Success stories

Richard Blaikie: New methods for nanolithography

Richard with his PhD student David Melville programming the electron lithography machine.

Richard Blaikie spends life straddled between physics and engineering. Trained as a physicist he likes to penetrate the fundamental issues of physics but in areas that have potential to be technologically relevant. He is now associate Professor of the Electrical and Electronic Engineering department at Canterbury University and deputy director of the MacDiarmid institute.

His work focuses on the development of lithographic techniques that permit the reduction in size of on-chip electronic elements to the nano-scale. There is a law in nanofabrication that the smaller you make things the more expensive they get. Richard has set out to defy this law.

Optical lithography- the traditional method.

“The traditional way of making semiconductor circuits since the 1950’s is to use optical lithography.” You start with a silicon substrate coated with a layer of light sensitive material called photo-resist. A mask with a pattern of transparent and opaque areas is placed over the top and the whole thing exposed to optical radiation. Exposed areas of photo-resist are removed in solution and the pattern can be transferred onto the underlying substrate using a variety of techniques. For a diagram showing the basic lithographic process see the article on Maan Alkaisi.

The method is conceptually very simple but when you try and shrink structures to the nano-scale wave-length limits are reached; diffraction causes blurring which destroys the pattern.

ENFOL - taking optical lithography to its limits

Richard set to work to discover what the smallest thing you can make using light. “I knew that people had used scanning near field optical microscopes. You get a light pipe with a very small hole at the end of it. Light leaks out of the hole and you can get resolution down to one twentieth of the wavelength. People have done…experiments using these to write out a pattern for nano-fabrication but it struck me as not being a very efficient way of going about it”.

One day at home Richard had an idea. “Why don’t you just do it all in parallel and have the aperture the shape you want then just print it all in one go rather than having to write your pattern point by point. ” It turned out that the idea was not that novel. It is called contact lithography and has been around for years “but no one had ever really tested the resolution limits of contact lithography so we just gave it a fancy name ‘ENFOL’ (Evanescent Near Field Lithography)and set to work. We’ve got (resolution) about one fifth to one seventh of the wavelength at the moment and we think we can go even smaller.”

Computer simulations

“Its hard work to get from the basic idea to demonstrating it…You can’t just think of ray traces…to get a really accurate answer you’ve got to use all of Maxwell’s equations and solve them.” Maxwell’s equations are the theoretical foundation for the entire field of electromagnetics. “There are very few cases where you can get an exact answer as a formula… that’s where modern computers are wonderful” You can plug Maxwell’s equations and the dimensions of the aperture into the computer and use it as a tool to work out solutions and run simulations and experiments.

The mysterious silver lens

"With ENFOL we are looking at a regime where you can’t have any gap whatsoever between your mask and your photoresist to get the best resolution. If the two are in intimate contact, you have what’s called evanescent fields close to the mask. These fields are usually ignored because they die off exponentially as you move away from the mask but Richard and his team have demonstrated a way of harnessing them and using them to project a near-field image. “We put a layer of silver between the mask and the imaging resist”. Its completely non-obvious but this layer acts as a lens, which refocuses the diffracting image.

The explanation is ‘negative refraction’. Rather than bending towards or away from the normal as in conventional optics, negatively refracted light rays bend completely on the other side of the normal. Light waves have even been known to bend backwards – “really strange phenomena!”

The idea of using materials with a negative refractive index as flat surfaced lenses was first put forward in the 1960’s and then theoretically investigated by John Pendry from Imperial College London in 2000. “ What John discovered is that (evanescent) waves very close to an object are very different for a material with negative refractive index. They actually grow rather than decaying and the consequence is that you have a material where they will be amplified. information will be amplified. So by putting a lens made from such a material between an object and an imaging plane you can have the fields decaying in the air around the lens, re-growing in the lens itself and decaying towards the image plane – and the net result is that you can get perfect reconstruction of the object.

Above: The silver lensing experimental setup. Below: Images formed with a 50-nm thick silver superlens, with periods of: (a) 500 nm, (b) 350 nm, (c) 290 nm, (d) 250 nm, (e) 200 nm and (f) 170 nm.
Figures reproduced from: D. O. S. Melville and R. J. Blaikie,

“Super-resolution imaging through a planar silver layer”, Opt. Express 13, 2127-2134 (2005). www.opticsexpress.org © 2005 Optical Society of America, Inc

“Its all to do with ‘surface plasmons’ – charge oscillating on the surface on the surface of the silver.” Radiation hitting the front face of the silver slab makes the electrons oscillate. “If the frequency is right, i.e. near plasma frequency of the silver, the plasmons on one side of the silver slab can induce another set of plasmons on the other side. These then act as tiny radiation sources or antennae that are phased in such a way as to form a real 1:1 magnified image on the other side of the silver slab. Magic really!”

Although this sort of experiment has been done using microwave frequencies, Richard’s group was the first to demonstrate that a flat slab of silver can act as a lens to re-form an image in the near field.The experimetns and simulations have largely been carried out by David Melville, who is doing this work for his PhD under Richard's supervision.

For their first experiments the silver lenses were still thick (in nano terms) at around 80nm to 120nm, and this has allowed for resolution down to 250nm. “This is smaller than the wavelength, but not yet below the diffraction limit which would require features smaller than half the wavelength.” They have now repeated the experiments for 50-nm thick silver and have imaged sub-diffraction-limited features; this is a key result, proving that Pendry’s controversial prediction can be achieved in a practical experiment.

The possible applications are numerous. “Optical microscopy, data storage and photolithography are the main application areas, particularly for visible or UV light. Also our work investigating resolution limits for near field photolithography and silver lensing would make high-resolution proximity lithography a real possibility. Industries that use optical imaging don’t like the close contact because it can damage the mask. So the advantage is that by allowing a small gap between the mask and the imaging resist, the damage to the expensive mask can be eliminated.”

As with any new technology “nobody could imagine all the applications that will result.”