MATISSE (the Multi AperTure mid-Infrared SpectroScopic Experiment)

Principal Investigator P.I. : Bruno Lopez
CO-Is: Thomas Henning (MPIA), Gerd Weigelt (MPIfR), Walter Jaffe (NOVA), Farrokh Vakili (INSU)

The MATISSE Consortium:

INSU: Observatoire de la Cote d'Azur and University of Sophia-Antipolis, Nice
Max-Planck-Institut für Astronomie (MPIA), Heidelberg
Max-Planck-Institut für Radioastronomie (MPIfR), Bonn
Netherlands Research School for Astronomy (NOVA), Leiden
Institut für Theoretische Physik und Astrophysik, Universität Kiel
Institut für Astronomie, Universität Wien

Instrument Responsible Andreas Glindemann
Instrument Engineer
Paul Bristow
Instrument Scientist Markus Schöller
Instrument Science Team Olivier Absil (Liege)
Hans-Ulrich Käufl (ESO, chair)
Mario Lattanzi (Torino)
Fabien Malbet (Grenoble)
Paulo J.V. Garcia (Porto University)
Huub Rottgering (Leiden)
Location VLTI Laboratory



  • Preliminary acceptance Chile: 2019
  • First light on telescope: Early 2018
  • Preliminary acceptance Europe: September 2017
  • Now
  • Final Design Review, March 2012
  • Optical and Cryogenics Final Design Review, September 2011
  • Preliminary Design Review, December 2010

Baseline Specification

Optical Throughput 15% (goal 25%) in L and N band
Wavelength coverage L, M and N band
Spectral Resolution 20< R <1000 in L band, 20 < R <550 in M band and 20 < R < 250 in N band
Field of View n/a
Spatial Sampling n/a
Interferometric Contrast 0.6 (goal 0.75) in L and N band
Observing modes High Sensitivity (HighSens) and Simultaneous Photometry (SiPhot)

Scientific Objectives

The objective of MATISSE is image reconstruction. It will extend the astrophysical potential of the VLTI by overcoming the ambiguities existing in the interpretation of simple visibility measurements. MATISSE will measure closure phase relations thus offering an efficient capability for image reconstruction.

The unique performance of MATISSE is partly related to the existence of the four large apertures of the VLT (UTs) that permits to push the sensitivity limits to values required by selected astrophysical programs such as the study of Active Galactic Nuclei and protoplanetary discs.

Moreover, the evaluated performance of MATISSE is linked to the availability of ATs which are relocatable in position in about 30 different stations allowing the exploration of the Fourier plane with up to 200 meters baseline length. Key science programs using the ATs cover for example the formation and evolution of planetary systems, the birth of massive stars as well as the observation of the high-contrast environment of hot and evolved stars.

During Phase A, three constituents of the planetary systems were idendified for which MATISSE will bring new insight:

  • Protoplanetary disks (T Tauri, HAeBe) and planetary debris disks (beta Pic type),
  • Minor bodies of our solar system: main belt asteroids and the comets,
  • Young giant planets and so-called hot Jupiter-like planets

The AMBER and MIDI instruments have started to observe the brightest protoplanetay disks in the infrared sky, approximately a dozen objects. The current capabilities of other observatories are similar (see e.g. Keck Interferometer)

.In addition, in our own solar system, a few asteroids have been observed by MIDI (Delbo et al. 2009, ApJ 694, 1228), demonstrating the feasibility to characterize solar system minor bodies with interferometry.

However, for extrasolar planet detections and/or characterization, interferometry has not reached enough dynamic range so far to successfully observe any of them (see e.g. Matter et al. 2010, A& A 515, 69 for MIDI, or Millour et al. 2008, SPIE 7013, 41 and Absil et al. 2010 in press for AMBER).

The second important science topic is active galactic nuclei. The nominal MATISSE sensitivity in blind mode at N-band is 0.6 Jy in 4-telescope mode, similar to the nominal MIDI sensitivity, although to date MIDI has been used to obtain correlated fluxes of AGNs down to 0.17 Jy. The nominal L-band sensitivity from the performance analysis is 0.1 Jy. For MIDI observations, various lists of AGNs have been assembled, often based on the list of Veron-City and Veron 2006 (A& A 455, 773).

