Volume 94 Issue 43 | p. 29
Issue Date: October 31, 2016

Cover Stories: 10 Start-ups to watch

Inpria

Bringing new chemistry to solve a chip-making problem
Department: Business
Keywords: semiconductors, photoresist, Moore’s law, tin oxide

The semiconductor industry has followed Moore’s law—the idea that the number of transistors in a computer chip will double every two years—faithfully since 1965. Lately, it hasn’t been easy.

As chip circuit lines approach 10-nm widths and the industry contemplates single-digit dimensions, lithographic drawing techniques based on 193-nm light aren’t cutting it anymore. So, the industry has been developing a new technique based on 13.5-nm extreme ultraviolet light. After many years and billions of dollars, EUV lithography is almost ready.

Inpria

Launched: 2007

Headquarters: Corvallis, Ore.

Focus: photoresists for EUV lithography

Technology: tin oxides that replace organic materials

Founders: Andrew Grenville and Douglas A. Keszler

Funding or notable partners: $21 million to date from investors including Air Liquide, Applied Materials, Intel, Samsung, and TOK

There’s a weak link, though: the photoresists that chip makers expose with the light and then etch away to create chip features. Developed to work in 193-nm light and chemically tweaked over the years as circuit lines shrink, today’s resists are struggling to keep up.

[+]Enlarge
Grenville
Credit: Inpria
Photo of Andrew Grenville, CEO of Inpria.
 
Grenville
Credit: Inpria

That’s where Inpria comes in. Launched in 2007 with technology developed in the Oregon State University labs of inorganic chemist Douglas A. Keszler, Inpria calls itself the first company to purpose-build a photoresist for EUV lithography.

When you do something from the ground up, you often do it differently. The veterans of photoresist chemistry—big firms such as Dow Chemical, JSR, and Shin-Etsu Chemical—base their products on acrylic polymers. Inpria, in contrast, is using inorganic chemistry—specifically tin oxide nanoclusters.

Inpria’s CEO, Andrew Grenville, who left Intel in 2007 to start the company with Keszler, explains that the large carbon chains in traditional resist molecules are getting too big relative to the feature sizes they need to pattern. The core molecules in Inpria’s tin oxides are one-fifth the size, he says.

In addition, Grenville says, the metal oxides are much more durable. They offer 10 times the etch selectivity of conventional polymers, he says, meaning the resist stands up well while a pattern is being etched into the substrate. “We’re putting the ‘resist’ back in ‘photoresist,’ ” he quips.

[+]Enlarge
Hard to resist
Because conventional resists have etch properties similar to the spin-on carbon substrate to be patterned, semiconductor makers typically incorporate two intermediate pattern-transfer layers. But these extra layers add costly steps and can lead to pattern-transfer errors. Inpria says the superior etch resistance of its inorganic photoresists permits chip makers to eliminate the extra layers, simplifying the process.
Credit: Inpria/Imec/C&EN/Shutterstock
A schematic of chip making using Inpria’s photoresist.
 
Hard to resist
Because conventional resists have etch properties similar to the spin-on carbon substrate to be patterned, semiconductor makers typically incorporate two intermediate pattern-transfer layers. But these extra layers add costly steps and can lead to pattern-transfer errors. Inpria says the superior etch resistance of its inorganic photoresists permits chip makers to eliminate the extra layers, simplifying the process.
Credit: Inpria/Imec/C&EN/Shutterstock

The electronics industry is starting to notice. In June, Inpria completed an $11 million financing round that included two repeat investors—the chip makers Intel and Samsung—and two new ones—the electronic materials suppliers Air Liquide and Tokyo Ohka Kogyo (TOK). Inpria says it will use the money to build a bigger resist facility at its Corvallis, Ore., headquarters.

As Inpria pushes ahead with tin oxides, traditional suppliers are trying to extend organic photoresist chemistry. For many years, they have employed acid-catalyzed chemical amplification to boost resist exposure. But the acid molecules can run rampant, causing roughness in circuit-line edges. For EUV, Dow’s solution is tethering the acid generator to the polymer. JSR is developing entirely new acid generators.

Mark Thirsk, a former electronic materials executive who is now with Linx Consulting, likes what he sees in Inpria’s inorganic approach, but he cautions that chip makers such as Intel and Samsung are comfortable with today’s acrylic chemistry and don’t especially like to change.

Grenville acknowledges that inorganic photoresists will be a brave new world for the chip industry. He also knows from his Intel years that the industry will change when given a good reason. Inpria, Grenville argues, offers that reason. “We’re coming in with a photoresist that unlocks the full potential of EUV lithography,” he says.

Know a start-up we should consider for next year's list? Nominate it here: cenm.ag/startupnom

 
Chemical & Engineering News
ISSN 0009-2347
Copyright © American Chemical Society
Leave A Comment