This paper reports on an experiment that demonstrates a superlens which is capable of breaking down the diffraction limit that has been a century-old obstacle in optics. This confirms the British physicist John Pendry’s theory of a superlens which he proposed in the year 2000. Though this paper shows superlens imaging at near-field, it is merely the first step towards a new optics and imaging.

The realization of the superlens will have a profound impact on many disciplines in science and technology. Using the surface excitations to significantly enhance a broadband of evanescent waves in a material that has either negative permitivity or permeability or both, one may be able to construct the superlens which has a unique capability to recover the lost evanescent waves, resulting in a new optics paradigm.

Making a flawless lens has long been a dream for lens makers. However, imaging with a regular lens has a fundamental constraint. The so-called optical "diffraction limit," where the smallest features can be seen by a lens, is about half of the light’s wavelength. This is because, during imaging, objects emit "evanescent" waves which carry the smallest details of an object, decay exponentially, and thus never make it to the image plane. This limit can now be lifted by using unique surface excitations that are able to greatly enhance those evanescent waves, forming an image that has a far better resolution.

I'd like to give credit to my hard-working team members—lead author Nicholas Fang (my former student who is now an Assistant Professor of mechanical engineering at the University of Illinois at Urbana-Champaign), Hyesog Lee, a graduate student, and Cheng Sun, a research scientist in my lab. We started this experiment soon after Pendry’s theoretical proposal of the perfect lens was first published in Physical Review Letters.

It took us quite a few years, however, to achieve quality results. In late 2002, we first observed the superlens effect with a 2D dot array object, in Fourier space, but the imaging of real space was still a bit fuzzy, due to the control of surface conditions that caused large noises. So, it took us another year to produce a much better superlens image. Our team developed a few key methods to overcome these problems which should prove useful in the future usage of this superlens.