A Self-Stabilizing Metasurface Laser Sail To Explore The Stars Joel Siegel University of Wisconsin Madison Physics Department 5/22/19 1
Laser Propelled Spacecraft Laser Sail High-Power Laser ~10 g, ~100 GW 4x4 m 2 Goal is to travel to Traveling at Alpha Centauri 1/5 th speed (~5 light years away) of light But how do we keep the sail in the beam? Starshot Breakthrough Initiative 2
How fast is 1/5 th the speed of light? β’ Took the first close up pictures of Pluto in 2015 β’ One of the fastest man made objects 747 SR-71 Blackbird New Horizons (550mph) (2,700 mph) (36,000 mph) 65xNew Horizons New Horizons (36,000mph) (177,000 mph) (2,365,000 mph) 65xNew Horizons 65xNew Horizons Laser Sail 3 (11,570,000 mph / (152,000,000 mph / Gif from Clay Bavor (2,365,000mph) 0.017c) 0.2c)
Optical Forces πΈ π πΈ π Force is determined by the reflected/refracted light π πΊ πΊ If we control how the light reflects/refracts, we can Force Force control the optical forces Laser Sail 4
Metasurface Based Laser Sail β’ Thin, lightweight structure with subwavelength scattering elements β’ Controls the phase and magnitude of reflected/refracted light Reflective Metasurface Focusing Lens Optical Vortex Beam Creation n ~ ΞΌ m ~cm Prism Beam Steering Metasurface Beam Steering Y. Yang, et. al. Nanoletters 14, 1394 (2014). Arbitrary wave-fronts can be generated with a metasurface 5 D. Fattal, et. al., Nature Photonics 4, 466 (2010).
Metasurface Example 1 Intensity Gaussian Beam 0 Forces Example Metasurface 6
Metasurface Motion Gaussian Beam Motion can be described by: Dynamic Force π π 2 π Metasurface flies ππ’ 2 = π· 1 π + π· 2 π Coefficients Moves back, but away also rotates π½ π 2 π Each metasurface/beam ππ’ 2 = π· 3 π + π· 4 π Offset the combination has Metasurface different coefficients Rotation Offset How can we control these coefficients to make a Example Metasurface metasurface that is stable? 7
Designing a Stable Sail Noise Applied to Beam Find a set of stable Create idealized sail to Simulate that sail using parameters generate those coefficients realistic components Sail to simulate π π 2 π with realistic ππ’ 2 = π« π π + π« π π components Failure Rate π½ π 2 π ππ’ 2 = π« π π + π« π π 1 Million CPU hours β’ β’ Requires many (>1 Million) computations with slight parameter variations Requires 1 large computation to produce this plot β’ β’ Human intuition for the sail designs Sail design chose from previous stage Over 1 million output files! β’ β’ Computation Requirements Computation Requirements Needed HPC Had to write a shell script to β’ β’ 1 CPU 80 CPUs to run β which βcatβ them all together in β’ β’ introduced 1MB of RAM 500 GB of RAM pieces me to HTC β’ β’ 3 GB of Disk 5 GB of Disk 8 β’ β’ Output of single file is ~10 bytes (0/1 for success/failure and an ID) Output is ~5 GB
Idealized Metasurface to Generate Stable Coefficients Two Offset Gaussians Inverted Cat Eye (ICE) Metasurface Partially Highly Reflective, Transmissive, Parabolic Normal Metasurface 9
Full-Wave Simulation Steering the Beam Reflected Beam front Simulated ICE Metasurface Resonators Partially Transmissive, Transmitted Beam front Parabolic Highly Reflective, Normal 504 ΞΌ m, 420 Resonators 10
Local Optical Forces on Metasurface Forces calculated at Overall good agreement each point on sail Full-Wave Sail is Stable! Full-Wave Full-Wave Full-Wave Full-Wave Ideal Ideal No Offset No Offset Offset Ideal Ideal Offset 11
Whatβs next? β’ Incorporate optimization techniques that can take advantage of throughput computing to more efficiently explore the parameter space β’ Generate a sail based on a set of dynamic force coefficients without human intuition β’ Improve the efficiency of our realistic Figure courtesy of Greg Holdman sails by using optimization based metastructures 3
Acknowledgments β’ Big Thank You to Christina Koch and Lauren Michael for helping me learn to use CHTC Brar Group Collaborators Mikhail A. Kats β UW Madison Sergey Menabde β KAIST Min Seok Jang β KAIST 13
14
Recommend
More recommend
