Calibration methods for high contrast imaging applications

1. Calibration methods for high contrast imaging applications Joseph J. Green, James K. Wallace and Michael Shao Jet Propulsion Laboratory California Institute of Technology Gemini ExAOC Conceptual Design Mid-Term Review HIA, Victoria, BC October 21-22, 2004

2. A menu of topics… • Static Speckle Calibration Experimental Work Status Quasi-Static Status • Quasi-Static and Dynamic Speckle Calibration Simulation Status • Shaped Pupil Selection and Its Impact on Calibration • Chromaticity of Static Speckles and Speckle Deconvolution • Chromaticity Issues with Static Speckle Control Gemini ExAOC Mid-Term Meeding, HIA

3. Experimental Work Status Static Speckle Calibration Gemini ExAOC Mid-Term Meeding, HIA

4. WFS VIA Coronagraphic Pupil Imaging Beam from telescope • Lyot coronagraphs that employ band-limited occulters can perfectly eliminate diffraction. • Any amplitude of phase variation across the entrance pupil to the coronagraph creates speckles • In the focal plane, the speckles scale both spatially (radially) and in intensity with the optical passband • In the pupil, however they do not radially streak making broadband WFS feasible. • To interpret the pupil intensity into wavefront error, we need to inject diversity Occulting Spot at Focal Plane Lyot Stop at Downstream Pupil HCIT Science Focal Plane Gemini ExAOC Mid-Term Meeding, HIA

5. WFS Using Occulter Focus-Diversity Lyot Pupil • Pupil images through the coronagraph are taken while the occulter is pistoned through focus • With 3 focus-diverse pupil images, the wavefront can be estimated. • The figures below are actual measurements from the TPF high contrast imaging testbed • 800nm, 40nm passband Actuator pokes as filtered by the coronagraph Actuator pokes observed using focus diversity -500µm 0µm +500µm WF Estimate NOTE: Different poke pattern Gemini ExAOC Mid-Term Meeding, HIA

6. Simulation Work Status Quasi-Static and Dynamic Speckle Calibration Gemini ExAOC Mid-Term Meeding, HIA

7. Multilayer Atmosphere Model • Simulates the temporal variations induced by the atmosphere capturing both the wavefront and the scintillation • Based on Cerro-Pachon model published on Gemini’s website Gemini ExAOC Mid-Term Meeding, HIA

8. Simultaneous Calibration of Science Imagery • Mach-Zender I/F enables temporally sampled measurements of the coronagraphic wavefront while integrating on the science target • Provides speckle model used for PSF subtraction • Provides updates to static calibration • MZ error modeling includes: • Coronagraph plug-in capacity • Realistic aberrations • Chromaticity effects • Spatial filter effects • Detector effect • Noise effects (coming soon) Gemini ExAOC Mid-Term Meeding, HIA

9. Contrast Improvement by MZ Cal Gemini ExAOC Mid-Term Meeding, HIA

10. Impact on Calibration Efforts Shaped Pupil Coronagraph Selection Gemini ExAOC Mid-Term Meeding, HIA

11. Ideal Shaped Pupil Performance l = 1.6µm Gemini ExAOC Mid-Term Meeding, HIA

12. TPF Shaped Pupil Designs • With shaped pupil coronagraphs, one may trade-off discovery space for mask throughput • Generally the IWA > 4 l/D unless the OWA is greatly constrained 18% 25% 27% Gemini ExAOC Mid-Term Meeding, HIA

13. Calibration Effort Impacts • Static Calibration (Day Time) • Occulter focus-diversity approach will not work as implemented • Even in Lyot coronagraph configuration it is problematic • Works best with band-limited occuter (not hard-edged hole) • Conventional focus diversity may work • Demonstrated very high accuracy on HCIT • l/10000 Wavefront Sensing repeatability - limited by photon noise • l /5000 Wavefront Control performance - limited by DM driver noise • How do get the diversity? • Translate FPA, Insert Lenses, Translate Calibration Source, Use DM • Quasi-Static and Dynamic Speckle (During Science) • No obvious problems with Mach-Zender other than it has to work in the presence of stronger uncalibrated static errors • Simulation work is ongoing Gemini ExAOC Mid-Term Meeding, HIA

14. TPF Simulations Chromaticity of Static Speckles and Speckle Deconvolution Gemini ExAOC Mid-Term Meeding, HIA

15. OAP-1 DM Surrogate OAP-2 FM-1 Total Amplitude 3.1x10-4% 0.23% 1.28% 0.18% 1.31% Phase 3.6nm 4.9nm 6.2nm 1.2nm 10.6nm Complex WFE from HCIT Front-End Optics • HCIT provides an example of how out-of-plane optics translates into complex exit-pupil variations • The translations of surface errors in the complex exit-pupil variations is a function of both wavelength and optic location within the propagation • It may be very hard decode errors based on observation of speckles at multiple wavelengths Gemini ExAOC Mid-Term Meeding, HIA

