Latest News
Web Date: April 11, 2012

Block Copolymer Helps Lithography Get Smaller

Materials Science: Material could shrink silicon structures in computer chips, leading to fast electronics
Department: Science & Technology
News Channels: Materials SCENE, Nano SCENE
Keywords: block copolymer, lithography, computer chip manufacturing
Compressed Storage
A new block copolymer forms 5-nm-wide cylinders (white spots), which are the some of the smallest features known for such materials.
Credit: Christopher Ellison
Micrograph of block copolymer nanostructures.
 
Compressed Storage
A new block copolymer forms 5-nm-wide cylinders (white spots), which are the some of the smallest features known for such materials.
Credit: Christopher Ellison

A new polymer combination could help electronics manufacturers build nanoscale stamps for speedy computer chips and high-capacity hard drives. The material self-assembles into some of the smallest structures known for such polymers (ACS Nano, DOI: 10.1021/nn300459r).

To produce computer chips, manufacturers use lasers to etch patterns into polymer-covered silicon wafers, in a process known as photolithography. The smallest silicon features this method can produce are about 20 nm wide, says Craig J. Hawker, of the University of California, Santa Barbara, who was not involved with the new study. To pack more wires, transistors, and junctions onto a chip, thus producing faster electronics, chip manufacturers need to create patterns with smaller features, says Christopher J. Ellison, of the University of Texas, Austin.

Researchers have tried to produce these super small structures by using an etching template made from block copolymers. In a block copolymer, each polymer strand has two parts, or blocks, each with different physical properties. The polymers are like linking a bubble of oil to a water droplet, Ellison says. In solution, the molecules’ waterlike sections cluster into nanosized structures, surrounded by a matrix of oily plastic, created by the other polymer block. Etching away one block of the copolymer creates a pattern of tiny features for lithography.

When designing a block copolymer, researchers control the size of the etched structures in the template by changing the length of the polymer strand and adjusting the polarity difference between the two polymer parts. In general, shorter strands form smaller clumps. As the strands get shorter, researchers must increase the polarity difference between the polymer parts. If the difference isn’t large enough, the strands’ waterlike and oily blocks start to mix and the structures don’t form.

To create super small features that would easily self-assemble, Ellison, C. Grant Willson, also at UT Austin, Redouane Borsali, of Centre de Recherches sur les Macromolécules Végétales in France, and their colleagues looked to design a new copolymer that kept the strands small and maximized the polarity difference. The result was a hydrophilic strand of sugars – either maltoheptose or a branched xyloglucan – connected to a hydrophobic string of silicon-tipped polystyrene.

On a silicon wafer, the sugar strands clumped into 5-nm-wide cylinders surrounded by a matrix of polystyrene. If each sugary cylinder was a bit of information on a hard drive disk, a square inch of disk area would contain 8 terabits of information, far greater than the 1 terabit per square inch manufacturers want to produce, Ellison says.

When the researchers blasted the patterned wafer with oxygen plasma, they removed the sugar from the surface, leaving behind the polystyrene. The result is a honeycomb of cylindrical cavities that could be used in subsequent chip production steps.

UC Santa Barbara’s Hawker applauds the researcher’s design strategy for the copolymer. “It’s a wonderful first step towards future potential commercialization” of copolymer lithography, he says.

One of the next challenges is to control the pattern formed during the polymers’ self-assembly, Hawker adds.

 
Chemical & Engineering News
ISSN 0009-2347
Copyright © American Chemical Society
Comments
todd (Wed Apr 11 21:49:13 EDT 2012)
Dang good resolution for AFM. I assume you will publish the conditions used to image.
Karol Vegso (Thu Apr 12 13:05:58 EDT 2012)
I am quite surprised what Craig J. Hawker said. The periodicity of 20 nm is quite good number. My estimation was 50 nm.
Karol Vegso (Fri Apr 13 06:51:55 EDT 2012)
Polymer science is simply amazing.

Leave A Comment

*Required to comment