Many of the methods we have described are
based on AFM technology, which currently suffers
from low throughput. Work is being done to increase
the scan speed using high-bandwidth mechanics
and by predistorting the drive for a step and
settle response faster than the mechanical
resonant frequency of the cantilevers would suggest.
Another effort to increase throughput is by creating parallel arrays of AFM tips, which have already
reached as many as 11 million tips for DPN
systems.
To improve e-beam and FIB systems, faster
beam-writing speeds using higher beam currents
and more sensitive resists are the order of the day.
In dynamic-stencil lithography, faster microelectromechanical
plates, smaller apertures, and large arrays
of apertures are where things are headed. In nanoimprint
techniques, efforts are focusing on better dimensional
stability and better overlay capabilities.
Nearly every area of applied physics and engineering
is involved in some aspect of nanomanufacturing.
Currently, each method fills a specific niche
in which it bests other approaches. By leveraging
the strengths of different technologies, researchers
are developing processes that combine multiple
techniques to write the highest-resolution masks
that can be projected or printed with high accuracy
and throughput. Clearly, the nanodevices of the future
will be built by optimizing and combining the
strengths of many technologies. There will never
again be a one-size-fits-all approach, as optical lithography
has been over the past four decades.