ESA’s Darwin space interferometer Huub Röttgering Sterrewacht Leiden.

ESA’s Darwin space interferometer

Huub RöttgeringSterrewacht Leiden

The InfraRed Space InterferometerDARWIN

• 2014• 6 1.5 m telescopes• Hexagonal configuration• Beam combiner• Passive cooling (40 K): 5-20 micron

Ringberg, 5-Sept-2003 Imaging with Darwin Page 3

Overview

Introduction– Timeline / status project– Relation with NASA’s Terrestrial Planet

Finder Imaging considerations Science

Ringberg, 5-Sept-2003 Imaging with Darwin Page 4

Finding and characterising exo-Earth’s– Nulling interferometry

Science

The Problem

Detecting light from planets beyond solar system is hard:– Planet emits few

photons/sec/m2 at 10 m– Parent star emits 106 more– Planet within 1 AU of star– Dust in target solar

system 300 brighter than planet

Finding a firefly next to a searchlight on a foggy night

Ringberg, 5-Sept-2003 Imaging with Darwin Page 6

Finding and characterising exo-Earth’s– Nulling interferometry– Atmosphere -> CO2

– Wet and pleasant H20– Life O3 (? / !)

High resolution and sensitive IR imaging– Cophasing using an off-axis reference star

Science

CO2

O3

H2O

(m)

6 8 10 12 14 16 18

Earth at 10pc

Ringberg, 5-Sept-2003 Imaging with Darwin Page 7

Darwin timeline 1993: Léger et al

– ``Darwin proposal’’ 2000 Presentation Alcatel system level study 2004 Results significant technology development

program (15 Meuro)– Optical components, coolers, thrusters, metrology,

control software, 2 breadboards … 2007 – SMART2 techno demonstration flight

– (mainly LISA technology) 2010 – SMART3 techno demonstration flight

– 2-3 space craft 2014 – launch

Ringberg, 5-Sept-2003 Imaging with Darwin Page 8

NASA’s TPF

Similar goals and timelines

1999:

IR interferometer with cooled 4x3.5 m mirrors and ~75-1000 m baseline

Vegetation edge

Blue sky

Earthspectrum

fromEarth-shine

2000

Variable-Pupil Coronagraph IR Nulling Interferometers

Large Aperture IR Coronagraph

SVS

coronagraphe

M2

M3

M1

Hyper-telescope

2001: 4 different studies

Variable-Pupil Coronagraph

IR Nulling Interferometers

•Coronagraph – Difficult•10-15 meter mirror with rms surface ~< 1 Å

–Deformable mirrors - control to <1 Å rms over wide range of scales

–Wavefront sensing - adequate for <1 Å control

•Interferometer - Complex–Cryogenic nulling - 10-5 or 10-6 depth across ~1 octave

–Wavefront & amplitude control - spatial filter in mid-IR (+ DM for low spatial freqs) + control of thermal & vibration effects + acc. amplitude measurement

–Beam transport issues (rejection of stray light at small angles)

2002: down selection for 2 concepts

Ringberg, 5-Sept-2003 Imaging with Darwin Page 12

Joined ESA/NASA mission MOU: aims for a joining in 2006 Plan

– Both sides continue technical studies– Regular scientific contact– Criteria to guide continuation after 2006

• #1: Sensitivity in finding and characterizing exoplanets• #2: Richness of astrophysical science opportunities• #3: Technology development needed• #4: Life-cycle costs• #5: Risk of cost, technology, schedule, on-orbit failures• #6: Reliability and robustness

Ringberg, 5-Sept-2003 Imaging with Darwin Page 13

Astrophysical imaging with Darwin

1. Imaging considerations2. Science

Röttgering et al. 2003, Heidelberg conference

Ringberg, 5-Sept-2003 Imaging with Darwin Page 14

Imaging performance at 10 micron

Sensitivity (Takajima and Matshura, 2001)• Limited by shot noise from the zodiacal background.• Similar to JWST

– Point source sensitivity• 1 hour, s/n=5: 2.5 microJy

– Image sensitivity • S_integrate/noise > 50 within FOV• > 2.5 microJy for a 100 hour

Resolution– Baselines up to 500 meter– 200 m baseline: 10 mas

• JWST 350 mas

Ringberg, 5-Sept-2003 Imaging with Darwin Page 15

Imaging considerations

PSF of an individual telescope: 1.4 arcsec– = maximum FOV for pupil combination

Mapsize (200 m baseline/telescope diameter) <~ 100 * 100 independent pixels

Complexity– per configuration maximum 6*5/2 = 15 uv points– number of uv-points >>~ number of image

parameters– for a complex map of 100 * 100 independent pixels:

• >>~ 666 configurations

Ringberg, 5-Sept-2003 Imaging with Darwin Page 16

Baseline dynamics

Fastest reconfiguration cycle takes about 16 hours

Snapshots will be taken “on the fly”

Basic reconfiguration approach

a single expansion up to baselines of 500 m and

contraction coupled to a 60o rotation

bang-bang thrust profile both radially and tangentially

<dB/dt> = 1.5 cm/s @ 1 mN

16

d’Arcio et al. 2001

Ringberg, 5-Sept-2003 Imaging with Darwin Page 17

UV coverageHexagonal array -> 9 independent visibilities per snapshot 600 snapshots, ~ 5400 uv points/reconfiguration cycle

-> Filling the UV plane is ’’easy’’ Ground based telescopes are ``fixed’’ (radio) Baseline/apertureis huge

17

d’Arcio et al. 2001

Ringberg, 5-Sept-2003 Imaging with Darwin Page 18

Issue: Cophasing How to phase-up the array not using the target?

