This image shows channels etched using sequential infiltration synthesis, which scientists at Argonne have used to create features that have high aspect ratios – that is, they are far deeper than wide. These crevasses will permit the creation of a new generation of semiconducting materials. |
When searching for the technology
to boost computer speeds and improve memory density, the best things come in
the smallest packages.
A relentless move toward smaller
and more precisely defined semiconductors has prompted researchers at the U.S.
Department of Energy’s (DOE) Argonne National Laboratory to develop a new
technique that can dramatically improve the efficiency and reduce the cost of
preparing different classes of semiconducting materials.
The new discovery meets certain
requirements of the international semiconductor “roadmap” all the way out to
2022—leapfrogging an anticipated ten years of progress with a single set of
experiments.
Most semiconductor patterns are
currently made using a process known as photolithography, in which portions of
a thin film are selectively removed to create a pattern. The pattern in this
film, known as a resist, is etched into the semiconductor by exposure to an
ionized gas. This gas also etches away the resist itself, reducing the number
of times the film can be used. Especially durable resists are known as hard
masks.
The drive to create smaller and
smaller semiconductor components is often limited by a phenomenon known as
domain collapse, said Argonne nanoscientist Seth Darling.
Conventional lithography—the technique used to make patterns in materials—attempts
to create features that are separated like the teeth of a comb. However, gaps
in the resist that are too deep tend to collapse inwards, which makes the
material useless.
“Engineers have tried many ways
of avoiding this collapse, but the industry is constantly running up against
it,” Darling said.
In 2010, Darling and his
colleagues developed a technique known as sequential infiltration synthesis
(SIS), which used gases to grow hard inorganic materials inside a soft polymer
film. The work was supported by the DOE Office of Science through Argonne’s Center for Nanoscale Materials and the Argonne-Northwestern Solar Energy Research
Center.
One of the most notable benefits
of SIS is that it eliminates the need for hard masks in photolithography,
according to Darling. “Hard masks are a real pain when it comes to semiconductor
processing—they’re expensive, complicated, reduce pattern quality, and add
extra steps,” he said.
According to Darling, sequential
infiltration synthesis has already been identified by leading semiconductor
companies as a technology with the potential to overcome several different
limitations.
In a recent experiment, Darling
and his Argonne colleagues showed that SIS can
actually eliminate pattern collapse, enabling the fabrication of materials that
have patterns with higher “aspect ratios,” which measures the height of a
feature divided by its width.
Generally speaking, lithography
seeks to create patterns with higher aspect ratios while using as little resist
as possible. “Usually, you need a certain thickness of the resist in order for
the process to work,” Darling said. “This new process enables us to do away
with a lot of that problem.”
“One of the biggest advantages
of this new study is that we’ve shown the possibility of using SIS for
photolithography, which is one of the most industrially important processes,”
Darling said. “As there’s more and more demand for better electronics, the
sizes of these semiconductors need to keep getting smaller and smaller, and it
becomes that much more important for us to meet and exceed the benchmarks that
we’ve set for ourselves.”
Source: Argonne National Laboratory