TY - GEN
T1 - Metasurface synthesis through multi-objective optimization aided inverse-design
AU - Campbell, Sawyer D.
AU - Whiting, Eric B.
AU - Werner, Douglas H.
AU - Werner, Pingjuan L.
N1 - Publisher Copyright:
© 2019 SPIE.
PY - 2019
Y1 - 2019
N2 - Advances in design and nanofabrication are enabling the ability to realize nanophotonic structures that achieve optical functionalities not possible with conventional devices. Some of the most interesting structures being developed are based on metasurfaces comprised of sub-wavelength unit cells intelligently patterned to locally manipulate an electromagnetic wavefront in a desired way. However, designing a metasurface to achieve a specific set of desired functionalities is extremely challenging especially in the presence of coupling between unit cells, which can affect system performance and lead to unexpected optical aberrations. To overcome this challenge, designers typically employ only canonical structures (e.g., loaded-dipoles, v-antennas, split-ring resonators) to synthesize a metasurface to achieve their desired functionality. However, metadevices based on these canonical structures do not always achieve optimal performance especially when broadband and/or wide-field-of-view functionality is desired. Additionally, different material combinations as well as fabrication effects and tolerances can make the optimal unit cell topology selection difficult. Therefore, the ability to generate a diverse set of unintuitive metasurface unit cells and evaluate their electromagnetic behaviors for achieving a specific functionality is highly desirable. Moreover, doing so in an efficient and intelligent manner is extremely important in order to maintain a tractable inverse design process. To this end, we employ a custom nanoantenna unit cell topology generation routine in conjunction with robust state-of-the-art multi-objective and surrogate-assisted optimization algorithms to find optimal designs that achieve an arbitrary number of user-specified performance criteria. Finally, when paired with advanced full-wave forward solvers these design and optimization algorithms enable an efficient inverse design toolkit for high performance metasurface synthesis.
AB - Advances in design and nanofabrication are enabling the ability to realize nanophotonic structures that achieve optical functionalities not possible with conventional devices. Some of the most interesting structures being developed are based on metasurfaces comprised of sub-wavelength unit cells intelligently patterned to locally manipulate an electromagnetic wavefront in a desired way. However, designing a metasurface to achieve a specific set of desired functionalities is extremely challenging especially in the presence of coupling between unit cells, which can affect system performance and lead to unexpected optical aberrations. To overcome this challenge, designers typically employ only canonical structures (e.g., loaded-dipoles, v-antennas, split-ring resonators) to synthesize a metasurface to achieve their desired functionality. However, metadevices based on these canonical structures do not always achieve optimal performance especially when broadband and/or wide-field-of-view functionality is desired. Additionally, different material combinations as well as fabrication effects and tolerances can make the optimal unit cell topology selection difficult. Therefore, the ability to generate a diverse set of unintuitive metasurface unit cells and evaluate their electromagnetic behaviors for achieving a specific functionality is highly desirable. Moreover, doing so in an efficient and intelligent manner is extremely important in order to maintain a tractable inverse design process. To this end, we employ a custom nanoantenna unit cell topology generation routine in conjunction with robust state-of-the-art multi-objective and surrogate-assisted optimization algorithms to find optimal designs that achieve an arbitrary number of user-specified performance criteria. Finally, when paired with advanced full-wave forward solvers these design and optimization algorithms enable an efficient inverse design toolkit for high performance metasurface synthesis.
UR - http://www.scopus.com/inward/record.url?scp=85076485433&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85076485433&partnerID=8YFLogxK
U2 - 10.1117/12.2529823
DO - 10.1117/12.2529823
M3 - Conference contribution
AN - SCOPUS:85076485433
T3 - Proceedings of SPIE - The International Society for Optical Engineering
BT - Metamaterials, Metadevices, and Metasystems 2019
A2 - Engheta, Nader
A2 - Noginov, Mikhail A.
A2 - Zheludev, Nikolay I.
A2 - Zheludev, Nikolay I.
PB - SPIE
T2 - Metamaterials, Metadevices, and Metasystems 2019
Y2 - 11 August 2019 through 15 August 2019
ER -