High-Contrast Imaging and the Direct Detection of Exoplanets Sandrine Thomas, Ruslan Belikov, and many collaborators Image of alpha cen

Is there another Earth out there? The question that we are trying to answer is nothing less than is there another Earth out there? Is there another Earth out there? Is there life on this over Earth? 2

Requirements for habitability not habitable (too large, hydrogen gas does not escape) habitable not habitable (too small to keep oxygen and water) Planet size: ~ 0.5 – 2 Earth size Temperature: 0-100 C Biomarkers: water and oxygen Credit: Petigura/UC Berkeley, Howard/UH-Manoa, Marcy/UC Berkeley H2O (Water) O2 (Oxygen) (Schematic representation only)

Spectroscopy: detecting biomarkers Ref: R. Hanel, GSFC O2 Iron oxides CO2 H2O EARTH-CIRRUS VENUS X 0.60 MARS EARTH-OCEAN H2O ice O3 Use either this slide or previous, depending on expected technical level of audience Detecting atmospheric oxygen and water likely indicates life (because very few non-biological processes can sustain an oxygen atmosphere)

Why Direct Planet Detection? How do planetary systems form and evolve? Which stars have planetary systems? What mechanism(s) create Jovian planets (accretion, migration, disk instability) How does dynamical evolution redistribute these planets? Physics of planets Planetary atmospheres Unexplored regime of 120 K < T < 600 K Domain of NH3 & H2O & clouds Just mention here about the different techniques: radial velocity, transits, microlensing, wabbling

861 confirmed planets (677 planetary systems) 2740 planet candidates (from Kepler mission) (Wobble method #2: Astrometry) Wobble method #1: Radial Velocity Astrometry: from 10pc away Direct detection

Direct Imaging Main Engineering Requirements Contrast ~1010 for Earth-like planets ~107 for young hot planets and disks Inner Working Angle The smaller the better! Typically 1-3 l/D required on missions

Beyond Kepler: Direct imaging missions 2010 2020 2030 Kepler Exo-C/S or AFTA (~1.4m / 2.4m, $1B / $2B+) New Worlds Telescope ($4-8m, $4B+) Small sats (0.25-0.7m, ~$5 – 200M) Earth-size Habitable zone Spectroscopy Two stars: aCen Earth-size Habitable zone No spectroscopy (biomarkers) Not nearby systems Another Earth? Earth-size Habitable zone Spectroscopy ~6-20 stars Earth-size Habitable zone Spectroscopy ~100s of Earths Simulation of an exo-Earth around aCen with a $1B mission (1.5m telescope) All these missions also do ground-breaking science on non-habitable planets

Ground based Instruments: Exoplanet direct imaging instruments GPI SPHERE Macintosh et al, 14 Beuzit et al, 14 P1640 SCExAO Hinkley et al, 08 Guyon et al, 10

High Contrast Imaging Like searching for a firefly next to a lighthouse in San Francisco from Boston => Very faint and small in comparison Upper Scorpius Lafreniere et al 2008 Beta Pictoris b Lagrange et al 2010 HR 8799 Marois et al 2008 Fomalhaut b Kalas et al 2008

Limitations Diffraction from the parent star Depends on telescope diameter and wavelength: FWHM=lambda/d Passives and active aberrations in the system Constant: static aberrations of the system 1Hz: drifts due to temperature, flexure, mechanical instability. 1KHz: from the ground: atmosphere Amplitude errors and Talbot effect that turns amplitude errors into phase errors 5 nm RMS non-common path WFE 1 nm RMS non-common path WFE

Solutions Diffraction: Coronagraphs Aberrations: Active and adaptive optics Wavefront sensor : Shack-Hartmann, curvature sensors Deformable mirror Control software DM system Coronagraph Science camera WFS Starlight rejected by coronagraph feedback Diagram of a direct imaging instrument from telescope

The Lyot Coronograph Sivaramakrishna, 2001

Principle of Pupil Apodization-type coronagraphs Image plane starlight Telescope pupil Resulting image

Apodizer manufacture: halftone technique using black chrome microdots The apodizer transmission is obtained by randomly distributing 2μm square dots over the glass. by JenOptiks.

The Apodized Lyot coronagraph Aime et Al. 2001 Pupil plane Focal plane Lyot stop Image

Challenge #2: optical aberrations Image plane starlight Telescope pupil Resulting image

Adaptive Optics System for Turbulence Correction Distorted wavefront Corrected wavefront Imager DM WFS Feedback

Image sharpening for slow changing aberrations Distorted wavefront Corrected wavefront Imager DM WFS Feedback

Speckle Nulling We need to know how the DM phase maps to the image location and intensity To calibrate location drive the DM at the highest spatial frequency To calibrate the intensity measure some of the spatial frequency and interpolate the rest

Lab Performance after Speckle Nulling Before static wavefront correction Example: Gemini Planet Imager, Macintosh B. Savransky, Thomas et al., 2012 After static wavefront correction

Image Sharpening: Ex: Electric Field Conjugation (Give’on et al Image Sharpening: Ex: Electric Field Conjugation (Give’on et al. 2008, Thomas et al. 2010) Calculate the electric field (Ef) in the focal plane Find the DM shape such that its effect in the plane of interest negates the electric field present in this plane Possible to correct both phase and amplitude G A + Ef = 0 Actuator commands Electric field in the focal plane Reconstruction matrix Improves the contrast by a factor 3, reaching 4.10^8. ONLY A FACTOR OF 3 ?? LOOKS BIGGER IN YOUR PLOT

