The discovery of new materials is being rapidly accelerated through the use of computational methods that can screen the constituent elements for a desired application. While simple materials systems and structures can be easily predicted using these methods, more complex systems can prove challenging due to the increased number of potential configurations. Evolutionary algorithms, a sort of “survival of the fittest” methodology that sequentially generates and weeds out potential solutions to converge on an optimized result, offer the possibility to accurately predict complex systems while reducing computation time. These design methods have been used in the past to predict metamaterial structures that can operate at terahertz frequencies but encounter problems at optical frequencies that require a finer computational grid to account for losses, leading to prohibitively longer computation times.
Teri W. Odom and colleagues at Northwestern University now report a bottom-up strategy that uses a custom-built evolutionary design algorithm to predict a new class of optical materials. These new “lattice opto-materials” are detailed in the November 7, 2014, issue of Nano Letters (DOI: 10.1021/nl5040573; p. 7195). Lattice opto-materials can be used to concentrate light into discrete focal points in the optical far-field and produce arbitrary three-dimensional (3D) light profiles. They function based on the discretization of a plasmonic film into a two-dimensional (2D) subwavelength grid of holes with diameters in the range of 100–200 nm. The researchers were able to design lattice opto-materials for a specific design criterion with vastly reduced computation time.
Lattice opto-materials can manipulate visible light to produce a variety of three-dimensional profiles from two-dimensional grids of nano-sized holes milled into a substrate. Reproduced with permission from Nano Lett. 14 (12) (2014), DOI: 10.1021/nl5040573; p. 7195. © 2014 American Chemical Society.
A number of optical systems were simulated based on this new design algorithm before being experimentally validated in the laboratory. The researchers used a focused ion beam to mill a 2D grid of holes and they meas-ured 3D optical profiles using confocal scanning optical microscopy. They found that experiments matched simulations accurately in a number of configurations. The researchers were able to generate a variety of 3D far-field light profiles with focal points varying in x, y, and z positions. These lattice opto-materials grant exceptional control over visible light, surpassing the capabilities of current microlenses, metalenses, and plasmonic lenses. With the inclusion of polarization-sensitive hole shapes, the possibility of dynamic control without physical alteration of the substrate is also possible. This new class of materials holds incredible potential for current and future applications, paving the way for enhanced flat optics for new imaging modalities.
