skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Surface modes at metallic an photonic crystal interfaces

Thesis/Dissertation ·
DOI:https://doi.org/10.2172/985309· OSTI ID:985309
 [1]
  1. Iowa State Univ., Ames, IA (United States)

A surface mode is an electromagnetic field distribution bounded at a surface. It decays exponentially with the distance from the surface on both sides of the surface and propagates at the surface. The surface mode exists at a metal-dielectric interface as surface plasmon (1) or at a photonic crystal surface terminated properly (34; 35; 36). Besides its prominent near-filed properties, it can connect structures at its propagation surface and results in far-field effects. Extraordinary transmission (EOT) and beaming are two examples and they are the subjects I am studying in this thesis. EOT means the transmission through holes in an opaque screen can be much larger than the geometrical optics limitation. Based on our everyday experience about shadows, the transmission equals the filling ratio of the holes in geometrical optics. The conventional diffraction theory also proved that the transmission through a subwavelength circular hole in an infinitely thin perfect electric conductor (PEC) film converges to zero when the hole's dimension is much smaller than the wavelength (40). Recently it is discovered that the transmission can be much larger than the the filling ratio of the holes at some special wavelengths (41). This cannot be explained by conventional theories, so it is called extraordinary transmission. It is generally believed that surface plasmons play an important role (43; 44) in the EOT through a periodic subwavelength hole array in a metallic film. The common theories in literatures are based on these arguments. The surface plasmons cannot be excited by incident plane waves directly because of momentum mismatch. The periodicity of the hole arrays will provide addition momentum. When the momentum-matching condition of surface plasmons is satisfied, the surface plasmons will be excited. Then these surface plasmons will collect the energy along the input surface and carry them to the holes. So the transmission can be bigger than the filling ratio. Based on this picture, they deduced naturally that when surface plasmons momentum-matching condition is satisfied, the surface plasmons are excited sufficiently and the transmission reaches its peak. I present a new theory from first principles to explain EOT through one-dimensional periodic subwavelength metallic slits in this thesis. This theory can also be extended to 2D hole arrays. I define the incident wavelengths that satisfy the momentum-matching condition as surface resonant wavelengths. I proved analytically that the transmission is actually zero at the surface resonant wavelengths. The correct logic is: When the momentum-matching condition is satisfied, the surface plasmons excited by each slit interfere constructively with each other, the total surface plasmons will go to infinity. But the law of nature forbids the infinity. The only solution is the surface plasmon excited by one slit is zero and all the energy is reflected. In my theory, the term corresponding to surface plasmons appear explicitly in the equations. So it confirms the importance of surface plasmons without any doubt. The theory divides the transmission process into two steps: energy collection process along the input surface and the propagation process in the slits. In the first process, the surface plasmons collect the energy along the input surface and carry them to the slits. This process happens efficiently at any wavelength other than the surface resonant wavelengths. So EOT can happen at almost any wavelength. After the energy enter the slits, the Fabry-Perot interference between the input and output surface decides how much energy is emitted from the slits. So the EOT wavelengths are decided by the Fabry-Perot resonances. I also use my theory to explain the data in literatures. The transmission spectra through 1D slits or 2D hole arrays in literatures agree with my theory very well. The new theory can explain a lot of experimental results published recently, such as the transmission through randomized hole arrays, the strong influence of the hole shape on the transmission peaks, and so on. Beaming is another far-field effect resulting from surface modes. Normally light coming from a subwavelength waveguide is diffracted to all angles. With the help of surface modes, we can confine the output field in a small angle interval. This phenomenon is called beaming (46). The principle of the beaming has been explained clearly in literatures (47). To achieve good beaming, a photonic crystal waveguide need a surface layer to support surface modes and a grating layer to coupling the evanescent surface modes into propagation modes. A metallic beaming structure is generally a subwavelength waveguide surrounded by periodic structures such as grooves or dielectric gratings (53; 54). The flat metal surface supports the surface mode, so additional surface layer is not necessary. The periodic structures work as the grating layer.

Research Organization:
Ames Lab., Ames, IA (United States)
Sponsoring Organization:
USDOE Office of Science (SC)
DOE Contract Number:
AC02-07CH11358
OSTI ID:
985309
Report Number(s):
IS-T 2664; TRN: US201016%%2177
Country of Publication:
United States
Language:
English