Beta Pictoris b with GPI Combined 30-min GPI image of Beta Pictoris. The spectral data have been median-collapsed into a synthetic broadband 1.5–1.8 μm channel. The image has been PSF subtracted using angular and spectral differential techniques. Beta Pictoris b is detected at a signal-to-noise of ∼ 100. Gemini Planet Imager Gemini NICI (Boccaleti et al) 1980 seconds 3952 seconds

Contrast vs. angular separation at H (1 Contrast vs. angular separation at H (1.6 μm) for a PSF-subtracted 30- min GPI exposure. Contrast is shown for PSF subtraction based on either a flat spectrum similar to a L dwarf or a methane-dominated spectrum (which allows more effective multiwavelength PSF subtraction.) For com- parison, a 45-min 2.1-μm Keck sequence is also shown. (Other high-contrast AO imaging setups such as Subaru HiCIAO, Gemini NICI, and VLT NACO have similar performance to that of Keck.) 0.4 arcsec = 10 lambda/d

The Ames Coronagraph Experiment (ACE)

The Ames Coronagraph Experiment (ACE) Laboratory Cooling system Wavefront Control: SN, EFC, LOWFS Not as mature => labs MEMS from Boston Micromachine Corporation On stepper and piezo stage

People and organizations partnering with ACE UofA Olivier Guyon Glenn Schneider Julien Lozi L3 Tinsley Jay Daniel Asfaw Bekele Lee Dettmann Bridget Peters Titus Roff Clay Sylvester NASA ARC Ruslan Belikov Thomas Greene Eugene Pluzhnik Sandrine Thomas Fred Witteborn Dana Lynch Paul Davis Eduardo Bendek Kevin Newman Princeton Jeremy Kasdin Bob Vanderbei David Spergel Alexis Carlotti STScI Laurent Pueyo JPL Brian Kern Andy Kuhnert John Trauger Wes Traub John Krist Marie Levine Stuart Shaklan K. Balasubramanian Now, how hard is this problem? Lockheed Martin Domenick Tenerelli Rick Kendrick Alan Duncan Wes Irwin Troy Hix

Contrast Achieved in air Median contrast of 4.06x10-7 between 1.2 and 2 l/d Simultaneously with 8.51x10-8 between 2 and 4 l/d Mask Inner Working Angle (IWA)=1.12 λ/d Average over an hour Speckle nulling, round mask

Stability => MEMS stable and reproducible correction Contrast remains under 10-6 in the inner region and under 10-7 in the outer region for over an hour, (represented as 1500 images). Same performance was obtained in three independent tests. Reapplying a MEMS map 1-day after the correction without changing the calibration keeps the results within 10%. => MEMS stable and reproducible correction

Vacuum tests Monochromatic result We achieved the same contrast as in air, but on a bigger zone: 1.2-12 λ/D instead of 1.2-4 λ/D We are close to the milestone in polychromatic (1.8e-7 at 5%, 3.2e-7 at 10% between 2 and 12 λ/D) Limited now by a manufacturing default of the focal plane mask (OD3 instead of OD5) A new mask will be manufacture for the last vacuum test, and help us achieve the milestone. Focal plane mask Polychromatic result with bandwidth of 5, 10, 20 and 40 %.

Status/Next steps Milestone #1 demonstration requirement met (exceeded) in air 1.8e-7, 1.2-2.0 l/D simultaneously with 6.5e-8, 2.0-4.0 l/D First vacuum tests started in January and are ongoing Milestone #2 (10% broadband light) to be pursued this year EXCEDE aggressive IWA technology development can be beneficially carried over to future larger scale coronagraphic missions with focus on exoplanets Most of the credit goes to: J. Lozi: LOWFS, DAQ, system calibration S. Thomas: Optical design, EFC wavefront control E. Pluzhnik: Optical assembly and alignment, SN E. Bendek: CAD design and layout T. Hix and the rest of the LM team: Vacuum hardware preparation and chamber operations

Adaptation to halo beam issues Will need: Contrast needed? 1e-7? How far from the core do you need to observe? Which wavelength? Dynamic range of the detector Potential issues: Static and dynamic aberrations (Stability?) The source is resolved. The source size varies with time (how much?) Noise Mie scattering? Figure 1. Lower-dynamic-range transverse beam profile measured in Jefferson Lab’s free-electron laser injector. The blue halo intensity is about 300 (arbitrary units) less than that of the green core. Image courtesy of Pavel Evtushenko Figure 1. Lower-dynamic-range transverse beam profile measured in Jefferson Lab’s free-electron laser injector. The blue halo intensity is about 300 (arbitrary units) less than that of the green core. Image courtesy of Pavel Evtushenko Christine Herman, 2001

Need contrast of about 1e-4/1e-5? Christine Herman, 2001

Conclusion Similar issues are seen by your groups and coronagraphy for astronomy. Need a translation between astronomy and particle physics

GPI Observations of HR 4796A 0.5” E Individual 60 s images One linear polarization shown. Waveplate rotates 0, 22.5, 45... & the parallactic angle changes Combined 12 minutes Total intensity Combined 12 minutes Linear polarized intensity Slide by Marshall Perrin