High transmission through subwavelength gratings inspires debate

The mechanism that allows high transmission through a type of metal film grating -- a phenomenon that should prove useful for displays and other electro-optic devices -- has become a source of contention. The majority of papers on the subject over the last two years have claimed that surface plasmons are responsible for allowing enhanced transmission: up to three orders of magnitude greater transmission than would be expected from conventional analysis. A significant proportion of these papers have come from scientists at the NEC Research Institute in Princeton, New Jersey. Last summer, another NEC researcher, Michael Treacy, proposed that surface plasmons were not involved and that diffraction alone could explain the effect. Now, in the latest paper on the subject, a group in Germany say they have unambiguously shown that surface plasmons are, in fact, responsible. The issue seems far from settled.

Figure 1. Arrays of nanoholes in optically thick metal films (this one is 200 nm thick with holes 150 nm across and a period of 900 nm) transmit much more light than would be expected by conventional analysis. However, scientists still disagree on which mechanism is responsible.

NEC researchers first reported the unusual transmission characteristics of metal films patterned with a periodic array of subwavelength holes in Nature in 1998. Later that year,¹ they went on to describe two mechanisms they thought were the cause. The minima seen in the pattern of transmitted light, they said, were caused by something called Wood's anomaly: an effect that occurs when light is diffracted at a tangent to the film. The maxima, on the other hand, were caused by surface plasmons: longitudinal waves of charge that are created at the interface between the metal and substrate. The dramatically high transmission occurred because plasmons created on either side of the substrate coupled together strongly, thus pulling the energy from one side to the other.

These theories have been expanded in other papers and by other researchers. The NEC group, having first shown that the transmission spectra contained features consistent with the surface plasmon theory, then demonstrated how the wavelength-selective transmission could be adjusted.² By using liquid crystal as the "substrate" (which sets many of the transmission properties of the grating), they could modulate transmission by simply changing the voltage, and without the need for polarizers to precondition the incoming beam or filter the light leaving the device. This property should make these gratings highly useful in displays and other spatial light modulators.

Another group, this time based at Imperial College, London, concentrated on doing experiments with slits.³ Based on their own models, they reached the conclusion that surface plasmons could indeed produce the coupling required to transfer light from the incoming surface of the element to the outgoing one. They also suggested another mechanism that could produce the same result -- coupling of incident plane waves with waveguide resonances in the slits -- but gave the theory no more than equal weight.

This seeming consensus was broken when Treacy argued that surface plasmons were not the cause of the increased transmission, but an effect of what was really causing it: Bloch waves.⁴ These occur when energy propagates through a lattice structure, and are caused by the interdependence of the different points in the array. According to Treacy, those Bloch waves that peak in one of the holes will only be weakly attenuated, whereas the others will be strongly absorbed. This, he says, points to diffraction as the cause. Treacy agrees that Wood's anomaly is indeed an important element, but says that this, too, backs up his theory that the entire mechanism is diffractive.

At the Ludwig-Maximilians-Universität München in Germany, researchers are not convinced. They used a scanning near-field optical microscope tip to illuminate the surface of a grating, demonstrating the creation of surface plasmons. Further, they showed that their propagation direction could be altered by changing the polarization of the incoming light. This, they say, allows the plasmon to be pointed at, and made to propagate through, a particular nano-hole (they call it nano-golf).

Treacy argues that the experimental results achieved in these experiments are just as consistent with the diffraction approach as the surface-plasmon interpretation, and is hoping to publish another paper on the subject soon. In any case, all the scientists involved agree on at least one thing: the enhanced-transmission phenomenon, however it is actually being achieved, will be a valuable tool for optical engineers.

References:

1. H.F. Ghaemi, Tineke Thio, D.E. Grupp, T.W. Ebbesen, and H.J. Lezec, Surface plasmons enhance optical transmission through subwavelength holes, Phys. Rev. B 58 (11), 15 September 1998.

2. C. Sönnichsen, A.C. Duch, G. Tae Jin Kim, Tineke Thio, T.W. Ebbesen, D.E. Grupp, H.J. Lezec, Control of optical transmission through metals perforated with subwavelength hole arrays, Opt. Lett. 24 (4), pp. 256-258, 15 February 1999.

3. J.A. Porto, F.J. García-Vidal, and J.B. Pendry, Transmission resonances on metallic gratings with very narrow slits, Phys. Rev. Lett. 83 (14), pp. 2845-2848, 4 October 1999.

4. M.M.J. Treacy, Dynamical diffraction in metallic optical gratings, Appl. Phys. Lett. 75 (5), 2 August 1999.Steininger, M. Koch, G. von Plessen, and J. Feldmann, Launching surface plasmons into nanoholes in metal films, Appl. Phys. Lett. 76 (2), pp. 140-142, 10 January 2000.