Optimization of Apodized Pupil Lyot Coronagraph for ELTs

Transcription

1 Optimization of Apodized Pupil Lyot Coronagraph for ELTs P. Martinez 1,2, A. Boccaletti 1, M. Kasper 2, P. Baudoz 1 & C. Cavarroc 1 1 Observatoire de Paris-Meudon / LESIA 2 European Southern Observatory 1

2 Context The Apodized Pupil Lyot Coronagraph (Aime et al. 2002, Soummer et al. 2003) Apodizer Solutions for obscured apertures (Soummer 2005) Lyot mask A A promising concept for ELTs Prototypes are being manufactured for several projects VLT/SPHERE, see poster by Boccaletti et al. Achievable contrasts strongly depend on telescope parameters An understanding of APLC behavior to these parameters is necessary 2

3 Goals Problematic: make a trade-off between performances & throughput Define optimal configuration in a general case Apodizer throughput does not evolve linearly with mask diameter Two apodizer regims exist (Soummer 2005) 1 st regim 2 nd regim 1 st regim 2 nd regim Identify critical telescope parameters Constraint: avoid signal to noise ratio estimation (right way but requires the telescope + instrument to be better defined) Intermediate solution optimization criterion (Boccaletti 2004): = Average contrast x Off-axis PSF attenuation 3

4 Central obscuration Each apodizer regim has an optimal configuration Maximum throughput APLCs: between these two regims 2 nd regim is clearly the more promising Optimal configurations: λ/d Optimal apodizer throughput: % Maximum throughput With our criterion the optimal apodizer/mask gets larger with the central obscuration while it is the opposite when only the throughput is considered 4

5 Secondary support Euro 50 like VLT like OWL like Assumptions: Central obstruction: 30 % (E-ELT) Optimal configuration for 30% : 4.7 λ/d Results: Number of spider arms does not influence too much the performances and the optimal configuration Performances degrade with spider width but optimal configuration is not influenced 5

6 Mirror segmentation Assumptions: Obstruction: 30% / ~ 750 hexagonal segments / no inter-segment gaps Amplitude segment defects Phase segment defects Reflectivity range: 0-5% (peak-to-valley) λ = 1.6µm Sensitive for configuration larger than 4 λ/d Segment reflectivity is not critical for the selection of optimal Apodizer/mask Both performances & optimal size are sensitive to phase aberrations Not relevant to optimize the APLC configuration 6

7 Chromatism Assumptions: Central obstruction: 30% No segmentation Central wavelength: λ 0 = 1.6 µm Results: Optimal configuration sensitive for R < 20 R [,, 20],, monochromatic optimization is sufficient. For smaller R, optimization with chromatism is needed For instance: R = 2 performances are increased by a factor 7 w.r.t monochromatic optimization 7

8 Application Application to futur Planet finder telescope designs: Design 1 Design 2 Diameter: 42 m (E-ELT) No segmentation Spider thickness: 50 cm R = 15 Design 1: APLC 3.5 λ/d Design 2: APLC 4.1 λ/d Optimized with throughputs Design 1: APLC 4.7 λ/d Design 2: APLC 4.3 λ/d Optimized with our criterion PSF PSF Coronagraphic PSF ~10 Coronagraphic PSF Optimized APLC allows the performance to be rather insensitive to telescope designs 8

9 Conclusion Sensitivity analysis was done for several important parameters (More details in Martinez et al. 2007, submitted to A&A) Optimal configurations have been evidenced Central obscuration is certainly the more critical parameter for the optimization of APLC Some parameters have small impact on the optimal configuration Manufacturing of selected prototype based on this study has started Tests with HOT, the High Order Testbench scheduled for the end of 2007 HOT : XAO ESO, including turbulence, Woofer/Tweeter, PWS, SWS 9