– Essential to• integrate longer than the coherence time of the interferometer

(~10 sec)• Measure complex visibilities (Amplitude and phase) needed for

imaging– Off-axis bright stars (there are enough!)

• Similar to PRIMA instrument for the VLTI (Quirrenbach, this meeting)

• Multiplexing in wavelengths has the advantage that science and reference beams travel along common path (Alcatel)

– Implementation1. Modification to the nulling beamcombiner (Alcatel)2. Separate imaging beamcombiner

How to get a large Field of View?– Mosaicing

• Expensive in time– homothetic mapping

• Relative complex• Pupil matching in

magnification and orientation before image plane combining

• Implementation

– Pupil matching/zooming optics at central beamcombiner

– Pupil matching/zooming at telescopes

(see d’Arcio and le Poole, 2003)

positioning stages

4kx4k detector

imaging telescope

Zoom optics(5-50x)

Light f rom telescope

Afocal zoomoptics (5-50x)

Lo

(fi xed)

Light f rom telescope

conventionalpupil mapping

variable magnification

positioning stages

4kx4k detector

imaging telescope

Zoom optics(5-50x)

Light f rom telescope

Afocal zoomoptics (5-50x)

Lo

(fi xed)

Light f rom telescope

conventionalpupil mapping

variable magnification

Physical processes observable at 6-20 micron– Molecules: Rotational and

vibrational lines • Temperatures, densities,

kinematics, Chemistry

– Ions: Forbidden fine-structure lines• Temperatures, densities,

kinematics, abundance's

– Dust: PAH features, continuum shape• Composition, temperature

– Late type stars: continuum (high z)• Spatial scales

ISO observationsStarburst galaxy

Circinus

Ringberg, 5-Sept-2003 Imaging with Darwin Page 21

Appropriate sensitivity and angular resolution ?

Star and planet formationAGN

Distant galaxies

Star and Planet formation

Sketch of scenario maybe in place (Shu et al. 87)

Vast range of conditions and relevant timescales– densities 10^4 - 10^13

/cm^3– temperatures 10 - 10,000 K – month - 10^6 years

Issues– density, temperature and

dynamical structure of disks?

– At what stage and when do planets form?

Compendium of Monnier and Millan-Gabet of K-band sizes of YSOs

Disk models of D’Alessio, Merin

An unphysical, unrealistic extrapolation-> fainter YSO are small (10-100 mas ?)

Log

Rad

ius[

mas

]

4

2

0-2 -1 0

Log flux @ 14 micron [Jy]

ISOCAM survey of your starclusters at 6.6 and 14.3 micron (Eiroa et al)

Darwin

MIDI

Ringberg, 5-Sept-2003 Imaging with Darwin Page 25

Active galactic Nuclei Zoo: Seyfert, Starburst

quasars ... – unification: orientation,

time-evolution, mass, spin

1000 times more AGN at z=2 than z=0

Every galaxy has a central massive Blackhole (?)

Issues:– Physics? When and how

do BH form?– Relation to Galaxy

formation?

Ringberg, 5-Sept-2003 Imaging with Darwin Page 26

AGN may contain dusty tori– can obscure the central QSO– feeds the massive Black Hole

Radiative transfer model of a dusty torus – size scales with QSO

luminosity– SED from = 1 - 300 m– morphologies at = 10 m

Models of Tori of Granato et al.

Adapted to NGC 1068, Heijligers etal.

Darwin observations of Tori D = 300 times the sublimation radius

NGC1068:

– Bight, low luminosity nearby AGN

• ~10 Jy: prime target for MIDI/VLTI in 2003

• 1.7 1031 erg/s/Hz at = 10 m

– (prime target for MIDI/VLTI in 2003)

Weak AGN observable up to z = 1 - 2

Stronger AGN up to z = 10NGC1068

0.01 0.1 1 10 redshift

1’’

0.1’’

0.01’’

50 Jy

5 JyL (

10

m [

1030

erg

/s/H

z]

Distant Galaxies When and how do galaxies form?

– Star formation history, galaxies shapes– Relation to black hole formation

8-10 meter telescopes: a few thousand with 3<z<6 and still counting– Hardly morphological information

Darwin: morphologies of the older stellar component– observe 2 micron rest == 10 micron for z=4

Semi-analytical models of galaxy formation as guidance – input: evolution of cold-dark matter halos, prescriptions for cooling, star

formation and feedback, dust…– output: large samples of mock galaxies and their properties (SF, mass,

type)

FIRES survey IsaacVLT

: 2.5^2 arcmin 96 h in J, H, K

HDFS limit in K = 24.4

mag Image HST

I+H+K Franx, Labbe,

Forster,schreiber, Rix, Rudnick, Röttgering, etal.

Ringberg, 5-Sept-2003 Imaging with Darwin Page 30

SED fittingwith galaxy templates

•Photometric redshift•Estimate 10 micron flux density

Rudnick, Labbe et al.

Ringberg, 5-Sept-2003 Imaging with Darwin Page 31

JWST resolutionAt 10 micron(0.35 arcsec)

Ringberg, 5-Sept-2003 Imaging with Darwin Page 32

100 hour, S_int/noise=50

F (

10

m )

J

y

(photometric) redshift

100 hourPointsource S/N=5

Ringberg, 5-Sept-2003 Imaging with Darwin Page 33

Conclusion

Darwin will be a powerful instrument for – Finding and characterizing exo-Earth– Astrophysical studies

Sensitivity is similar to JWST– Cophasing is an important issue

Size scales, AGN, YSOs, distant galaxies are appropriate– Case for larger fields

2025Terrestrial planet imager?20 8-m telescopes