The facts
On 23 September 2026, NASA announced that it had selected PRIMA (PRobe far-Infrared Mission for Astrophysics) as the very first mission in a new class, the “Probe Explorers”. The telescope is due to launch in 2033, for a mission planned to last five years.
PRIMA carries a 1.8-metre aluminium primary mirror, actively cooled to about 4.5 kelvins, or about −268.6 °C. Two instruments will be installed: a camera, PRIMAger, and a spectrometer, FIRESS, which splits light to identify its components. Together they will observe at wavelengths between 24 and 235 micrometres, in the far infrared.
The cost of the mission is capped at $1.2 billion, not counting the launch or off-project expenses. NASA’s Jet Propulsion Laboratory (JPL) leads the mission, with the Goddard and Marshall centres. Seven partner agencies are taking part, including CNES for France, as well as the Italian, German, Canadian, South Korean, Japanese and British agencies. The Max Planck Institute for Astronomy, for example, will supply two steerable mirrors that adjust the instruments’ field of view. PRIMA now enters Phase B, the preliminary design phase.
The context
The Probe Explorers class was recommended in 2021 by the US astronomers’ decadal survey, Astro2020. It is meant to fill the gap between very large observatories, such as the James Webb Space Telescope, and small missions. In October 2024, NASA had funded two competing studies, of $5 million each: AXIS, an X-ray observatory, and PRIMA. The cost cap was then set at $1 billion, for a launch in 2032.
Far-infrared light is almost entirely absorbed by the Earth’s atmosphere, so it has to be observed from space. The last space observatory able to do so, the European Space Agency’s Herschel, stopped operating in 2013. NASA did operate the Spitzer infrared telescope until 2020, but the liquid helium that cooled its instruments ran out as early as 2009: its detectors at the longest wavelengths were then no longer cold enough to observe. Such a telescope must also be very cold, because its own heat would radiate at these wavelengths and blind its detectors.

Why it matters
Far-infrared light passes through the dust clouds that hide the birth of stars and the growth of black holes at the heart of galaxies. PRIMA is meant to help understand how galaxies have evolved, how dust and heavy elements built up in the Universe, how planets form, and even where the Earth’s water comes from.

Thanks to its cooled mirror, the telescope will be far more sensitive than its predecessors. According to the University of Maryland, at certain wavelengths PRIMA will be able to obtain observations as deep as those of Herschel up to six orders of magnitude faster, that is, a million times more quickly. “It is rare for an instrument that goes into service to be several orders of magnitude more capable than the one that came before it. Even a factor of 10 is rare. Several orders of magnitude is almost unheard of,” says Sylvain Veilleux, professor of astronomy at the University of Maryland and head of PRIMA’s science working group on active galactic nuclei.
Archive note
- Selected on 23 September 2026, launch targeted for 2033 for a five-year mission.
- 1.8 m mirror cooled to around 4.5 K; cost capped at $1.2 billion, excluding launch.
- First far-infrared space observatory since Herschel (2009–2013).
Sources
- NASA, NASA Selects Far-Infrared Telescope as First in New Mission Class, 23 Sept 2026 (primary source)
- CNES, Feu vert pour PRIMA, futur observatoire de l'infrarouge lointain, 24 Sept 2026 (primary source)
- CNES, PRIMA : l'infrarouge lointain, une expertise française au service de la science internationale, 24 Sept 2026
- Université du Maryland, département d'astronomie, Two UMD Astronomers on PRIMA Team Selected for New Class of NASA Astrophysics Missions, 23 Sept 2026
- Max-Planck-Gesellschaft (MPIA), NASA selects new far-infrared telescope PRIMA with MPIA providing crucial hardware, 25 Sept 2026
- NASA Science, NASA Selects AXIS and PRIMA for Astrophysics Probe Explorers Studies, 7 Oct 2024
- NASA Jet Propulsion Laboratory, NASA's Spitzer Telescope Warms Up To New Career, 6 May 2009
- SatNews, NASA Selects Far-Infrared PRIMA Telescope as Initial Billion-Dollar Astrophysics Probe Explorer Mission, 26 Sept 2026
- Clubic, La NASA tient déjà son prochain grand télescope spatial, il s'appelle PRIMA et va observer l'Univers invisible, 24 Sept 2026
- GNT (Génération-NT), La NASA annonce son nouveau télescope spatial PRIMA, 27 Sept 2026
- Ciel et Espace, Fin de la mission Herschel : un premier bilan, 30 Apr 2013
- Cité de l'espace, L'héritage infrarouge de Spitzer, 4 Jun 2020
Image and video credits
- The Taurus cloud seen in the far infrared by Herschel (ESA), the kind of observation PRIMA will take up again.: ESA/Herschel/NASA/JPL-Caltech, acknowledgement R. Hurt (JPL-Caltech), CC BY-SA 3.0 IGO · source
- Spitzer, NASA’s infrared telescope operated from 2003 to 2020 (artist’s impression).: NASA/JPL-Caltech, public domain (NASA) · source
- Stellar nurseries hidden by dust (the W3/W4/W5 complex), observed by Herschel.: ESA/Herschel/NASA/JPL-Caltech, acknowledgement R. Hurt (JPL-Caltech), CC BY-SA 3.0 IGO · source
Corrections
- : We attributed to Alberto Bolatto the sentence about an instrument “several orders of magnitude” more capable; it is by Sylvain Veilleux, professor of astronomy at the University of Maryland and head of PRIMA’s science working group on active galactic nuclei. We have also restored the figure on which it rests: at certain wavelengths, PRIMA will be able to observe up to six orders of magnitude (a million times) faster than Herschel at equal depth.
- : Updated: the conversion of 4.5 kelvins is now given precisely (about −268.6 °C), the point about Spitzer clarified, the title of one source corrected, French-language sources added (CNES, Clubic, Ciel et Espace, Cité de l’espace, GNT).
— The Bunker archivists
Method: at least 3 sources including the official one, cross-checked with the help of an AI (Claude), checked and approved by a human on 28 Sept 2026 in the French original; this English version is an AI translation. Our method
Spotted a mistake? Report it: we correct in public.
