In the fifty years since Richard Feynman's eloquent assertion that there is
Plenty of Room at the Bottom the world of nano-engineering has matured so that
sophisticated systems can now be manufactured with nano-scale dimensions in
large quantities at low cost-flash memory chips are a classic example, with
their multi-billion-bit storage capacity on thumbnail-sized slivers of
silicon.
|
Evanescent Near Field Optical Lithography
(ENFOL) is a low-cost high resolution technique of
lithography which is capable of patterning features beyond the diffraction limit
of light. ENFOL requires the use of conformable masks for intimate contact due
to its use of the evanescent near field. |
Optical lithography has driven many of the advances in nano-scale
manufacturing, with its ability to print ever smaller features as the technology
matures. Lens-based optical systems are typically used, with improved resolution
usually being achieved by lowering the wavelength-current systems use 193nm
illumination from ArF excimer lasers. By applying some additional tricks, these
sophisticated projectors can be pushed to print dense lines narrower that 45nm,
which is less that 1/4 of that wavelength.
But these systems are highly advanced and expensive, so are only suited to
high-volume manufacturing. And getting the resolution below 20nm introduces
additional complexities and expense. There is an inverse law of
nano-manufacturing that seems to be at play here, where the cost of the printing
tool goes up inversely with the resolution. There is no better example of this
that with the extreme ultraviolet (EUV) systems that are being tested for
20nm-scale chip manufacturing-these use 13nm wavelength x-ray illumination and
state-of-the-art vacuum-based reflective optical systems, and the development
costs to date for the engineering test systems has been many billions of
dollars.
Is this inverse law immutable, or are there other ways to optically print
nano-scale patterns in moderate volumes at low cost? One technique that has been
explored by Richard Blaikie and his team at the University of Canterbury
in New Zealand is Evanescent Near Field Optical Lithography (ENFOL). As a simple
extension of contact lithography (Fig. 1), the expensive lenses are thrown away
and the mask pattern is directly transferred onto the substrate-by keeping the
mask and substrate in intimate contact we capture the evanescent light (Fig. 2)
that is locked within the near-field region of the mask, and nanoscale
resolution is possible even with relatively long wavelengths. Using the 436nm
line from a mercury lamp Blaikie's team proved the concept by printing sub-100nm features and, more recently, Ito and co-workers at
Canon Inc. have printed 32nm lines using a wavelength of 365nm.
|
| Figure 1. The ENFOL process. A
patterned mask is held in intimate contact with an ultra-thin photoresist layer.
UV illumination generates high-resolution, near-field (evanescent) light that is
captured by the resist. |
|
| Figure 2. The difference between
propagating light and evanescent light. Diffraction of a propagating wave from a
sub-wavelength grating produces exponentially decaying (evanescent) near-fields
trapped within a wavelength of the grating. These contain the
high-spatial-frequency information about the structure of the
grating. |
The contact requirement for ENFOL means that it will not be a direct plug-in
replacement for the mature projection printers, but it certainly adds another
viable technique to the nano-engineer's toolbox when it comes to knowing how to
manufacture niche products at moderate volumes. It sits alongside techniques
such as nano-imprint lithography (NIL) and dip-pen
nanolithography (DPN) in this toolbox, which are already being used
extensively for prototyping and manufacturing sophisticated nano-scale photonic,
electronic or biosensing systems.
And there is more to be explored for ENFOL and related techniques. Recently
Blaikie and others have shown that adding a silver
'superlens' to the mask can provide a much-needed protective layer above the
substrate, whilst maintaining resolution. And techniques for using reversible
bleaching in dye layers to project the high-resolution evanescent fields onto
the substrate have recently been pioneered by Rajesh
Menon at MIT (now at Utah), to combine the best features of ENFOL with the
convenience and maturity of traditional far-field projection optical
lithography.
So 50 years on, the words of Dr Seuss may be as prophetic as Feynman's here,
with the line From near to far, from here to there, funny things are everywhere
... in his classic One Fish, Two Fish, Red Fish, Blue Fish (Beginner Books,
1960). Near or far, funny things are everywhere in the world of nano-scale
optics, where the wavelength is no longer seen as fundamentally limiting the
length scale at which light can be used for nano-engineering.
Copyright AZoNano.com, Prof. Richard Blaikie (University of
Canterbury)