Instrument Description

MATISSE is a mid-infrared spectro-interferometer combining the beams of up to 4 UTs/ATs of the VLTInterferometer.

The number of combined beams is 4. The instrument will be able to operate with 3 or 2 beams. The instrument sensitivity, sampling and throughput are optimized for L and N. The L band is specified from 3.2 to 3.9 μm and the N band from 8.0 to 13.0 μm. MATISSE will operate also in M band, from 4.5 to 5.0 μm. The L, M and N bands can be observed simultaneously.

The instrument will be able to observe with different spectral resolutions. 2 spectral resolutions are possible in N band (R ≈ 30, R ≈ 200) and 3 in L&M bands (R ≈ 30, R ≈ 500 for L and M, R ≈ 1000 for L only). Due to readout time, the full simultaneous coverage of the L&M bands in low and medium resolutions, and the full coverage of the L band in high spectral resolution require an external fringe tracker.

MATISSE will measure: coherent flux, visibilities, closure phases and differential phases. Differential visibilities can also be derived. These quantities will be measured as a function of the wavelength in the selected spectral bands and resolutions.

MATISSE will have an imaging mode (2D image observation without dispersion) for field acquisition and a non-interferometric imaging mode (photometric channels) for flux measurements. It will have also internal devices allowing detector calibration (flat-fielding, bad pixel map), relative flux calibration, wavelength calibration, and instrumental contrast measurement.

MATISSE is a four-beam experiment with a multi-axial global combination. The interferometric beam and the photometric beams receive respectively I and P fraction of the incoming flux (observations without photometric channels are also possible). For an observation with 4 telescopes with photometric channels, 5 images are produced on the detector (4 photometric channels and the interferometric one).During observations with 4 telescopes, the interferogram (in each spectral channel) contains a pattern with 6 fringe periods and is dispersed in the spectral direction. The spatial size of this interferometric channel is larger than the photometric ones in order to optimize the sampling of the 6 fringe structures. The beam combination is made by the camera optics. At this level, the beam configuration is non redundant (separation B between beams equal to 3D, 9D and 6D where D is the spatial diameter of the beam) in order to avoid crosstalk between the fringe peaks in the Fourier space.

The Fourier transform of each spectral column of the interferometric image is thus composed by 6 fringe peaks centered at different frequencies Bij/λ (3 D/λ, 6 D/λ, 9 D/λ, 12 D/λ, 15 D/λ, 18 D/λ) and a low frequency peak that contains the object photometry and the thermal background coming from the 4 telescopes. In order to measure the coherent fluxes with a good accuracy, the design of MATISSE is based on the use of spatial filters, including image and pupil stops.

In order to measure closure and differential phases with a good accuracy, a beam commutation can be made in order to reduce the effect of the instrumental defects on the useful signal.

To measure the coherent fluxes and all the derived interferometric measures such as the differential visibility and phase and the closure phase, the key problem is to eliminate the cross talks between the low frequency peak and all the other peaks that introduce sensitivity of the fringe peaks to variations of the thermal background. Two methods are combined in MATISSE to ensure this result with a large margin: spatial modulation like in AMBER combined with temporal modulation like in MIDI. In addition, to measure the absolute visibility we also have to find the true source photometry. To do that, it is necessary to separate the stellar flux from the sky background, using chopping.

Some devices such as artificial sources, hot screen, optics for flat field or pupil visualization, special material for spectral calibration are implemented in the instrument in order to perform alignment, test, maintenance, calibration and acquisition operations.MATISSE will operate with 2 modes:

  • High Sensitivity mode: this mode has no photometry and all photons are collected in the interferometric beam. This maximizes the sensitivity and also the SNR on the coherent flux and the differential and closure phases. Chopping is optional in this mode
  • Simultaneous Photometry mode: this mode uses photometry (2/3 of flux in the interferometric channel and 1/3 in the photometric ones) and chopping to measure the average source photometry and therefore extract the visibility from the coherent flux (the chopping period is longer than the coherence time and hence the chopping has no influence on the limiting magnitude). For an observation with 4 telescopes with photometric channels, 5 images are produced on the detector (4 photometric channels and the interferometric one).

Instrument Design

The figure shows a 3D optical layout for the set of four beams inside the cryostat for the LM band.


References and external links

Matisse Consortium