16. Compensations for out-of-plane optical errors leads to chromatics limitations to contrast We may see to create a fixed corrector conjugate to Gemini’s tertiary Chromatic Limitation of Pupil Correctors Gemini ExAOC Mid-Term Meeding, HIA

17. TPF Simulations Chromaticity Issues with Static Speckle Control Gemini ExAOC Mid-Term Meeding, HIA

18. Deformable Mirrors Controlling Static Speckles • Michelson interferometer wavefront control configuration • Enables control of amplitude and phase speckles over the entire field of view. (Full-Dark-Hole) 8nm rms 0.2% rms Science Focal Plane Gemini ExAOC Mid-Term Meeding, HIA

19. Case 1a: OPD Errors at a Pupil R=Infinity • With only OPD errors at a pupil a DM at a pupil can perfectly correct them within its limited controllable spatial frequencies • This pupil based correction does not take into account speckles resulting from high-order interactions of the high-frequency (uncontrollable) speckles R=20 R=4 Gemini ExAOC Mid-Term Meeding, HIA

20. Case 1b: OPD Errors at a Pupil R=Infinity • Using a focal-plane speckle driven control to optimize the Michelson WFC system, these speckles can be fully controlled at a single wavelength • The correction, however, is accurate over a finite bandwidth. R=20 R=4 Gemini ExAOC Mid-Term Meeding, HIA

21. Case 2: OPD and Transmission Errors R=Infinity • Introducing transmission errors in the reflectivity of the optics results in additional chromaticity of the control • Even with transmission errors only at a pupil, these amplitude speckles do not scale with wavelength and limit the bandwidth of the correction R=20 R=4 Gemini ExAOC Mid-Term Meeding, HIA

22. Case 3: OPD, Transmission and Occulter Errors R=Infinity • With the addition of occulter imperfections, the chromaticity of the compensation worsens • The occulter errors include: • Maximum Optical Density = 8 • Occulter substrate has OPD errors linear with OD (0.1 rad/OD) R=20 R=4 Gemini ExAOC Mid-Term Meeding, HIA

23. Another View of Chromaticity 3-Color Composition of Broadband contrast Broadband contrast Case 2: R=4 Case 3: R=4 Gemini ExAOC Mid-Term Meeding, HIA

24. TPF HCIT Experiments Demonstration of Extreme Wavefront Sensing & Control Using Focus Diversity Gemini ExAOC Mid-Term Meeding, HIA

25. WFS Repeatability Experiment (~July 2003) HCIT • Collected WFS datasets repeatedly for about 40 hours • 26 runs total (24 complete datasets) • Vacuum tank pumped down to 10 mTorr • Pump was on during the experiment • Temperature remained stable to better than 100 mC Surface Gauge Gemini ExAOC Mid-Term Meeding, HIA

26. Implications of WFS Repeatability for High Contrast Imaging • If we consider the WFS repeatability to be the limiting factor for correcting the wavefront • Its average energy level at a radial spatial frequency would yield a contrast • After 3 cycles/pupil, the achievable level of contrast is between 10-9 and 10-10 10-9 Contrast 10-10 Contrast TPF Requirements 3 l/D Gemini ExAOC Mid-Term Meeding, HIA

27. Calibration of 32x32 DM for HCIT WFC To enable the maximum correction of the HCIT WFE, the DM actuator mapping to the WFS coordinate system must be accurately calibrated We applied dilute poke patterns and observed the change in the wavefront using focus diverse phase-retrieval Using several such patterns, we calibrated the mapping between the HCIT WFS and the DM coordinate systems and derived the control algorithm Change in HCIT Wavefront as Measured by Phase-Retrieval Differential Command to DM

28. HCIT Wavefront Sensing and Control It. 1 It 8 • After the DM was calibrated we conducted an iterative sequence of WFS and WFC. • With each iteration, improved estimates of the true focus and F/# for HCIT were used to improve the defocus level selection and defocus symmetry in the subsequent focus-diverse data collection • In this experiment, we have chosen to comply with the conservative inter-actuator position differential constraints. Complying with this operational rule prevents the full correction of 1 actuator within the clear aperture. Sequence of Post- Control Wavefront Estimates Sequence of WFE Power Spectral Densities

29. Increasing WFS/C Iterations Accumulated WFE within ensquared region (rms waves) WFE PSD Extent of PSD integrated region as a fraction of controllable limit Analysis of WFS/C Performance • The set of curves depicts the state of the WFE on HCIT over the iterations of WFS/C. • Each curve represents the accumulated WFE as a function of the size of the ensquared region within the PSD of the wavefront error. • The WFS/C process achieved a correction of HCIT WFE to about the l/5000 level within the controllable passband • Beyond this Nyquist limit for the DM, the WFE is a combination of the uncontrollable WFE within HCIT and the DM print-through.