IEEE Spectrum February, 2014 - 58

created a special photoresist with the right refractive index to replace
that oil, so that the resist could be dropped directly onto the lens.
Without the aforementioned two-photon absorption, of course,
this scheme wouldn't be a good idea at all. Photons coming out of
the lens would cure the photoresist all along the beam line, caus-
ing the photoresist to stick to the lens and destroying the lens after
its first use. But recall that with two-photon absorption, we can
localize the curing process in three dimensions.
This "dip-in" approach also enables us to build structures on sur-
faces other than glass. And because the resist and the lens are no
longer separated by a thick glass substrate, we can extend the focus
a good 10 times as far into a sample, creating 3-D metamaterials
arrays as tall as 1 mm. That's getting close to a height where we can
begin thinking about constructing macroscopic metamaterials-
objects that are large enough to hold in your hand-rather than
just thin, delicate coatings.
Of course, there's still room for improvement. The resolu-
tion of direct laser writing is only barely sufficient to make light-
manipulating metamaterials that can operate at visible frequencies.
Some more-complex structures, such as forests of metal helices,
are still out of reach, and making them will likely require new
photoresists. But in principle at least, there is no obvious reason
we won't succeed.
We'll also have to tackle the issue of writing speed. On that
front, we've actually been making good progress. In research

58

|

FEB 2014

|

north amErican

|

SPEctrUm.iEEE.orG

labs, the laser is fixed, and structures are drawn by moving the
sample around using high-precision 3-D piezoelectric actuators.
These are commonly used at speeds of about 100 micrometers
per second. For many structures, that translates to half an hour
or so of work to create a metamaterial with a 100- by 100-µm
footprint. But recently, Nanoscribe and others have adopted
2-D galvanometer-based laser scanning systems that can write
much faster, with speeds upwards of 1 meter per second. Using
these lithography instruments, which are now commercially
available, we can reduce the fabrication times by a factor of a
hundred or more.
Metamaterials are just the start. With the recent speed boost and
with further improvements to spatial resolution, we think that direct
laser writing might one day compete with planar electron-beam
lithography, which is used today to make the stencil-like masks
that pattern circuits on logic and memory chips.
And there's more. By incorporating active elements, such as
nanodiamonds engineered to emit single photons, we could even
envision creating 3-D optical systems for chip-based quantum in-
formation processing.
We're only beginning to explore what can be done with these
new fabrication tools. But in the future, there may be no limit to
how sophisticated and intricately structured matter can become. n
Post your comments at http://spectrum.ieee.org/metamaterials0214
http://spectrum.ieee.org/metamaterials0214 http://SPEctrUm.iEEE.orG
Table of Contents for the Digital Edition of IEEE Spectrum February, 2014
