion mechanism, allowing it to be removed from the beam for standard (non-SPOL) observations. The purpose of the Wollaston is to deviate the “o” and “e” polarized spectra in opposite directions, achieving a “throw” sufficient to separate them into two separate images on the science detector.
Macro-slicer Subsystem
The next key component for MIRADAS is the “macro-slicer”. This is an image slicing mini-IFU which takes the input focal plane — a sparsely-filled pseudo- longslit — and slices it into a different, nearly-filled, pseudo-longslit of length ∼ 60 mm × 0.14 mm (i.e. dividing each slitlet into three (3) for up to 36 narrower slitlets oriented lengthwise). This is the same approach used for the FISICA and FRIDA image slicers.


The input optics of the MIRADAS spectrograph contains a "slit slicer" which reformats each probe's field into 3 end-to-end slices which are in a fixed 3.7" x 0.4" format. Effectively, it's an image slicing mini-IFU which takes the input focal plane (a sparsely-filled pseudo-longslit), and slices it into a different, nearly-filled, pseudo-longslit of length ∼ 60 mm × 0.14 mm. In other words, the macro-slicer is a stack of six advanced image slicer Integral Field Units (IFUs). Like other advanced image slicer IFUs, there are three sets of mirrors that work together to geometrically rearrange the loosely packed inputs from the probe arms into a tightly packed pseudo-slit. The macro-slicer also passively keeps the spectral resolution of MIRADAS fixed at R ~ 20000 in seeing down to 0.4 arcseconds. Each slice is geometrically rearranged end-to-end by the three pupil mirrors, and re-imaged onto the field mirrors (which are co-located at the image plane generated by the slicer and pupil mirrors). The field mirrors are unpowered fold mirrors and redirect the light into the spectrograph module.
Filter Subsystem
The MIRADAS order-sorting filters are located in 3 wheels located in the diverging f /7.14 focal plane. This allows the filter sizes to be such that 11 filter positions can be fit into each wheel. As a result, thirty (30) such filters can fit into three wheels of ∼ 360 mm diameter (roughly similar to the FLAMINGOS -2 filter wheels).
Dispersers
A single echelle grating serves as the primary dispersion optic for ALL the different modes. In the “full multiplex” MXS mode, with all probes feeding into the instrument, we will only use a single order of the echelle. Thus, the “cross-disperser” will simply be a flat fold mirror (and the order sorting filter has already blocked any “off-order” light from reaching the grating). For single-object (SO) mode, the cross-disperser will be a grating with sufficient dispersion to shift light by about 80 pixels per order, providing an octave (factor of two) of instantaneous bandpass in the range 1 – 2.5 μm. There will be two selectable configurations available in SO mode: either ∼ 1.0 – 1.8 μm (SO-short), or ∼ 1.3 – 2.5 μm (SO-long).
The Instrument Calibration Module
The Instrument Calibration Module (ICM) is a separate module which will be mounted in the Folded Cassegrain E (FC-E) focus: With its deployable mirror, the light from a variety of calibration lamps can be fed into the beam, to obtain arc lamp spectra (for wavelength calibration) or halogen flats. In the ICM at the FCE focus there wil be 3 Uranium-Neon and 3 Uranium-Argon lamps mounted, as well as 3 incandescent lamps for flat fields. Uranium lamps have been chosen over Thorium lamps because of their richness in emission lines in the near-infrared. These will be complemented by the (brighter) Neon and Argon lines.
Observing Modes
Single-object Long-wavelength Mode (SO-Long)
For the SO-long mode, we find an optimal solution for a grating with 90 lines/mm. This projects all of orders 14 through 25 onto the detector mosaic, thus covering 1.34 – 2.50 μm. This is nearly a full octave in wavelength — the theoretical maximum possible coverage. Furthermore, it covers the full long-wavelength bandpass transmitted by the Earth’s atmosphere. At the short wavelength end, it provides ample overlap with the SO-short mode, so there are no gaps in coverage between the two modes.

Single-object Short-wavelength Mode (SO-Short)
The SO-short mode is more challenging because the same factor of two in wavelength occupies more spectral orders. This requires more detector area, and also means that cross-dispersion causes a greater tilt of the orders — also “eating” more detector space. For the SO-short mode, we find an optimal solution for a grating with 150 lines/mm. This projects all of orders 20–32 onto the detector mosaic, thus covering 1.04 μm to 1.78 μm. This provides ample overlap with the SO-long mode at the long wavelength end (nearly 0.5 μm) and covers out to near the end of the atmospheric H-band (thus, extending further into the longer wavelengths by an order would not provide any more scientifically useful data). At the short wavelength end, it approaches to near the nominal 1.0 μm short-wavelength goal, and at 1.04 μm cut-on it easily captures the He I 1.083 μm line (an important stellar diagnostic feature).

Maximum-multiplex Mode (MMX)
For the MMX mode, the cross-disperser is replaced by a flat mirror (also located in the grating carousel). The resulting spectra will be a single order (selected by an order-sorting filter), without any tilt, and with orders separated by the probe separation. As expected, we can cleanly separate the spectra for all 12 MXS probes with the probe spacing.

Spectropolarimetry
MIRADAS will have the capability to do spectropolarimetry (SPOL) in single-object mode. This approach meets the vast majority of the science goals (which are largely single-object oriented, within a given field of view), while providing a basically feasible approach. In the MIRADAS design, the "o" and "e" beams are no longer separated by an order height, but effectively the input slit to the spectrograph is comprised of interleaved "o" and "e" rays (see the sketch below).

This allows the spectrograph to effectively "ignore" the SPOL beam separation, and we can carry out cross-dispersed spectroscopy in this mode. For either linear or circular spectro-polarimetry with MIRADAS, the MXS probe will be placed underneath the half-wave plate (HWP). Spectro-Polarimetry Science Working Group has identified a MXS probe configuration (and thus location in the field of view) which optimally minimizes the instrumental polarization.

Scientific Cases
- Massive Stars in the Milky Way
- Chemo-Dynamics of the Inner Milky Way
- Building Blocks of Galaxy Evolution at Intermediate Redshift
- Infrared Spectro-Polarimetry: New Windows on Stellar Astrophysics
In addition to the Design Reference Cases, the MIRADAS Science Team have also identified many additional science cases which will make excellent scientific use of the MIRADAS capabilities determined by the Design Reference Case drivers, ranging from exoplanets to stellar magnetic fields to Galactic archeology of the Milky Way to compact objects and relativistic astrophysics to intermediate-redshift galaxies.
MIRADAS contact persons at GTC
contact | email @ gtc.iac.es |
Stefan Geier - main contact | stefan.geier |
Riccardo Scarpa | riccardo.scarpa |
Useful Documents
More Information
Last modified: 26 September 2024